Pembrolizumab plus High-Dose IL-2 in Advanced Clear Cell Renal Cell Carcinoma: Six-Year Survival Outcomes and Molecular Signatures from a Phase 2 Trial

Pembrolizumab plus High-Dose IL-2 in Advanced Clear Cell Renal Cell Carcinoma: Six-Year Survival Outcomes and Molecular Signatures from a Phase 2 Trial | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Pembrolizumab plus High-Dose IL-2 in Advanced Clear Cell Renal Cell Carcinoma: Six-Year Survival Outcomes and Molecular Signatures from a Phase 2 Trial Jad Chahoud, Jeffrey Johnson, Justtin Miller, Michael Schell, and 19 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7698261/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract Prolonged or indefinite systemic therapy remains standard for advanced clear cell renal cell carcinoma (ccRCC), often resulting in cumulative toxicities and treatment burden. We conducted a single-arm phase 2 trial of a fixed-duration regimen of pembrolizumab plus high-dose interleukin-2 in treatment-naive advanced ccRCC. Among 26 patients treated, after a median follow-up of 76.4 months, the objective response rate was 73%, with complete responses in 42%. Median overall survival was not reached, with 5-year survival of 73%; median progression-free survival was 19.3 months; and median treatment-free survival was 23.8 months, with 42% of patients remaining treatment-free at 5 years. No grade 5 adverse events occurred, and no patient with durable disease control experienced persistent grade ≥2 toxicities. Correlative analyses identified immune patterns associated with durable benefit, including enrichment of CD16⁺ natural killer cells, suppression of PD-1⁺ T-cell frequencies, and coordinated chemokine, complement, and PKC/TGF-β pathway activation. ClinicalTrials.gov identifier: NCT02964078. Health sciences/Medical research/Clinical trial design/Clinical trials/Phase II trials Biological sciences/Cancer/Cancer therapy/Cancer immunotherapy Biological sciences/Cancer/Tumour biomarkers Biological sciences/Cancer/Tumour immunology RCC immunotherapy cytokine treatment-free survival pembrolizumab interleukin-2 combination immunotherapy durable survival durable response long-term follow up multi-omics proteomics transcriptomics Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 1. Introduction In 2025, kidney cancers, including renal cell carcinoma (RCC) and cancers of the renal pelvis, are estimated to affect approximately 80,980 individuals in the United States, with approximately 20-30% of patients presenting with metastatic disease at the time of diagnosis. 1,2 The therapeutic landscape for the first-line treatment of metastatic RCC (mRCC) has evolved dramatically, with contemporary standard-of-care regimens involving continuous, or prolonged, dosing of either dual immune checkpoint inhibitor (ICI) therapy or combination ICI plus tyrosine kinase inhibitor (TKI) therapy. 3 For example, the CheckMate 214 trial evaluated dual immune checkpoint inhibition with nivolumab (anti-PD-1) plus ipilimumab (anti-CTLA-4), and found improved overall survival compared to sunitinib across all risk groups of advanced renal cell carcinoma patients with 8 years of follow up. 4 ICI-TKI combinations, including pembrolizumab-axitinib, nivolumab-cabozantinib, and pembrolizumab-lenvatinib have also exhibited improved survival outcomes in the KEYNOTE-426, CheckMate-9ER, and CLEAR trials, respectively, with median follow-up intervals ranging from 33 to 67.2 months. 5–10 Despite these advances, current therapeutic approaches require prolonged treatment administration. Dual checkpoint inhibition may entail up to two years of continuous dosing, while ICI-TKI combinations may continue indefinitely. 3 This treatment paradigm presents challenges for patients with mRCC, who face chronic symptoms including fatigue, pain, dyspnea, and insomnia that significantly impact quality of life. 11 Prolonged treatment exposure may compound these burdens by increasing the risk of chronic treatment toxicities, imposing additional financial and psychosocial stress, and escalating healthcare resource utilization, with contemporary first-line immunotherapy combinations incurring costs exceeding $260,000-$325,000 annually. 12–16 In response to these concerns, treatment-free survival (TFS), which quantifies the duration of disease control maintained without additional treatment, has emerged as a patient-centered metric for evaluating therapeutic regimens. 17 Contemporary efforts to characterize TFS in the CheckMate 214 trial revealed that 18% of patients treated with nivolumab plus ipilimumab (nivo+ipi) remained treatment-free at the 5-year timepoint. 18 Moreover, the nivo+ipi population had a 5-year mean overall survival of 40.7 months and mean TFS of 11.1 months. 18 The discrepancy between overall and treatment-free survival highlights the need for novel therapeutic strategies that can achieve durable objective responses while enabling sustained treatment-free intervals. Interleukin-2 (IL-2), approved by the United States Food and Drug Administration in 1992, was the first immunotherapy to demonstrate efficacy in metastatic RCC. 19 While high-dose IL-2 can produce durable complete responses in select patients, its widespread use has been limited by acute toxicity. 20 Mechanistically, ICI and cytokine therapies demonstrate complementary approaches to immune activation. Pembrolizumab, an anti-PD-1 ICI, acts to overcome immune tolerance by inhibiting PD-1/PD-L1 interaction, enhancing T-cell recognition of tumor antigens, and inducing intratumoral lymphocyte penetration; while IL-2 promotes proliferation, cytotoxic capacity, and survival of activated T- and NK-cells, potentially amplifying the effector response. 21–23 Recently, alternative approaches to harness IL-2-mediated immune activation have been explored. The phase 3 PIVOT-09 trial evaluated bempegaldesleukin (BEMPEG), a pegylated IL-2 prodrug in combination with nivolumab, an anti-PD-1 ICI, for treatment-naïve advanced ccRCC. 24 Despite its design to overcome limitations of conventional IL-2 therapy, BEMPEG plus nivolumab did not improve ORR or OS versus TKI monotherapy in patients with intermediate and poor-risk disease. In contrast to the simultaneous administration of a modified cytokine prodrug plus PD-1 inhibition, our approach employs sequential immunotherapy administration with high-dose IL-2. Here, we report long-term survival outcomes from the first prospective trial evaluating short-course pembrolizumab plus high-dose IL-2 in treatment-naive advanced ccRCC. This time-limited approach addresses the critical unmet need for effective therapy that enables sustained treatment-free survival. 25 With over 6 years of follow-up, we assess the durability of survival outcomes and identify molecular signatures associated with sustained response. 2. Results Participants Between April 2017 and October 2018, 27 patients were enrolled in the study. One enrollment-eligible patient was later ineligible due to progression of disease prior to study day one, and did not receive any treatment on study, resulting in a treatment population of 26 patients (Fig. 1a). The median age at the time of enrollment was 60.5 years (range: 40-74). 22 patients (85%) were male, and 4 patients (15%) were female. International Metastatic RCC Database Consortium (IMDC) risk classification identified 6 patients (23%) with favorable risk, 19 (73%) with intermediate risk, and 1 (4%) with poor risk disease. Prior to study enrollment, 21 patients (81%) had undergone radical nephrectomy, 1 patient (4%) had undergone partial nephrectomy, and 4 patients (15%) had not undergone nephrectomy. Baseline demographic and clinicopathologic data are displayed in Table 1a. Table 1a. Baseline Characteristics of Patients Treated with Pembrolizumab Plus IL-2 Stratified by Treatment-Free Survival Characteristic TFS > 5 years (N=11) TFS 6 months – 5 years (N=10) TFS < 6 months (N=5) Total (N=26) Median Age, years (range) 65 (38-74) 58 (40-67) 60 (48-66) 61 (38-74) Sex, n (%) Male 10 (38) 9 (35) 3 (12) 22 (85) Female 1 (4) 1 (4) 2 (8) 4 (15) Race, n (%) White 11 (42) 8 (31) 5 (19) 24 (92) Asian 0 (0) 1 (4) 0 (0) 1 (4) Unknown 0 (0) 1 (4) 0 (0) 1 (4) Ethnicity, n (%) Not Hispanic/Latino 11 (42) 8 (31) 4 (15) 23 (89) Hispanic/Latino 0 (0) 2 (8) 1 (4) 3 (12) IMDC Score, n (%) Favorable 3 (12) 2 (8) 1 (4) 6 (23) Intermediate 7 (27) 8 (31) 4 (15) 19 (73) Poor 1 (4) 0 (0) 0 (0) 1 (4) Primary Tumor Grade, n (%) Grade 2 3 (12) 2 (8) 0 (0) 5 (19) Grade 3 7(27) 5 (19) 2 (8) 14 (54) Grade 4 1 (4) 2 (8) 1 (4) 4 (15) Histological Type, n (%) Clear Cell RCC 11 (42.3) 10 (38.5) 5 (19.2) 26 (100.0) Distant Metastatic Site, n (%) Lung/Pleura 6 (23) 7 (27) 4 (15) 17 (65) Lymph Node 14 (54) Liver 2 (8) 1 (12) 0 (0) 3 (11) Bone 2 (8) Other a 7 (27) Includes brain, pancreas, chest wall, gluteal, paraspinal, soft tissue of hip, retroperitoneum. Table 1b. Radiological Response and Efficacy Outcomes Efficacy outcomes Radiological Response Following Treatment, n % Complete response (CR) 11 (42%) Partial response (PR) 8 (31%) Stable disease (SD) 5 (19%) Progressive disease (PD) 2 (8%) Overall Response Rate (ORR), n (%) CR/PR 19 (73%) Disease Control Rate (DCR), n (%) CR/PR/SD 24 (92%) Median Survival Outcomes, n months (IQR) Overall Survival Progression-free Survival Treatment-free Survival Not reached (54 - NR mo.) 19.3 mo. (8.4 - NR mo.) 23.8 mo. (12.9 - NR mo.) 5-year Survival Outcomes, % (95% CI) Overall Survival Progression-Free Survival rate 73% (52 - 86%) 42% (24 - 60%) Treatment-Free Survival 42% (24 - 60%) IMDC Favorable Risk, n months (IQR) Median Overall Survival Median Progression-free Survival Not reached Not reached (12.2 - NR mo.) Median Treatment-free Survival Not reached (9.7 - NR mo.) IMDC Intermediate/Poor-Risk, n months (IQR) Median Overall Survival Not reached (26.7 - NR mo.) Median Progression-free Survival Median Treatment-free Survival 17.3 mo. (7.9 – NR mo.) 23.7 mo. (13.9 - NR mo.) Efficacy Outcomes Patients received a median of 12 pembrolizumab doses (Interquartile Range [IQR]: 10-12, range: 3-12), with 14/26 patients (54%) completing all twelve planned doses. Of the 40 planned IL-2 doses, patients received a median of 25.5 doses (IQR: 18-33, range: 0-38). After a median follow-up duration of 76.4 months (IQR: 73.4-80.4 months) at the time of data cutoff, 19/26 (73%) of patients with therapy-naive disease achieved an objective response to pembrolizumab plus IL-2 therapy as assessed by the investigator (95% confidence interval [CI]: 52%-88%). The study therefore met its primary endpoint with an ORR exceeding the pre-specified threshold of 45%. Best response as determined per RECIST 1.1 criteria was a complete response (CR) in 11 patients (42%), partial response (PR) in 8 patients (31%), stable disease (SD) in 5 patients (19%), and progressive disease (PD) in 2 patients (8%), resulting in a disease control rate (DCR) of 92% (95% CI: 75%-99%; Table 1b). At data cutoff, median overall survival (OS) was not reached (NR) (IQR: 53.7-NR months), median progression-free survival (PFS) was 19.3 months (IQR: 8.4-NR months), and median treatment-free survival (TFS), defined as time from completion of pembro IL-2 to next line of therapy, date of last follow up, or death, was 23.8 months (IQR: 12.9-NR months). OS rates at the 1-, 3-, and 5-year follow-up timepoints were 100% (95% CI: 100%-100%), 76.9% (95% CI: 56-89%), and 73% (95% CI: 52-86%), respectively. PFS rates at the 1-, 3-, and 5-year timepoints were 62% (95% CI: 40-77%), 42% (95% CI: 24-60%), and 42% (95% CI: 24-60%), respectively. TFS rates were 77% (95% CI: 56-89%) at 1 year, and 42% (95% CI: 24-60%) at both the 3- and 5-year timepoints. Time-to-event outcomes and individual patient responses are illustrated in Fig. 2. Safety No grade 5 safety events occurred throughout the duration of the study as assessed per Common Terminology Criteria for Adverse Events (CTCAE) v4.0 grading criteria. No new or unexpected adverse events (AEs) have occurred following the previous study report. 25 None of the 11 patients with durable disease control, without the need for additional systemic therapy beyond this trial, had any persistent grade 2 or higher AEs after a median follow-up duration of 76.4 months. Biomarker Outcomes Characterization of Response Categories Based on Treatment-Free Survival In this study, we retrospectively defined clinical response categories based on treatment-free survival duration to capture the durability of treatment benefit and identify patients achieving sustained disease control. TFS was calculated from the date of treatment discontinuation to the date of subsequent systemic therapy initiation, disease progression requiring intervention, or death from any cause. Upon analysis of TFS outcomes in the patient population (n=26), three distinct response categories emerged that exhibited clear separation in clinical outcomes. 11/26 patients (42%) demonstrated TFS >5 years (extreme responders), 10/26 patients (39%) had TFS between 6 months and 5 years (intermediate responders), and 5/26 patients (19%) had TFS <6 months (non-responders), suggesting primary non-response or rapidly progressive disease despite combination immunotherapy. To generate hypotheses regarding biological mechanisms underlying extreme response, we performed transcriptomic, proteomic, and flow cytometric analyses comparing patient subgroups defined by treatment-free survival duration. Integrated Time-Independent Transcriptomic and Proteomic Analysis Identifies Distinct Immune Signatures in Exceptional Responders To capture biological differences that include patients with early treatment discontinuation, we performed time-independent transcriptomic and proteomic analyses averaging expression across all available timepoints. 19/26 patients (73%) had evaluable data. Gene expression analysis of normalized NanoString data revealed 20/249 genes (8.0%) with statistically significant differential expression (p < 0.05) between extreme and non-responders (Fig. 3a). No genes in this comparison survived false discovery rate (FDR) correction. Extreme responders exhibited significantly greater expression of 19 genes, predominantly chemokines, inflammatory cytokines, and immune effector molecules. Of the significantly upregulated genes in extreme responders, the largest log 2 fold changes in expression were observed in IL-8 (log 2 [FC] = 1.44), IL-1B (log 2 [FC] = 0.94), TLR4 (log 2 [FC] = 0.93), MEF2C (log 2 [FC] = 0.92), and CXCL2 (log 2 [FC] = 0.86). Time-independent proteomic analysis comparing extreme and non-responders identified 22/276 proteins (7.97%) with statistically significant differential expression (p < 0.05; Fig. 3d). No proteins in this comparison survived FDR correction. Among the significantly downregulated proteins in extreme responders, the largest log 2 fold changes were observed in MIC-A/B (log 2 [FC] = -1.04), MetAP2 (log 2 [FC] = -0.88), ADA (log 2 [FC] = -0.78), SOD1 (log 2 [FC] = -0.74), and GMZA (log 2 [FC] = -0.73). Gene set enrichment analysis (GSEA) was conducted to identify pathway signatures distinguishing response groups (Fig. 3h-j). Transcriptomic GSEA revealed that extreme responders demonstrated significantly lower protein serine/threonine kinase activity compared to non-responders (q < 0.05, normalized enrichment score [NES] = -1.53), with trends toward enhanced chemokine receptor binding and chemokine activity (NES = 1.49, q = 0.07). When compared to non-responders, intermediate responders showed significantly reduced serine/threonine kinase activity (q < 0.05) and stress-responsive signaling pathways involving MAPK, NF-κB, and TGF-β cascades (q < 0.05). Proteomic GSEA showed that extreme responders demonstrated significantly greater innate immunity pathway activity compared to intermediate responders (NES = 1.65, q = 0.03) and trended toward enhanced chemokine signaling pathways (q < 0.1). Time-Dependent Gene Expression Analysis Reveals Immune Modulation Associated with Treatment We performed time-dependent analysis of immune gene expression across treatment phases (baseline, pembrolizumab monotherapy, IL-2 addition, and post-IL-2; mean: 3.1 timepoints per patient) to capture dynamic patterns underlying sequential administration of pembrolizumab and IL-2. In a pooled analysis of all response groups, the administration of IL-2 produced the most pronounced transcriptional response, with six genes significantly altered (B2W4 vs. B2W2, n=15 paired samples; q 0.6). Five genes were upregulated, including chemokine receptors and ligands (CCR3, CCL3), inflammatory mediators (IL-11), and immune effector molecules (CSF1, HMGB2), while CD40 was downregulated (Fig. 4a). In contrast to the robust IL-2 signature, analyses of pembrolizumab monotherapy (B2W2 vs. B1W1) and post-IL-2 phases (B4W1 vs. B2W4) showed no significant transcriptional changes. Cross-sectional transcriptomic analyses revealed distinct gene expression patterns between response groups at sequential correlative timepoints. Following pembrolizumab monotherapy (B2W2), both extreme and intermediate responders exhibited significant C3 upregulation compared to non-responders (2.2-fold, p = 0.019 and p = 0.003, respectively) (Fig. 4b). Extreme responders uniquely displayed enhanced CCL11 expression (2.1-fold, p = 0.012 versus non-responders). IL-2 administration (B2W4) generated marked transcriptional differences (Fig. 4c, 4d). Extreme responders demonstrated coordinated upregulation of pro-inflammatory chemokines compared to non-responders, including IL-8 (5.6-fold, p = 0.007), IL-1β (3.1-fold, p = 0.012), CXCL1/2/3 (2.3-2.8-fold, p < 0.03), and CCL7 (2.1-fold, p = 0.0004). Treatment responders exhibited significant CXCL9 upregulation (1.7-fold in extreme and 2.0-fold in intermediate responders, p < 0.04). Intermediate responders showed greater expression of IL-8 (9.8-fold, p = 0.006), FOS (4.5-fold, p = 0.026), CSF1 (1.6-fold, p = 0.01) and IL-1α (2.2-fold, p = 0.026) compared to extreme responders. Durable responses may depend not just on peak inflammation, but on sustained innate/adaptive immune engagement and complement activation. To identify differential treatment response pathways between groups, we performed difference-in-differences analyses comparing gene expression changes during sequential treatment phases. During the IL-2 treatment phase, extreme responders showed significantly enhanced upregulation of PKC isoforms (PRKCA: Δlog₂[FC] = 0.72, p = 0.005; PRKCB: Δlog₂[FC] = 0.40, p = 0.007), BCL6 (Δlog₂[FC] = 0.74, p = 0.014), MEF2A (Δlog₂[FC] = 0.44, p = 0.003), and TGFB1 (Δlog₂[FC] = 0.79, p = 0.016) compared to non-responders. Extreme responders also demonstrated enhanced TGFBR1 upregulation versus intermediate responders (Δlog₂[FC] = 0.56, p = 0.019). Following IL-2 discontinuation, extreme responders showed upregulation of inflammatory mediators (IL1A: Δlog₂[FC] = 1.56, p = 0.026; IL17A: Δlog₂[FC] = 1.29, p = 0.022) and complement factors (C5, C6, CFB: Δlog₂[FC] = 0.55-1.73, p < 0.03) compared to intermediate responders. Extreme responders exhibited both early and sustained transcriptional programs marked by innate immune activation, chemokine signaling, PKC and TGF-β pathway engagement, and complement activation, which differentiated them from intermediate responders. These findings suggest IL-2–induced immune remodeling as a critical determinant of therapeutic depth and durability. Peripheral Blood Immune Profiling Identifies Patterns Associated with Response Groups Flow cytometry analysis of peripheral blood immune cell populations was performed to identify immune states associated with durable treatment-free survival. Pooled analysis across all timepoints revealed that the frequency of CD16+ natural killer (NK) cells within the total circulating immune cell population was markedly elevated in treatment responders compared to non-responders, with intermediate responders demonstrating a 6.05-fold higher frequency (q = 1.77×10 -4 ) and extreme responders showing a 4.20-fold higher frequency (q = 1.86×10 -2 ) (Fig. 5a). Additionally, circulating CD4+ helper T-cell and CD19+ B-cell frequencies were elevated in extreme compared to intermediate responders (1.38-fold, q = 1.11×10 -4 ; 1.93-fold, q = 4.85×10 -4 , respectively; Fig. 5b,c). Longitudinal analysis revealed treatment-induced immune dynamics that varied by response group. Extreme responders demonstrated significant increases in CD15+ myeloid-derived suppressor cell (MDSC) populations from baseline to post-IL-2 (B1W1 to B2W4: 5.03-fold, p = 0.0391). Intermediate and non-responders exhibited no statistically significant temporal changes in any immune population analyzed. Among 25/26 patients (96.1%) with evaluable PD-1 expression data across the four collection timepoints, we observed distinct patterns that correlated with clinical outcomes. Longitudinal analysis revealed that patients achieving extreme response demonstrated suppression of PD-1+ cell frequencies across B1W1 through B4W1. In contrast, non-responders demonstrated more variable PD-1 expression levels throughout treatment (Fig. 5d). Furthermore, baseline PD-1+ expression levels demonstrated elevated median expression in non-responders compared to extreme and intermediate responders across T-cell populations, including live PD-1+ T-cells, FoxP3+ regulatory T-cells (Treg/PD-1+), helper T-cells (Th/PD-1+) and cytotoxic T-cells (Tc/PD-1+) (Fig. 5e). While no individual results achieved statistical significance, these patterns were consistently observed across T-cell subsets. 3. Discussion Despite improved outcomes with the advent of immunotherapy, advanced renal cell carcinoma represents a significant therapeutic challenge. Current standard-of-care treatment approaches, including dual checkpoint inhibition or checkpoint inhibitor-tyrosine kinase inhibitor combinations, often require prolonged or indefinite therapy resulting in cumulative treatment-related morbidity, economic burden, and impact on quality of life. The implications of prolonged therapy are substantial, with contemporary studies demonstrating that immunotherapy combinations incur significant annual healthcare costs and increase emergency department visits and hospitalizations. 16,26–28 Moreover, a recent systematic review and meta-analysis identified patient-reported outcomes as an independent prognostic predictor of overall survival, highlighting the importance of treatment approaches that preserve quality of life. 29 These findings underscore the need for treatment strategies that can achieve durable disease control without requiring indefinite therapy administration. Consequently, treatment-free survival has emerged as a patient-centered endpoint that incorporates dual goals of durable disease control and freedom from ongoing treatment burden. 17,18,30,31 The present study evaluated survival outcomes following a novel short-course regimen of combination ICI-cytokine therapy, in the form of pembrolizumab plus IL-2, in patients with advanced ccRCC. This study highlights long-term follow up beyond a median of 6 years, allowing for longitudinal assessment of the durable response in extreme responders clinically. Our results demonstrate that a finite course of pembrolizumab plus IL-2 can achieve exceptional treatment-free survival outcomes, with 42.3% (95% CI: 23.5-60.0%) of patients remaining free of both disease progression and subsequent therapy administration after over 6 years of follow-up. Notably, these patients experienced no persistent grade ≥2 toxicities, suggesting that durable disease control can be achieved without the chronic morbidity associated with continuous immunotherapy approaches. 32,33 After a median follow-up duration of 76.4 months, 11/26 (42.3%) patients achieved CR, 8/26 (30.8%) achieved PR, and 5/26 (19.2%) maintained SD, contributing to a DCR of 92.3%. The pembrolizumab plus IL-2 regimen demonstrated a complete response rate substantially higher than other contemporary combination immunotherapy regimens, which have reported CR rates ranging from 8-16%. 7,34,35 Moreover, TFS rates were durable over long-term observation, with 76.9% (95% CI: 55.7-88.9%) of patients remaining treatment-free at 1 year from therapy completion, and 42.3% (95% CI: 23.5-60.0%) maintaining treatment-free status at both the 3- and 5-year follow-up timepoints. While direct cross-trial comparisons require caution due to differences in patient populations and study designs, our treatment-free survival outcomes compare favorably to contemporary standards, despite our study’s more limited sample size. The 5-year landmark analysis of CheckMate 214 demonstrated that, following two years of ipilimumab plus nivolumab treatment, 18% of patients maintained treatment-free survival at 5 years, with a mean allocation of 16.0 months on protocol therapy, 11.1 months treatment-free, and 13.7 months surviving after subsequent therapy initiation. 18 The 43-month follow-up analysis of the KEYNOTE-426 trial of pembrolizumab plus axitinib versus sunitinib reported a median of 20 pembrolizumab administrations and 15 (range: 0.03–48) months of axitinib. 36 While TFS was not a systematically reported endpoint from the trial, 204/432 patients (47.2%) in the combination arm received subsequent therapy at any time following study treatment discontinuation. 36 In contrast, the pembrolizumab plus IL-2 regimen required substantially less treatment exposure with longer treatment-free intervals. Patients received a median of 12 pembrolizumab doses and 25.5 IL-2 doses in this study. Following discontinuation of treatment, the median TFS time was 23.8 months with 42.3% of patients remaining treatment-free after 5 years. Moreover, compared to previous trials of high-dose IL-2 monotherapy, our trial demonstrates superior response rates. A study of 155 mRCC patients treated with high dose IL-2 monotherapy found a CR rate of 7% and ORR of 21%, while a contemporary retrospective analysis of 810 mRCC patients from the PROCLAIM registry, most of whom were treated with the 14-doses-in-a-row schedule, found a CR rate of 5% and ORR of 24.7%. 37,38 In contrast, our pembrolizumab plus IL-2 combination regimen achieved a CR rate of 42.3% and ORR of 73.1%, representing approximately 5-fold and 3-fold improvements, respectively. Nevertheless, these encouraging signals must be interpreted within the context of our single-arm design and limited sample size. Prospective randomized comparison would be required to definitively establish superiority over alternative treatment approaches. Our findings of durable response with pembrolizumab plus high-dose IL-2 are particularly notable when contextualized with data from the phase 3 PIVOT-09 trial of BEMPEG plus nivolumab (BEMPEG+nivo) versus TKI in advanced ccRCC. 24 Despite both approaches targeting the IL-2 pathway with PD-1 inhibition, BEMPEG plus nivolumab achieved a 23.0% ORR and 2.0% complete response rate in intermediate/poor-risk patients, while our sequential approach delivered substantially improved response and survival outcomes across IMDC risk categories. 24 These divergent outcomes likely result from differences in immune receptor engagement and treatment administration strategy. While BEMPEG demonstrates selective CD122/CD132 binding to preferentially activate CD8+ T-cells and NK-cells over Tregs, aldesleukin engages the trimeric IL-2 receptor (CD25/CD122/CD132). 39,40 This broader receptor engagement may provide more potent stimulation to tumor-reactive CD8+ T cells that transiently upregulate CD25 during activation, supporting their optimal expansion, function, and memory formation. 40–43 Interestingly, Eisner and colleagues demonstrated that responders to IL-2 therapy paradoxically exhibited higher baseline expression of immunosuppressive markers (such as MDSCs, Tregs, and neutrophils) that typically predict poor response to PD-1 inhibition. 44 Our sequential approach with pembrolizumab priming followed by high-dose IL-2 may more effectively leverage the distinct and complementary mechanisms of ICI and cytokine therapies than the simultaneous administration of BEMPEG+nivo in PIVOT-09. Our correlative analyses, enabled by over six years of follow-up on this phase 2 trial, provide hypothesis-generating insights into mechanisms underlying durable treatment-free survival through characterization of extreme responders and their distinct biomarker profiles. Although only a portion of individual biomarkers survived multiple testing correction, convergent evidence from flow cytometry, transcriptomic, and proteomic platforms suggests the presence of biologically coherent patterns warranting further validation. Longitudinal flow cytometry revealed that extreme responders demonstrated suppression of PD-1+ cell frequencies in peripheral blood across T-cell subsets from baseline through completion of treatment, while non-responders demonstrated variable PD-1 population frequencies. This PD-1 suppression may suggest successful immune reconditioning, with chronically exhausted T-cells transitioning toward functional competence. Notably, both extreme and intermediate responders demonstrated marked enrichment of circulating CD16+ NK-cells which may serve as effector mechanisms for sustained antitumor immunity. Recent single-cell analyses in ccRCC have demonstrated that advanced disease progression is characterized by depletion of functional, cytotoxic NK-cells and enrichment of dysfunctional, tissue-resident NK-cell populations with impaired degranulation capacity. 45 We highlight that the preservation or enhancement of circulating CD16+ NK-cells in extreme responders may suggest maintenance of cytotoxic immune surveillance capabilities that could contribute to durable disease control. While the expansion of CD15+ MDSCs may traditionally be associated with immune suppression, their increase exclusively in extreme responders during IL-2 therapy may reflect transient neutrophilic activation or immune homeostasis, rather than functional suppression. Our findings support a mechanistic model of controlled immune reconditioning, whereby PD-1 blockade facilitates immune reinvigoration without overstimulation or clearance of exhausted lymphocyte populations, allowing for IL-2–mediated immune amplification in a biologically contained and clinically tolerable fashion. 46 These immune cell population changes aligned with molecular signatures in extreme responders. Temporal analysis revealed distinct gene expression signatures that differentiated response groups, with extreme responders demonstrating coordinated upregulation of pro-inflammatory chemokines following IL-2 administration, suggesting enhanced T-cell trafficking. Moreover, time-independent comparisons of response groups identified elevated immune recruitment chemokines (such as IL-8 and CXCL1/2) combined with reduced cellular stress markers (such as MIC-A/B and GMZA) in extreme responders, potentially representing optimal immune homeostasis for sustained antitumor surveillance. 47 A study of ipilimumab in hematologic cancers similarly demonstrated chemokine upregulation (CXCL2, IL-8) and reduced soluble MIC-A in treatment responders, supporting conserved mechanisms of immune cell recruitment and cellular stress modulation. 48 Furthermore, a pharmacodynamic study of PD-1 blockade demonstrated that nivolumab induces CXCL9 and CXCL10 production in mRCC tumors. 23 The sequential treatment strategy of this study may play a role in achieving immune homeostasis, with pembrolizumab initially rescuing exhausted T-cells from PD-1-mediated suppression, creating permissive conditions for subsequent IL-2-mediated activation without overwhelming the overall system. This controlled immune reconditioning could explain both the durable treatment-free survival and the favorable toxicity profile compared to high-dose IL-2 monotherapy or prolonged checkpoint inhibition. This immune remodeling was supported by multi-omic signals including innate immune priming via CD16⁺ NK-cells, sustained chemokine and cytokine induction, and enhanced expression of PKC and TGF-β pathway genes. The reduction in cellular stress ligands such as MIC-A/B and downregulation of PD-1 expression further suggest an immunologically poised but restrained state, enabling effective tumor control without immune toxicity. While further validation in independent cohorts is needed, these findings suggest a biological framework for finite combination immunotherapy through immune-checkpoint inhibition and cytokine chemotaxis and identify potential biomarkers for patient selection and treatment optimization. Recent work has highlighted the potential of IL2RG-related gene signatures as prognostic and predictive biomarkers in ccRCC, which could be further explored in the context of pembrolizumab plus IL-2 therapy to identify patients most likely to achieve durable response. 49 Several important limitations must be acknowledged. The non-randomized, single-arm design of this study prevents definitive attribution of outcomes to the combination regimen versus individual components or patient selection factors. Our study population of 26 patients is appropriate for initial statistical exploration of biomarkers, rather than generalizability to broader populations. While we had a limited number of patients with poor-risk disease, the distribution between favorable and intermediate risk groups is reflective of the real-world IMDC risk distribution observed in clinical practice for RCC patients. 50 Our single-center experience may not reflect outcomes achievable in broader community practice, particularly given the specialized requirements for IL-2 administration. 51 However, for appropriately selected patients, the potential for durable treatment-free survival may justify additional upfront resource intensity, particularly when weighed against the cumulative burden and costs of indefinite therapy. Future studies could explore less toxic IL-2 receptor-targeted cytokines such as IL-2 fusion proteins or monoclonal antibody complexes that may maintain beneficial immunologic effects while reducing specialized administration requirements. 52 In summary, our findings demonstrate that time-limited pembrolizumab plus IL-2 therapy is not only feasible and tolerable, but also capable of delivering durable clinical benefit, with high objective and complete response rates, prolonged overall survival, and extended treatment-free survival. These results establish a clinical and mechanistic rationale for time-limited immunotherapy that is capable of inducing sustained tumor control. Future studies should validate biomarker signatures in independent cohorts, develop immune profiling–based patient stratification tools, and incorporate randomized comparisons with standard-of-care regimens. Importantly, integration of patient-reported outcomes and health economic analyses will be essential to fully characterize the clinical and societal value of finite immunotherapy strategies with long-term disease control. 4. Methods Patients Eligible patients were aged 18 years or older with a diagnosis of metastatic renal cell carcinoma containing a clear cell component and with measurable disease per Response Evaluation Criteria in Solid Tumors (RECIST; version 1.1). Patients were required to have an Eastern Cooperative Oncology Group (ECOG) performance status of 0 or 1 and adequate organ and marrow function. Additionally, patients were required to undergo exercise or thallium stress testing (unless deemed exempt after evaluation by a cardiologist) and pulmonary function testing within 3 months of treatment initiation. Patients were excluded from participation in the study if they had active brain metastases, active autoimmune disease requiring systemic treatment within the past 2 years, received more than 1 systemic therapy in the past year, IL-2 therapy within 1 year, or anti-PD-1/PD-L1 agents at any time. Comprehensive inclusion and exclusion criteria are detailed in the study protocol in Supplementary Information. All patients provided written informed consent before any screening or study procedures took place. Trial Design and Treatments This was a single-arm, single-institution phase 2 trial (ClinicalTrials.gov identifier: NCT02964078) investigating pembrolizumab and high dose IL-2 for the treatment of advanced clear cell renal cell carcinoma. The trial was conducted at the H. Lee Moffitt Cancer Center (Tampa, FL) with the approval of the Moffitt Cancer Center institutional review board (protocol: MCC-18536). The study architecture involved a maximum of four treatment blocks, each comprised of three 200 mg doses of pembrolizumab administered intravenously at 21 (±3) -day intervals. During the first and fourth blocks, patients were treated with pembrolizumab monotherapy. During the second and third blocks, two cycles of 5-in-a row high-dose (HD) IL-2 were administered between consecutive pembrolizumab doses. Each cycle of HD IL-2 consisted of up to five 600,000 IU/kg doses of IL-2 given every 8 hours for a total of 20 doses planned for block 2 and block 3, respectively. The protocol did not permit replacement of missed treatment doses. Computed tomography (CT) radiographic assessment of disease was obtained at baseline prior to treatment, and subsequent assessments were obtained at 9-week intervals while on treatment and at 2–3-month intervals for at least two years or until progression of disease after discontinuing treatment. Tumor-response assessments were performed at each imaging timepoint according to Response Evaluation Criteria in Solid Tumors (RECIST) 1.1 criteria. Patients achieving a complete response (CR) at the first RECIST assessment timepoint were eligible to continue with pembrolizumab monotherapy. Patients with stable or progressive disease at this timepoint proceeded to treatment block 2. After completing the first treatment block, patients were permitted to proceed to subsequent treatment blocks in the absence of disease progression or intolerable toxicity. Peripheral blood samples for correlative analyses were collected at baseline (B1W1), after four cycles of pembrolizumab (B2W2), after 10 doses of IL-2 (B2W4), and after IL-2 was completed (B4W1). Trial Oversight A Protocol Monitoring Committee (PMC) was responsible for monitoring the safety data from this study as per the standard Moffitt Cancer Center Data Safety Monitoring Plan for investigator-initiated studies. The PMC monitors adverse event reporting, data and safety monitoring, and internal audit findings. Upon review, the PMC may approve a study for continuation, recommend revisions, or suspend or close the protocol. Pembrolizumab was provided by Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc., Kenilworth, NJ, USA. The conduct of this investigation abided by all applicable local regulatory requirements and was performed in accordance with the Good Clinical Practice (GCP) Guidelines of the International Council for Harmonization of Technical Requirements for Pharmaceuticals for Human Use (ICH) and the Declaration of Helsinki. All authors confirm that the trial was conducted with respect to the standards of Good Clinical Practice. All authors had access to the data, participated in the writing and reviewing of this manuscript, approved the submitted manuscript, and confirmed the accuracy and completeness of the data. Statistical Analyses Following a Simon 2-stage minimax design, the study evaluated the first 15 patients for objective response, with early termination specified if ≤ 3 responses were observed (stage 1 futility boundary). Upon passing this threshold, accrual continued to a total of 24 evaluable patients. The null hypothesis (objective response rate ≤ 20%) would be rejected in favor of the alternative hypothesis (response rate ≥ 45%) if ≥ 8 cumulative responses were observed across both stages (alpha=0.10, power=0.90). Dose-limiting toxicities were monitored using a sequential boundary method that accommodated variable follow-up durations as enrollment progressed. 53 The design implemented a Pocock-type stopping boundary with maximum 5% probability of boundary crossing when the true dose-limiting toxicity rate was at the pre-specified acceptable threshold of 15%. All patients who received at least one dose of pembrolizumab were included in the safety analysis. Outcomes and Assessments The primary objective of this study was twofold: to evaluate anti-tumor activity, targeting an objective response rate (ORR) ≥45% in the ever-treated population, and to assess treatment safety using a sequential toxicity boundary method. Secondary objectives were survival endpoints (assessed using Kaplan-Meier analysis) including overall survival (OS), measured from enrollment to death from any cause; progression-free survival (PFS), defined as time from enrollment to first documented disease progression per RECIST 1.1 criteria or death from any cause; and treatment-free survival (TFS), calculated from completion of study treatment to initiation of next therapy, death, or last follow-up (in patients not receiving subsequent therapy). Exploratory endpoints included correlative biomarker analyses. All tumor-response assessments were performed in accordance with RECIST 1.1 criteria. Complete response was defined as disappearance of all target and non-target lesions. Partial response was defined as a 30% or more decrease in the sum of diameters of target lesions, taking as a reference the baseline sum diameters. Progressive disease was defined as a ≥20% increase in the sum of diameters of target lesions and unequivocal progression of existing non-target lesions or appearance of a new lesion. Stable disease was defined as insufficient increase or decrease to qualify as complete or partial response or disease progression. Correlative Analyses Flow cytometry staining and analysis Peripheral blood was collected from clinical trial participants in up to three 10 mL sodium heparin vacutainers. The blood was then transferred and centrifuged for 5 minutes at 500xg to collect plasma. This plasma was stored at -80°C. Peripheral blood mononuclear cells (PBMCs) were then isolated from the cell pellet using Ficoll-Paque density gradient centrifugation, washed twice in phosphate-buffered saline (PBS) with 2% fetal bovine serum (FBS), and then cryopreserved using Recovery Freeze Media (Thermo Fisher Scientific, Waltham, Massachusetts), overnight controlled rate freezing and then LN2 storage. For flow batch runs, cells were rapidly thawed at 37°C, washed twice with Fc blocking buffer (5% Human Serum inactivated) and resuspended at 1 × 10 6 cells/mL in staining buffer. A multicolor flow cytometry panel was designed to assess immune cell populations, using titrated antibodies against CD19 (SJ25C1, BUV496), HLA-DR (G46-6, BV650), CD14 (M5E2, BUV805), CD15 (W6D3, BV711), CD45 (HI30, BV786), CD11b (D12, BB700), CD33 (WM53, PE), CD279/PD-1 (J43, BUV737), CD274/PD-L1 (MIH1, PE-Cy7), CD56/NCAM (NCAM16.2, BV605), CD16 (B73.1, APC), CD3 (UCHT1, BUV395), CD8 (HIT8α, R718), CD4 (SK3, BV480), CD127/IL-7Rα (HIL-7R-M21, BV421), CD25/IL-2Rα (M-A251, BB515), FOXP3 (259D/C7, PE-CF594), and Fixable Viability Stain 780 (all from BD Biosciences, Franklin Lakes, New Jersey). Antibody concentrations were optimized via serial dilutions on PBMCs from normal donor leukopaks (purchased from AllCells, Huntsville, Alabama) to maximize signal-to-noise ratio. Cells were stained with antibody cocktails for 20 minutes at 4°C, with FOXP3 staining performed post-fixation/permeabilization using the BD Cytofix/Cytoperm kit, and viability staining applied prior to fixation. Single-color and fluorescence-minus-one (FMO) controls, prepared using leukopak-derived PBMCs, were used to establish compensation matrices and gating boundaries, respectively, with rainbow beads (Spherotech, Lake Forest, Illinois) ensuring instrument standardization. Samples were acquired on a BD FACSSymphony A5 (LSRFortessa) with BD FACSDiva Software (Version 8.0.1), collecting at least 50,000 events per sample. Data were analyzed in FCS Express 7 (De Novo Software, Pasadena, California), using FMO-guided hierarchical gating to identify T cells (CD3+), B cells (CD19+), monocytes (CD14+), neutrophils (CD15+), and NK cells (CD56+CD16+), with dead cells excluded via viability staining. Expression of activation (HLA-DR, PD-1, PD-L1) and regulatory markers (CD25, FOXP3) was quantified as mean fluorescence intensity or percentage of positive cells. See Extended Data Fig. 1 for flow cytometry sequential gating/sorting strategies. Transcriptomic and Proteomic Assays Transcriptomic (NanoString nCounter® PanCancer Immune Profiling Panel for Gene Expression) and proteomic (Olink® Target 96 Cardiometabolic Analysis Service USD, Target 96 Immuno-Oncology Analysis Service USD and Target 96 Oncology II Analysis Service) assays were performed. NanoString gene expression data were analyzed using nSolver 4.0 software, with normalization performed against six housekeeping genes (PGK1, CLTC, HPRT1, GUSB, GAPDH, TUBB). Classification of Response Groups by TFS To classify patient response, we established discrete categories based on duration of treatment-free survival. Upon examination of the TFS distribution in our cohort, three distinct clusters of patients emerged with natural separation points at approximately 6 months and 5 years. Based on these observed groupings, we classified patients as extreme responders (TFS >5 years), intermediate responders (TFS between 6 months and 5 years), and non-responders (TFS <6 months). This classification approach was determined through inspection of the TFS distribution, which revealed clear multi-modal separation. These groupings were established prior to molecular analyses to avoid bias in the interpretation of gene expression patterns. Time Independent Molecular Analyses Differential gene and protein expression analysis was performed using the Mann-Whitney U test for non-parametric comparisons or Welch's t-test when statistical assumptions were met (normality via Shapiro-Wilk test, equal variance via Levene's test, n ≥ 15). Gene Set Enrichment Analysis (GSEA) Comprehensive pathway enrichment analysis was performed using the GSEApy implementation of Gene Set Enrichment Analysis. 54,55 For each platform and comparison, ranked gene/protein lists were generated based on log2 fold change values. This ranking approach maintains the directionality of expression changes while being compatible with the non-parametric differential expression analysis. Pre-ranked GSEA was conducted using GSEApy version 1.1.8 with the following parameters: Gene ranking: Log2 fold change of difference in group means Permutation number: 5,000 Minimum gene set size: 15 Maximum gene set size: 500 Weighted score type: 1 Random seed: 42 for reproducibility Gene set databases interrogated included Gene Ontology Biological Process 2021, Gene Ontology Molecular Function 2021, KEGG 2021 Human, WikiPathway 2021 Human, and Reactome 2022. Pathway significance was assessed using nominal p-values and tiered FDR q-values, with pathways considered significantly enriched at FDR < 0.05, nominally enriched at FDR < 0.10, and exploratory enrichment at FDR < 0.25. Time-Dependent Molecular Analyses To characterize dynamic patterns of immune gene expression across treatment phases, we performed comprehensive time-dependent analyses using multiple complementary approaches. Control genes (e.g., NEG_, POS_) and housekeeping genes (PGK1, CLTC, HPRT1, GUSB, GAPDH, TUBB) were systematically excluded from analyses. Data were pooled across all patients regardless of response category to identify general treatment effects. For each treatment phase (pembrolizumab monotherapy: B2W2 vs B1W1; IL-2 addition: B2W4 vs B2W2; post-IL-2: B4W1 vs B2W4), paired t-tests were conducted for genes with sufficient paired expression data (≥3 paired samples per timepoint comparison). Multiple testing correction was applied using the Benjamini-Hochberg false discovery rate (FDR) method. Genes were considered significantly altered if they met both statistical (FDR 0.6) significance thresholds. This pooled approach allowed identification of consistent transcriptional changes induced by each treatment phase across the entire cohort. We performed cross-sectional transcriptomic analyses to reveal distinct gene expression patterns between response groups at specific timepoints. At each timepoint (B1W1, B2W2, B2W4, B4W1), expression levels were compared between extreme, intermediate, and non-responders using either Mann-Whitney U tests (non-parametric) or Welch's t-tests based on statistical assumption testing (normality via Shapiro-Wilk test, equal variance via Levene's test). A minimum of 3 samples per group was required for analysis. Following pembrolizumab monotherapy (B2W2), comparisons identified early molecular signatures associated with treatment outcomes. Similarly, after IL-2 administration (B2W4), analyses determined whether response groups exhibited different gene expression profiles potentially associated with IL-2 responsiveness. Expression differences were considered significant at p < 0.05, with FDR correction applied for multiple testing across genes within each comparison. To investigate whether temporal expression patterns differed between response groups, we performed difference-in-differences analyses comparing the magnitude of expression change between response groups. For each treatment transition (pembrolizumab effect: B2W2 vs B1W1; IL-2 effects: B2W4 vs B2W2; post-IL-2 effect: B2W4 vs B4W1; overall effect: B4W1 vs B1W1), we calculated individual patient treatment effects as the change in gene expression between timepoints. Treatment effects were then compared between response group pairs (extreme vs non-responders, intermediate vs non-responders, extreme vs intermediate responders) using Welch's t-tests. This approach identified genes where the treatment response magnitude differed significantly between response categories, requiring a minimum of 2 patients per group with paired timepoint data. Missing data were handled using available case analysis, including all patients with data at each timepoint rather than restricting to complete cases. For patients with multiple samples at the same timepoint, duplicate measurements were aggregated using median values. Longitudinal analyses only included patients with paired data at both relevant timepoints for each comparison. For all time-dependent analyses, statistical significance was assessed at α = 0.05. Effect sizes were calculated as Cohen's d for t-tests and fold change metrics for biological interpretation. Individual patient trajectories were tracked across timepoints with group means and standard errors calculated. Results were stratified into significance tiers: highly significant (FDR < 0.05), nominally significant (FDR < 0.10), and exploratory (FDR < 0.25) given the nature of this biomarker discovery study. Flow Cytometry Analyses Nine immune markers (CD4, CD8, CD19, CD16, CD25, CD15, CD33, CD14, CD11b) were analyzed using four measurement types: cell frequency as percentage of parent population, percentage of total cells, median fluorescence intensity (MFI), and geometric mean fluorescence. Log₂ transformation was applied to fluorescence measurements to normalize distributions. Between-group comparisons used the Mann-Whitney U test (for n < 15 or non-normal distributions) or Welch's t-test following statistical assumption testing (Shapiro-Wilk, Levene's tests). A minimum of 5 samples per group was required. Multiple testing correction used the Benjamini-Hochberg FDR for between-group comparisons and Bonferroni correction for pairwise temporal comparisons. Effect sizes included fold change, Cohen's d, Glass's delta, and Cliff's delta. For baseline comparisons of PD-1 across treatment response groups, Kruskal-Wallis tests assessed overall group differences, followed by pairwise Mann-Whitney U tests when significant (p < 0.1 exploratory threshold). For longitudinal analysis of subjects with complete four-timepoint data (B1W1, B2W2, B2W4, B4W1), Friedman's non-parametric repeated measures ANOVA tested overall temporal changes. Duplicate measurements were aggregated using median values. Pairwise temporal comparisons used Wilcoxon signed-rank tests for all timepoint combinations (B1W1 vs. B2W2, B2W2 vs. B2W4, B2W4 vs. B4W1, B2W2 vs. B4W1, and B4W1 vs. B1W1). Effect sizes were calculated as r = |Z|/√N, where Z is the standardized test statistic. For all flow cytometry analyses, statistical significance was assessed at α = 0.05. Individual patient trajectories were tracked across timepoints with group means and standard errors calculated. Effect sizes and confidence intervals accompanied p-values for clinical interpretation. Software Specifications, Data Processing, and Quality Control Statistical analyses and figure generation were performed with R version 4.0.2 and Python version 3.11.5 using the pandas (v2.1.4), numpy (v1.24.3), scipy.stats (v1.15.2), statsmodels (v0.14.0), matplotlib (v3.7.2), seaborn (v0.13.2), scikit-learn (v1.3.0), and GSEApy (v1.1.8) modules. Platform-specific quality control included: (1) assessment of data completeness and missing value patterns, (2) parameter availability across timepoints, (3) sample size distribution validation, and (4) control gene identification and exclusion. The code used for data analysis is available at https://github.com/MCC18536/Pem-IL2/. Declarations Data Availability Aggregate patient-related information has been made available on ClinicalTrials.gov (https://clinicaltrials.gov/ct2/show/results/NCT02964078). Additional requests for raw and analyzed data can be referred to the corresponding author (J.C.) and will be reviewed promptly as part of a data transfer agreement per H. Lee Moffitt Cancer Center institutional policies, to determine whether the request is subject to any clinical trial patient confidentiality or intellectual property requirements. Funding Financial support by contract with Clinigen, Inc. Pembrolizumab furnished by Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc., Kenilworth, NJ, USA. IND and regulatory support through Moffitt Cancer Center. Role of the Funding Source The funder of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report. Reporting Summary Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article. Acknowledgements We thank the patients and their families, investigators, coinvestigators, and the study team for their support and contributions to this research. We thank Michael Holman (Prometheus) for his advocacy for the project at its early stages. References Siegel, R. L., Kratzer, T. B., Giaquinto, A. N., Sung, H. & Jemal, A. Cancer statistics, 2025. CA Cancer J Clin (2025) doi:10.3322/caac.21871. Znaor, A., Lortet-Tieulent, J., Laversanne, M., Jemal, A. & Bray, F. International variations and trends in renal cell carcinoma incidence and mortality. Eur Urol 67 , 519–530 (2015). Rathmell, W. K. et al. Management of Metastatic Clear Cell Renal Cell Carcinoma: ASCO Guideline. J Clin Oncol 40 , 2957–2995 (2022). Tannir, N. M. et al. Nivolumab plus ipilimumab versus sunitinib for first-line treatment of advanced renal cell carcinoma: extended 8-year follow-up results of efficacy and safety from the phase III CheckMate 214 trial. Ann Oncol 35 , 1026–1038 (2024). Rini, B. I. et al. Pembrolizumab plus Axitinib versus Sunitinib for Advanced Renal-Cell Carcinoma. New England Journal of Medicine 380 , 1116–1127 (2019). Choueiri, T. K. et al. Nivolumab plus Cabozantinib versus Sunitinib for Advanced Renal-Cell Carcinoma. New England Journal of Medicine 384 , 829–841 (2021). Motzer, R. J. et al. Nivolumab plus Ipilimumab versus Sunitinib in Advanced Renal-Cell Carcinoma. New England Journal of Medicine 378 , 1277–1290 (2018). Rini, B. I. et al. Pembrolizumab plus axitinib versus sunitinib for advanced clear cell renal cell carcinoma: 5-year survival and biomarker analyses of the phase 3 KEYNOTE-426 trial. Nat Med 1–10 (2025) doi:10.1038/s41591-025-03867-5. Motzer, R. J. et al. Nivolumab plus cabozantinib versus sunitinib in first-line treatment for advanced renal cell carcinoma (CheckMate 9ER): long-term follow-up results from an open label, randomised, phase 3 trial. Lancet Oncol 23 , 888–898 (2022). Motzer, R. J. et al. Characterization of Responses to Lenvatinib plus Pembrolizumab in Patients with Advanced Renal Cell Carcinoma at the Final Prespecified Survival Analysis of the Phase 3 CLEAR Study. Eur Urol 86 , 4–9 (2024). de Groot, S. et al. Health-related quality of life and its determinants in patients with metastatic renal cell carcinoma. Qual Life Res 27 , 115–124 (2018). Cronin, C., O’Sullivan, A. & Naidoo, J. Long-Term Toxicity of Immune Checkpoint Inhibitors—Time to Widen the Lens. JAMA Oncology 11 , 371–372 (2025). Khan, H. M., Ramsey, S. & Shankaran, V. Financial Toxicity in Cancer Care: Implications for Clinical Care and Potential Practice Solutions. JCO 41 , 3051–3058 (2023). Gupta, A., Eisenhauer, E. A. & Booth, C. M. The Time Toxicity of Cancer Treatment. JCO 40 , 1611–1615 (2022). George, S. et al. The Impact of Adverse Events on Health Care Resource Utilization, Costs, and Mortality Among Patients Treated with Immune Checkpoint Inhibitors. Oncologist 26 , e1205–e1215 (2021). Shah, N. J. et al. Health Care Resource Use for Modern First-Line Treatments in Metastatic Renal Cell Carcinoma. JAMA Network Open 7 , e2422674 (2024). Regan, M. M. et al. Treatment-Free Survival: A Novel Outcome Measure of the Effects of Immune Checkpoint Inhibition-A Pooled Analysis of Patients With Advanced Melanoma. J Clin Oncol 37 , 3350–3358 (2019). Mantia, C. M. et al. Treatment-free survival and partitioned survival analysis of patients with advanced renal cell carcinoma treated with nivolumab plus ipilimumab versus sunitinib: 5-year update of CheckMate 214. J Immunother Cancer 12 , e009495 (2024). Klapper, J. A. et al. High-dose interleukin-2 for the treatment of metastatic renal cell carcinoma. Cancer 113 , 293–301 (2008). Atkins, M. B., Regan, M. & McDermott, D. Update on the role of interleukin 2 and other cytokines in the treatment of patients with stage IV renal carcinoma. Clin Cancer Res 10 , 6342S–6S (2004). Muhitch, J. B. & and Schwaab, T. High-dose IL-2 for Metastatic Renal Cell Carcinoma: Can the First Antitumor Immunotherapy be Reinvented? Immunotherapy 6 , 955–958 (2014). Khoja, L., Butler, M. O., Kang, S. P., Ebbinghaus, S. & Joshua, A. M. Pembrolizumab. Journal for ImmunoTherapy of Cancer 3 , 36 (2015). Choueiri, T. K. et al. Immunomodulatory Activity of Nivolumab in Metastatic Renal Cell Carcinoma. Clin Cancer Res 22 , 5461–5471 (2016). Tannir, N. M. et al. Bempegaldesleukin Plus Nivolumab Versus Sunitinib or Cabozantinib in Previously Untreated Advanced Clear Cell Renal Cell Carcinoma: A Phase III Randomized Study (PIVOT-09). Journal of Clinical Oncology (2024) doi:10.1200/JCO.23.02082. Chatzkel, J. et al. Coordinated Pembrolizumab and High Dose IL-2 (5-in-a-Row Schedule) for Therapy of Metastatic Clear Cell Renal Cancer. Clinical Genitourinary Cancer 20 , 252–259 (2022). Kalinich, M. et al. Prediction of severe immune-related adverse events requiring hospital admission in patients on immune checkpoint inhibitors: study of a population level insurance claims database from the USA. J Immunother Cancer 9 , e001935 (2021). Zheng, Y. et al. Real-World Clinical and Economic Outcomes in Selected Immune-Related Adverse Events Among Patients with Cancer Receiving Immune Checkpoint Inhibitors. Oncologist 26 , e2002–e2012 (2021). Wang, L. X. et al. Health care utilization and steroid-refractory toxicities from immune checkpoint inhibitors. Cancer 126 , 322–328 (2020). Huang, R. S. et al. Patient-Reported Outcomes as Prognostic Indicators for Overall Survival in Cancer: A Systematic Review and Meta-Analysis. JAMA Oncol (2025) doi:10.1001/jamaoncol.2025.3153. Chang, E. et al. A Pooled Analysis of Treatment-Free Survival in Advanced Renal Cell Carcinoma. Clin Cancer Res 30 , 3282–3286 (2024). Gupta, M. et al. Treatment-Free Survival After Nivolumab vs Pembrolizumab vs Nivolumab-Ipilimumab for Advanced Melanoma. JAMA Netw Open 6 , e2319607 (2023). Fletcher, K. & Johnson, D. B. Chronic immune-related adverse events arising from immune checkpoint inhibitors: an update. J Immunother Cancer 12 , e008591 (2024). H, S. et al. Immune-related adverse events and kidney function decline in patients with genitourinary cancers treated with immune checkpoint inhibitors. European journal of cancer (Oxford, England : 1990) 157 , (2021). Choueiri, T. K. et al. Lenvatinib plus pembrolizumab versus sunitinib as first-line treatment of patients with advanced renal cell carcinoma (CLEAR): extended follow-up from the phase 3, randomised, open-label study. Lancet Oncol 24 , 228–238 (2023). Yanagisawa, T. et al. Updated systematic review and network meta-analysis of first-line treatments for metastatic renal cell carcinoma with extended follow-up data. Cancer Immunol Immunother 73 , 38 (2024). Plimack, E. R. et al. Pembrolizumab Plus Axitinib Versus Sunitinib as First-line Treatment of Advanced Renal Cell Carcinoma: 43-month Follow-up of the Phase 3 KEYNOTE-426 Study. European Urology 84 , 449–454 (2023). Yang, J. C. et al. Randomized Study of High - Dose and Low - Dose Inter leukin - 2 in Patients With Metastatic Renal Cancer. J Clin Oncol 21 , 3127–3132 (2003). Fishman, M. et al. Overall survival by clinical risk category for high dose interleukin-2 (HD IL-2) treated patients with metastatic renal cell cancer (mRCC): data from the PROCLAIMSM registry. j. immunotherapy cancer 7 , 84 (2019). Sharma, M. et al. Bempegaldesleukin selectively depletes intratumoral Tregs and potentiates T cell-mediated cancer therapy. Nat Commun 11 , 661 (2020). Charych, D. et al. Modeling the receptor pharmacology, pharmacokinetics, and pharmacodynamics of NKTR-214, a kinetically-controlled interleukin-2 (IL2) receptor agonist for cancer immunotherapy. PLoS One 12 , e0179431 (2017). Beebe, K. D. et al. The Immunogenomic Landscape of Peripheral High-Dose IL-2 Pharmacodynamics in Patients with Metastatic Renal Cell Carcinoma: A Benchmark for Next-Generation IL-2-Based Immunotherapies. J Immunol 213 , 29–39 (2024). LaPorte, K. M., Hernandez, R., Santos Savio, A. & Malek, T. R. Robust IL-2-dependent antitumor immunotherapy requires targeting the high-affinity IL-2R on tumor-specific CD8+ T cells. J Immunother Cancer 11 , e006611 (2023). Wu, W. et al. IL-2Rα-biased agonist enhances antitumor immunity by invigorating tumor-infiltrating CD25+CD8+ T cells. Nat Cancer 4 , 1309–1325 (2023). Eisner, J. R. et al. Distinct Predictive Immunogenomic Profiles of Response to Immune Checkpoint Inhibitors and IL2: A Real-world Evidence Study of Patients with Advanced Renal Cancer. Cancer Research Communications 2 , 894–903 (2022). Xu, W. et al. Progressive natural killer cell dysfunction in advanced-stage clear-cell renal cell carcinoma and association with clinical outcomes. ESMO Open 10 , 104105 (2025). Ando, S. & Araki, K. CD8 T-cell heterogeneity during T-cell exhaustion and PD-1-targeted immunotherapy. Int Immunol 34 , 571–577 (2022). Ha, H., Debnath, B. & Neamati, N. Role of the CXCL8-CXCR1/2 Axis in Cancer and Inflammatory Diseases. Theranostics 7 , 1543–1588 (2017). Davids, M. S. et al. Ipilimumab for Patients with Relapse after Allogeneic Transplantation. New England Journal of Medicine 375 , 143–153 (2016). Hu, W. et al. Prognostic and therapeutic relevance of IL2RG-related LncRNAs in clear cell renal cell carcinoma. Sci Rep 15 , 29651 (2025). Ernst, M. S. et al. Outcomes for International Metastatic Renal Cell Carcinoma Database Consortium Prognostic Groups in Contemporary First-line Combination Therapies for Metastatic Renal Cell Carcinoma. European Urology 84 , 109–116 (2023). Dutcher, J. P. et al. High dose interleukin-2 (Aldesleukin) - expert consensus on best management practices-2014. J Immunother Cancer 2 , 26 (2014). Tanigawa, K. & Redmond, W. L. Current landscape and future prospects of interleukin-2 receptor (IL-2R) agonists in cancer immunotherapy. Oncoimmunology 14 , 2452654. Ivanova, A., Qaqish, B. F. & Schell, M. J. Continuous Toxicity Monitoring in Phase II Trials in Oncology. Biometrics 61 , 540–545 (2005). Subramanian, A. et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A 102 , 15545–15550 (2005). Fang, Z., Liu, X. & Peltz, G. GSEApy: a comprehensive package for performing gene set enrichment analysis in Python. Bioinformatics 39 , btac757 (2023). Additional Declarations Yes there is potential Competing Interest. I have no competing interests. All disclosures are on page 1 of the manuscript. Supplementary Files SupplementalFigure1.pdf Gating Straregy Protocol.pdf Trial Protocol Cite Share Download PDF Status: Under Review Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7698261","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":526070845,"identity":"0645cc64-5745-43ff-a861-9ef7fe89fdd9","order_by":0,"name":"Jad Chahoud","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA+0lEQVRIiWNgGAWjYBAC9gYGA4afDQwJQLbhAwYGOX42QloYgVoYeyFajA2AWLKNjZmwFmZWiBYzCZCWBoJa2g9vYGbcYZfHL5G8repGhYEEn3z/AYaPe2pxa+lJK2BmPJNcLDkjrex2zhkDCTagLYwznh3H47AcA2bGNubEDTdyzG7ntv2pA2lh5jlwDLeW/jcgLfVgLcW5bRBb8GoRnAG0hbXtMFgLM5KWGpxapCWeFRzsbTueOLPnWbE0xC/JBgdnHDiAUwsff/LGBz/bqhP72ZM3fs4Bhph888GHDz4cqMOpBQQwDQSKHMarBSvAb8soGAWjYBSMKAAAQmhTWdCtdi8AAAAASUVORK5CYII=","orcid":"","institution":"Moffitt Cancer Center","correspondingAuthor":true,"prefix":"","firstName":"Jad","middleName":"","lastName":"Chahoud","suffix":""},{"id":526070846,"identity":"cc1e5d71-0cf7-43f9-9840-a2434fce5a15","order_by":1,"name":"Jeffrey Johnson","email":"","orcid":"","institution":"Moffitt Cancer Center","correspondingAuthor":false,"prefix":"","firstName":"Jeffrey","middleName":"","lastName":"Johnson","suffix":""},{"id":526070847,"identity":"8a0e6030-173e-42bc-8013-b11a125ca7ca","order_by":2,"name":"Justtin Miller","email":"","orcid":"","institution":"Moffitt Cancer Center","correspondingAuthor":false,"prefix":"","firstName":"Justtin","middleName":"","lastName":"Miller","suffix":""},{"id":526070848,"identity":"f783bd36-1995-4f0a-bf12-db3442dd20e0","order_by":3,"name":"Michael Schell","email":"","orcid":"https://orcid.org/0000-0003-4594-6601","institution":"Moffitt Cancer Center","correspondingAuthor":false,"prefix":"","firstName":"Michael","middleName":"","lastName":"Schell","suffix":""},{"id":526070849,"identity":"a49c1a98-f0ef-4036-a922-54062a61630a","order_by":4,"name":"Xiaoqing Yu","email":"","orcid":"https://orcid.org/0000-0003-4585-9372","institution":"Department of Biostatistics, Yale School of Public Health","correspondingAuthor":false,"prefix":"","firstName":"Xiaoqing","middleName":"","lastName":"Yu","suffix":""},{"id":526070850,"identity":"46007c7c-a244-4513-84b5-d93944b1733c","order_by":5,"name":"Gabriel Roman Souza","email":"","orcid":"","institution":"Moffitt Cancer Center","correspondingAuthor":false,"prefix":"","firstName":"Gabriel","middleName":"Roman","lastName":"Souza","suffix":""},{"id":526070851,"identity":"473cd15c-5782-4919-9b20-e99f205f7a63","order_by":6,"name":"Sarah Mizelle","email":"","orcid":"","institution":"Moffitt Cancer Center","correspondingAuthor":false,"prefix":"","firstName":"Sarah","middleName":"","lastName":"Mizelle","suffix":""},{"id":526070852,"identity":"64c5facb-909c-453d-a731-e03f338d73ec","order_by":7,"name":"Keerthi Gullapalli","email":"","orcid":"","institution":"Moffitt Cancer Center","correspondingAuthor":false,"prefix":"","firstName":"Keerthi","middleName":"","lastName":"Gullapalli","suffix":""},{"id":526070853,"identity":"7bdaad4d-01b6-4065-81a4-660bc85e6542","order_by":8,"name":"Adnan Fazili","email":"","orcid":"","institution":"Moffitt Cancer Center","correspondingAuthor":false,"prefix":"","firstName":"Adnan","middleName":"","lastName":"Fazili","suffix":""},{"id":526070854,"identity":"45260620-a3f6-406d-93cf-84ccad0c3bbe","order_by":9,"name":"Firas Hatoum","email":"","orcid":"","institution":"Moffitt Cancer Center","correspondingAuthor":false,"prefix":"","firstName":"Firas","middleName":"","lastName":"Hatoum","suffix":""},{"id":526070855,"identity":"24136999-ac08-479e-8045-5f5bd499a81b","order_by":10,"name":"Rohit Jain","email":"","orcid":"","institution":"Weill Cornell Medicine","correspondingAuthor":false,"prefix":"","firstName":"Rohit","middleName":"","lastName":"Jain","suffix":""},{"id":526070856,"identity":"645baf3f-b2f9-4bcc-a8b2-b9c709297a7d","order_by":11,"name":"Jonathan Chatzkel","email":"","orcid":"","institution":"Moffitt Cancer Center","correspondingAuthor":false,"prefix":"","firstName":"Jonathan","middleName":"","lastName":"Chatzkel","suffix":""},{"id":526070857,"identity":"3aeb7a6c-10e3-45d6-8110-9440ed3cf9dc","order_by":12,"name":"Ling Cen","email":"","orcid":"","institution":"Moffitt Cancer Center","correspondingAuthor":false,"prefix":"","firstName":"Ling","middleName":"","lastName":"Cen","suffix":""},{"id":526070858,"identity":"cec758a6-3384-4b82-9643-391b60ff0e79","order_by":13,"name":"Jasreman Dhillon","email":"","orcid":"https://orcid.org/0000-0001-6785-3487","institution":"Department of Pathology, Moffitt Cancer Center","correspondingAuthor":false,"prefix":"","firstName":"Jasreman","middleName":"","lastName":"Dhillon","suffix":""},{"id":526070859,"identity":"48ffaf6c-96c1-4f4a-8272-aa7cb49f3104","order_by":14,"name":"Christopher Cubitt","email":"","orcid":"https://orcid.org/0000-0001-6343-7451","institution":"Moffitt Cancer Center","correspondingAuthor":false,"prefix":"","firstName":"Christopher","middleName":"","lastName":"Cubitt","suffix":""},{"id":526070860,"identity":"8b2b832e-9c4f-438d-b124-2900efbb6c35","order_by":15,"name":"Jiqiang Yao","email":"","orcid":"","institution":"Moffitt Cancer Center","correspondingAuthor":false,"prefix":"","firstName":"Jiqiang","middleName":"","lastName":"Yao","suffix":""},{"id":526070861,"identity":"6f809dfe-d983-4b82-8b29-96a0b835ad09","order_by":16,"name":"Junmin Whiting","email":"","orcid":"","institution":"Moffitt Cancer Center","correspondingAuthor":false,"prefix":"","firstName":"Junmin","middleName":"","lastName":"Whiting","suffix":""},{"id":526070862,"identity":"6a9366b2-20c4-455d-9203-36b114cd9d7c","order_by":17,"name":"Jennifer Swank","email":"","orcid":"","institution":"Moffitt Cancer Center","correspondingAuthor":false,"prefix":"","firstName":"Jennifer","middleName":"","lastName":"Swank","suffix":""},{"id":526070863,"identity":"ab1a5139-02a8-46f9-b5fd-9693efe1645f","order_by":18,"name":"Ghazal Jameel","email":"","orcid":"","institution":"Moffitt Cancer Center","correspondingAuthor":false,"prefix":"","firstName":"Ghazal","middleName":"","lastName":"Jameel","suffix":""},{"id":526070864,"identity":"f30af8d9-5c5a-49c4-8974-aae5754a6de2","order_by":19,"name":"Jingsong Zhang","email":"","orcid":"https://orcid.org/0000-0003-1192-5116","institution":"H. Lee Moffitt Cancer Center and Research Institute","correspondingAuthor":false,"prefix":"","firstName":"Jingsong","middleName":"","lastName":"Zhang","suffix":""},{"id":526070865,"identity":"eb26e646-8c0e-4cff-becf-0e2298194f34","order_by":20,"name":"Xuefeng Wang","email":"","orcid":"https://orcid.org/0000-0001-5775-408X","institution":"H. Lee Moffitt Cancer Center \u0026 Research Institute","correspondingAuthor":false,"prefix":"","firstName":"Xuefeng","middleName":"","lastName":"Wang","suffix":""},{"id":526070866,"identity":"28d5210d-fa37-4284-a08b-c570dd88aa27","order_by":21,"name":"Philippe Spiess","email":"","orcid":"","institution":"Lee Moffitt Cancer Center and Research Institute","correspondingAuthor":false,"prefix":"","firstName":"Philippe","middleName":"","lastName":"Spiess","suffix":""},{"id":526070867,"identity":"d2eb7df6-57a5-413e-ab4e-b233eb5a63ef","order_by":22,"name":"Mayer Fishman","email":"","orcid":"","institution":"University of South Florida Morsani College of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Mayer","middleName":"","lastName":"Fishman","suffix":""}],"badges":[],"createdAt":"2025-09-24 01:20:31","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7698261/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7698261/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":93029417,"identity":"90a02c03-1e0a-4467-bc9e-cde00f4af4f7","added_by":"auto","created_at":"2025-10-08 09:55:50","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":241878,"visible":true,"origin":"","legend":"\u003cp\u003eStudy design and patient disposition. (A) CONSORT diagram showing patient enrollment and disposition through the study. (B) Schematic overview of correlative analyses, including RECIST imaging assessments, NanoString gene expression analysis, Olink proteomics, and flow cytometry for immune cell profiling. Illustration of combination cytokine and immune checkpoint therapy resulting in tumor cell lysis. (C) Trial schema showing treatment blocks and assessment schedule. Block 1 consisted of three cycles of 200 mg pembrolizumab every 3 weeks followed by RECIST 1.1 assessment. Patients with complete response (CR) or stable disease (SD) during block 1 proceeded to Block 2, which included pembrolizumab with high-dose IL-2 (600,000 IU/kg, up to 5 doses per cycle for 2 cycles) followed by maintenance pembrolizumab and RECIST assessment. Patients with response of SD or better (SD+; includes SD, partial response [PR], or CR) proceeded to block 3. Patients with progressive disease (PD) discontinued treatment. After completion of block 3, patients with SD+ continued to block 4 and those with PD discontinued treatment. During block 4, patients received up to 3 cycles of pembrolizumab maintenance until end of treatment and underwent imaging at treatment discontinuation. Red diamonds indicate timepoints for peripheral blood collection for correlative analyses (baseline [B1W1], block 2 week 2 [B2W2], block 2 week 4 [B2W4], and block 4 week 1 [B4W1]).\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-7698261/v1/5536351a7e211bed078d45d5.png"},{"id":93029416,"identity":"728dd55c-908d-40d5-a03b-be9728f74ebd","added_by":"auto","created_at":"2025-10-08 09:55:50","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":282779,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eClinical outcomes and treatment responses.\u003c/strong\u003e Kaplan-Meier survival curves showing overall survival (A), progression-free survival (B), and treatment-free survival (C) for all patients (N=26). Median survival times with 95% confidence intervals are indicated. Numbers at risk are shown below each plot. (D) Waterfall plot showing maximum percent change in target lesion measurements from baseline for each patient, colored by best overall response. (E) Swimmer plot depicting individual patient treatment duration and best overall response. Each bar represents one patient, with colors indicating response category: red (progressive disease, PD), blue (stable disease, SD), green (partial response, PR), and purple (complete response, CR). \u0026nbsp;(F) Spider plot demonstrating percent change in tumor size over time for each patient as determined by RECIST 1.1 assessment. Vertical line indicates end of treatment on trial. Horizontal dotted red line represents progression of disease while the dotted blue line represents partial response. (G) Distribution of best overall responses across the study cohort. (H) Representative CT images from a selected patient demonstrating durable complete response, with baseline scan (top) and follow-up imaging after \u0026gt;6 years post-treatment discontinuation (bottom).\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-7698261/v1/475b1abdcffa4e9cee5517f1.png"},{"id":93029419,"identity":"f6c229d5-0468-40b5-8661-6161764ee040","added_by":"auto","created_at":"2025-10-08 09:55:50","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":354042,"visible":true,"origin":"","legend":"\u003cp\u003eTranscriptomic and proteomic analysis reveals differential expression patterns associated with treatment response. (A) Heatmap showing time-independent differential gene expression between extreme responders (EXT), intermediate responders (INT), and non-responders (NON). (B) Volcano plot displaying statistical significance versus fold change for all genes comparing extreme responders to non-responders. Red points indicate genes with p-value \u0026lt; 0.05 and |log2[FC]| \u0026gt; 0.6. (C) Violin plots comparing expression levels of selected differentially expressed genes between response groups. (D) Heatmap showing time-independent differential protein expression between response groups. (E) Volcano plot for protein expression comparing extreme responders to non-responders. displaying statistical significance versus fold change for all genes comparing extreme responders to non-responders. Red points indicate genes with p-value \u0026lt; 0.05 and |log2[FC]| \u0026gt; 0.6. (F) Violin plots comparing protein levels of selected differentially expressed proteins between response groups. Red text indicates Benjamini-Hochberg false discovery rate q \u0026lt; 0.05. (G) Illustration of treatment response groupings. (H-J) Selected Gene Set Enrichment Analysis (GSEA) plots show running enrichment scores (top) and gene rank metrics (bottom). (H) NanoString EXT vs. NON protein serine/threonine kinase pathway enrichment (NES = -1.53, q = 0.02). (I) NanoString EXT vs. NON chemokine activity enrichment (NES = 1.49, q = 0.07). (J) Olink EXT vs. INT SARS-CoV-2 innate immunity evasion and cell-specific immune response enrichment (NES = 1.65, q = 0.03). NES: normalized enrichment score; FDR: false discovery rate; FC: fold change.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-7698261/v1/f32e6540d35382d61ba7bca4.png"},{"id":93029418,"identity":"859fabde-06e2-4517-a89c-f1da89c7d31e","added_by":"auto","created_at":"2025-10-08 09:55:50","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":215925,"visible":true,"origin":"","legend":"\u003cp\u003eDifferential gene expression analysis across multiple timepoints. (A) Volcano plot showing differentially expressed genes at block 2 week 4 (B2W4, post-10 doses of IL-2) versus block 2 week 2 (B2W2, post-4 doses of pembrolizumab), with significant genes (FDR \u0026lt; 0.05, |log2FC| \u0026gt; 0.6) highlighted. (B) Volcano plot of extreme versus non-responders at B2W2 with significant genes (p \u0026lt; 0.05, |log2FC| \u0026gt; 0.6) highlighted. (C) Volcano plot of extreme versus non-responders at B2W4 with significant genes (p \u0026lt; 0.05, |log2FC| \u0026gt; 0.6) highlighted. (D) Heatmap of gene expression at B2W4 illustrating patterns of key inflammatory mediators and chemokines across different experimental conditions. Blue indicates downregulation while red indicates upregulation of gene expression.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-7698261/v1/fd7570aa63d4acb935ddb9aa.png"},{"id":93030078,"identity":"9fd7d016-b50b-4312-ae06-9f52fa0ed384","added_by":"auto","created_at":"2025-10-08 10:03:50","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":277078,"visible":true,"origin":"","legend":"\u003cp\u003eFlow cytometry analysis of immune cell populations and PD-1 expression dynamics. (A) Violin plots showing CD16+ NK cell frequencies between response groups (EXT, extreme responders; INT, intermediate responders; NON, non-responders). (B) Violin plots showing CD4+ T-cell frequencies between response groups (C) Violin plots showing CD19+ B-cell frequencies between response groups. (D) Longitudinal PD-1 expression trajectories across immune cell subsets (Live+/PD-1+ Tc/PD-1+, Th/PD-1+) stratified by response category. Green lines represent extreme responders, blue lines represent intermediate responders, and red lines represent non-responders. Error bars indicate standard mean error. (E) Heatmap of baseline PD-1 expression in individual patients across immune cell subsets, with color intensity representing expression levels.\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-7698261/v1/4272811f212605b28d773e4e.png"},{"id":98625444,"identity":"efc06a9c-90d6-475d-b742-92420624ef15","added_by":"auto","created_at":"2025-12-19 17:09:07","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2484300,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7698261/v1/b4d08612-d539-4871-a44e-9d2fe8a69b48.pdf"},{"id":93030079,"identity":"d66f2b0b-a3d3-4532-b55b-ef58a0720534","added_by":"auto","created_at":"2025-10-08 10:03:50","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":421272,"visible":true,"origin":"","legend":"Gating Straregy","description":"","filename":"SupplementalFigure1.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7698261/v1/50662a22e88986a124d44b5e.pdf"},{"id":93029422,"identity":"58a83714-91eb-44f7-a0f5-b7cd54e66407","added_by":"auto","created_at":"2025-10-08 09:55:50","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":1048172,"visible":true,"origin":"","legend":"Trial Protocol","description":"","filename":"Protocol.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7698261/v1/a621bbc88513627deb83cf7d.pdf"}],"financialInterests":"\u003cb\u003eYes\u003c/b\u003e there is potential Competing Interest.\nI have no competing interests. All disclosures are on page 1 of the manuscript.","formattedTitle":"Pembrolizumab plus High-Dose IL-2 in Advanced Clear Cell Renal Cell Carcinoma: Six-Year Survival Outcomes and Molecular Signatures from a Phase 2 Trial","fulltext":[{"header":"1.\tIntroduction","content":"\u003cp\u003eIn 2025, kidney cancers, including renal cell carcinoma (RCC) and cancers of the renal pelvis, are estimated to affect approximately 80,980 individuals in the United States, with approximately 20-30% of patients presenting with metastatic disease at the time of diagnosis.\u003csup\u003e1,2\u003c/sup\u003e The therapeutic landscape for the first-line treatment of metastatic RCC (mRCC) has evolved dramatically, with contemporary standard-of-care regimens involving continuous, or prolonged, dosing of either dual immune checkpoint inhibitor (ICI) therapy or combination ICI plus tyrosine kinase inhibitor (TKI) therapy.\u003csup\u003e3\u003c/sup\u003e For example, the CheckMate 214 trial evaluated dual immune checkpoint inhibition with nivolumab (anti-PD-1) plus ipilimumab (anti-CTLA-4), and found improved overall survival compared to sunitinib across all risk groups of advanced renal cell carcinoma patients with 8 years of follow up.\u003csup\u003e4\u003c/sup\u003e ICI-TKI combinations, including pembrolizumab-axitinib, nivolumab-cabozantinib, and pembrolizumab-lenvatinib have also exhibited improved survival outcomes in the KEYNOTE-426, CheckMate-9ER, and CLEAR trials, respectively, with median follow-up intervals ranging from 33 to 67.2 months.\u003csup\u003e5–10\u003c/sup\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eDespite these advances, current therapeutic approaches require prolonged treatment administration. Dual checkpoint inhibition may entail up to two years of continuous dosing, while ICI-TKI combinations may continue indefinitely.\u003csup\u003e3\u003c/sup\u003e This treatment paradigm presents challenges for patients with mRCC, who face chronic symptoms including fatigue, pain, dyspnea, and insomnia that significantly impact quality of life.\u003csup\u003e11\u003c/sup\u003e Prolonged treatment exposure may compound these burdens by increasing the risk of chronic treatment toxicities, imposing additional financial and psychosocial stress, and escalating healthcare resource utilization, with contemporary first-line immunotherapy combinations incurring costs exceeding $260,000-$325,000 annually.\u003csup\u003e12–16\u003c/sup\u003e In response to these concerns, treatment-free survival (TFS), which quantifies the duration of disease control maintained without additional treatment, has emerged as a patient-centered metric for evaluating therapeutic regimens.\u003csup\u003e17\u003c/sup\u003e Contemporary efforts to characterize TFS in the CheckMate 214 trial revealed that 18% of patients treated with nivolumab plus ipilimumab (nivo+ipi) remained treatment-free at the 5-year timepoint.\u003csup\u003e18\u003c/sup\u003e Moreover, the nivo+ipi population had a 5-year mean overall survival of 40.7 months and mean TFS of 11.1 months.\u003csup\u003e18\u003c/sup\u003e The discrepancy between overall and treatment-free survival highlights the need for novel therapeutic strategies that can achieve durable objective responses while enabling sustained treatment-free intervals.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eInterleukin-2 (IL-2), approved by the United States Food and Drug Administration in 1992, was the first immunotherapy to demonstrate efficacy in metastatic RCC.\u003csup\u003e19\u003c/sup\u003e While high-dose IL-2 can produce durable complete responses in select patients, its widespread use has been limited by acute toxicity.\u003csup\u003e20\u003c/sup\u003e Mechanistically, ICI and cytokine therapies demonstrate complementary approaches to immune activation. Pembrolizumab, an anti-PD-1 ICI, acts to overcome immune tolerance by inhibiting PD-1/PD-L1 interaction, enhancing T-cell recognition of tumor antigens, and inducing intratumoral lymphocyte penetration; while IL-2 promotes proliferation, cytotoxic capacity, and survival of activated T- and NK-cells, potentially amplifying the effector response.\u003csup\u003e21–23\u003c/sup\u003e Recently, alternative approaches to harness IL-2-mediated immune activation have been explored. The phase 3 PIVOT-09 trial evaluated bempegaldesleukin (BEMPEG), a pegylated IL-2 prodrug in combination with nivolumab, an anti-PD-1 ICI, for treatment-naïve advanced ccRCC.\u003csup\u003e24\u003c/sup\u003e Despite its design to overcome limitations of conventional IL-2 therapy, BEMPEG plus nivolumab did not improve ORR or OS versus TKI monotherapy in patients with intermediate and poor-risk disease. In contrast to the simultaneous administration of a modified cytokine prodrug plus PD-1 inhibition, our approach employs sequential immunotherapy administration with high-dose IL-2.\u003c/p\u003e\n\u003cp\u003eHere, we report long-term survival outcomes from the first prospective trial evaluating short-course pembrolizumab plus high-dose IL-2 in treatment-naive advanced ccRCC. This time-limited approach addresses the critical unmet need for effective therapy that enables sustained treatment-free survival.\u003csup\u003e25\u003c/sup\u003e With over 6 years of follow-up, we assess the durability of survival outcomes and identify molecular signatures associated with sustained response.\u003c/p\u003e"},{"header":"2.\tResults","content":"\u003ch1\u003eParticipants\u003c/h1\u003e\n\u003cp\u003eBetween April 2017 and October 2018, 27 patients were enrolled in the study. One enrollment-eligible patient was later ineligible due to progression of disease prior to study day one, and did not receive any treatment on study, resulting in a treatment population of 26 patients (Fig. 1a). The median age at the time of enrollment was 60.5 years (range: 40-74). 22 patients (85%) were male, and 4 patients (15%) were female. International Metastatic RCC Database Consortium (IMDC) risk classification identified 6 patients (23%) with favorable risk, 19 (73%) with intermediate risk, and 1 (4%) with poor risk disease. Prior to study enrollment, 21 patients (81%) had undergone radical nephrectomy, 1 patient (4%) had undergone partial nephrectomy, and 4 patients (15%) had not undergone nephrectomy. Baseline demographic and clinicopathologic data are displayed in Table 1a.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 1a. Baseline Characteristics of Patients Treated with Pembrolizumab Plus IL-2 Stratified by Treatment-Free Survival\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"593\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eCharacteristic\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eTFS \u0026gt; 5 years (N=11)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eTFS 6 months – 5 years\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(N=10)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eTFS \u0026lt; 6 months\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(N=5)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eTotal\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(N=26)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eMedian Age, years (range)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e65 (38-74)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e58 (40-67)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e60 (48-66)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e61 (38-74)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eSex, n (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Male\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e10 (38)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e9 (35)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e3 (12)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e22 (85)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Female\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1 (4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1 (4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e2 (8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e4 (15)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eRace, n (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; White\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e11 (42)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e8 (31)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e5 (19)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e24 (92)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Asian\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0 (0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1 (4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0 (0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1 (4)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Unknown\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0 (0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1 (4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0 (0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1 (4)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eEthnicity, n (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Not Hispanic/Latino\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e11 (42)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e8 (31)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e4 (15)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e23 (89)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Hispanic/Latino\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0 (0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e2 (8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1 (4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e3 (12)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eIMDC Score, n (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Favorable\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e3 (12)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e2 (8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1 (4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e6 (23)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Intermediate\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e7 (27)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e8 (31)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e4 (15)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e19 (73)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Poor\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1 (4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0 (0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0 (0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1 (4)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ePrimary Tumor Grade, n (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Grade 2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e3 (12)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e2 (8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0 (0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e5 (19)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Grade 3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e7(27)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e5 (19)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e2 (8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e14 (54)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Grade 4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1 (4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e2 (8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1 (4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e4 (15)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eHistological Type, n (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Clear Cell RCC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e11 (42.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e10 (38.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e5 (19.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e26 (100.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eDistant Metastatic Site, n (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Lung/Pleura\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e6 (23)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e7 (27)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e4 (15)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e17 (65)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Lymph Node\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e14 (54)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Liver\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e2 (8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1 (12)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0 (0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e3 (11)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Bone\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e2 (8)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Other\u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e7 (27)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eIncludes brain, pancreas, chest wall, gluteal, paraspinal, soft tissue of hip, retroperitoneum.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 1b. Radiological Response and Efficacy Outcomes\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eEfficacy outcomes\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eRadiological Response Following Treatment, n %\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Complete response (CR)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e11 (42%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Partial response (PR)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e8 (31%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Stable disease (SD)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e5 (19%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Progressive disease (PD)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e2 (8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eOverall Response Rate (ORR), n (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; CR/PR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e19 (73%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eDisease Control Rate (DCR), n (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; CR/PR/SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e24 (92%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eMedian Survival Outcomes, n months (IQR)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Overall Survival\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Progression-free Survival\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Treatment-free Survival\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNot reached (54 - NR mo.)\u003c/p\u003e\n \u003cp\u003e19.3 mo. (8.4 - NR mo.)\u003c/p\u003e\n \u003cp\u003e23.8 mo. (12.9 - NR mo.)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e5-year Survival Outcomes, % (95% CI)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Overall Survival\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Progression-Free Survival rate\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e73% (52 - 86%)\u003c/p\u003e\n \u003cp\u003e42% (24 - 60%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Treatment-Free Survival\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e42% (24 - 60%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eIMDC Favorable Risk, n months (IQR)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e Median Overall Survival\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Median Progression-free Survival\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eNot reached\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eNot reached (12.2 - NR mo.)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Median Treatment-free Survival\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eNot reached (9.7 - NR mo.)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eIMDC Intermediate/Poor-Risk, n months (IQR)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Median Overall Survival\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eNot reached (26.7 - NR mo.)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Median Progression-free Survival\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Median Treatment-free Survival\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e17.3 mo. (7.9 – NR mo.)\u003c/p\u003e\n \u003cp\u003e23.7 mo. (13.9 - NR mo.)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cstrong\u003eEfficacy Outcomes\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePatients received a median of 12 pembrolizumab doses (Interquartile Range [IQR]: 10-12, range: 3-12), with 14/26 patients (54%) completing all twelve planned doses. Of the 40 planned IL-2 doses, patients received a median of 25.5 doses (IQR: 18-33, range: 0-38). After a median follow-up duration of 76.4 months (IQR: 73.4-80.4 months) at the time of data cutoff, 19/26 (73%) of patients with therapy-naive disease achieved an objective response to pembrolizumab plus IL-2 therapy as assessed by the investigator (95% confidence interval [CI]: 52%-88%). The study therefore met its primary endpoint with an ORR exceeding the pre-specified threshold of 45%. Best response as determined per RECIST 1.1 criteria was a complete response (CR) in 11 patients (42%), partial response (PR) in 8 patients (31%), stable disease (SD) in 5 patients (19%), and progressive disease (PD) in 2 patients (8%), resulting in a disease control rate (DCR) of 92% (95% CI: 75%-99%; Table 1b).\u003c/p\u003e\n\u003cp\u003eAt data cutoff, median overall survival (OS) was not reached (NR) (IQR: 53.7-NR months), median progression-free survival (PFS) was 19.3 months (IQR: 8.4-NR months), and median treatment-free survival (TFS), defined as time from completion of pembro IL-2 to next line of therapy, date of last follow up, or death, was 23.8 months (IQR: 12.9-NR months). OS rates at the 1-, 3-, and 5-year follow-up timepoints were 100% (95% CI: 100%-100%), 76.9% (95% CI: 56-89%), and 73% (95% CI: 52-86%), respectively. PFS rates at the 1-, 3-, and 5-year timepoints were 62% (95% CI: 40-77%), 42% (95% CI: 24-60%), and 42% (95% CI: 24-60%), respectively. TFS rates were 77% (95% CI: 56-89%) at 1 year, and 42% (95% CI: 24-60%) at both the 3- and 5-year timepoints. Time-to-event outcomes and individual patient responses are illustrated in Fig. 2.\u0026nbsp;\u003c/p\u003e\n\u003ch1\u003eSafety\u003c/h1\u003e\n\u003cp\u003eNo grade 5 safety events occurred throughout the duration of the study as assessed per Common Terminology Criteria for Adverse Events (CTCAE) v4.0 grading criteria. No new or unexpected adverse events (AEs) have occurred following the previous study report.\u003csup\u003e25\u003c/sup\u003e None of the 11 patients with durable disease control, without the need for additional systemic therapy beyond this trial, had any persistent grade 2 or higher AEs after a median follow-up duration of 76.4 months.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eBiomarker Outcomes\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eCharacterization of Response Categories Based on Treatment-Free Survival\u003c/p\u003e\n\u003cp\u003eIn this study, we retrospectively defined clinical response categories based on treatment-free survival duration to capture the durability of treatment benefit and identify patients achieving sustained disease control. TFS was calculated from the date of treatment discontinuation to the date of subsequent systemic therapy initiation, disease progression requiring intervention, or death from any cause. Upon analysis of TFS outcomes in the patient population (n=26), three distinct response categories emerged that exhibited clear separation in clinical outcomes. 11/26 patients (42%) demonstrated TFS \u0026gt;5 years (extreme responders), 10/26 patients (39%) had TFS between 6 months and 5 years (intermediate responders), and 5/26 patients (19%) had TFS \u0026lt;6 months (non-responders), suggesting primary non-response or rapidly progressive disease despite combination immunotherapy. To generate hypotheses regarding biological mechanisms underlying extreme response, we performed transcriptomic, proteomic, and flow cytometric analyses comparing patient subgroups defined by treatment-free survival duration.\u003c/p\u003e\n\u003cp\u003eIntegrated Time-Independent Transcriptomic and Proteomic Analysis Identifies Distinct Immune Signatures in Exceptional Responders\u003c/p\u003e\n\u003cp\u003eTo capture biological differences that include patients with early treatment discontinuation, we performed time-independent transcriptomic and proteomic analyses averaging expression across all available timepoints. 19/26 patients (73%) had evaluable data. Gene expression analysis of normalized NanoString data revealed 20/249 genes (8.0%) with statistically significant differential expression (p \u0026lt; 0.05) between extreme and non-responders (Fig. 3a). No genes in this comparison survived false discovery rate (FDR) correction. Extreme responders exhibited significantly greater expression of 19 genes, predominantly chemokines, inflammatory cytokines, and immune effector molecules. Of the significantly upregulated genes in extreme responders, the largest log\u003csub\u003e2\u003c/sub\u003e fold changes in expression were observed in IL-8 (log\u003csub\u003e2\u003c/sub\u003e[FC] = 1.44), IL-1B (log\u003csub\u003e2\u003c/sub\u003e[FC] = 0.94), TLR4 (log\u003csub\u003e2\u003c/sub\u003e[FC] = 0.93), MEF2C (log\u003csub\u003e2\u003c/sub\u003e[FC] = 0.92), and CXCL2 (log\u003csub\u003e2\u003c/sub\u003e[FC] = 0.86).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTime-independent proteomic analysis comparing extreme and non-responders identified 22/276 proteins (7.97%) with statistically significant differential expression (p \u0026lt; 0.05; Fig. 3d). No proteins in this comparison survived FDR correction. Among the significantly downregulated proteins in extreme responders, the largest log\u003csub\u003e2\u003c/sub\u003e fold changes were observed in MIC-A/B (log\u003csub\u003e2\u003c/sub\u003e[FC] = -1.04), MetAP2 (log\u003csub\u003e2\u003c/sub\u003e[FC] = -0.88), ADA (log\u003csub\u003e2\u003c/sub\u003e[FC] = -0.78), SOD1 (log\u003csub\u003e2\u003c/sub\u003e[FC] = -0.74), and GMZA (log\u003csub\u003e2\u003c/sub\u003e[FC] = -0.73).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eGene set enrichment analysis (GSEA) was conducted to identify pathway signatures distinguishing response groups (Fig. 3h-j). Transcriptomic GSEA revealed that extreme responders demonstrated significantly lower protein serine/threonine kinase activity compared to non-responders (q \u0026lt; 0.05, normalized enrichment score [NES] = -1.53), with trends toward enhanced chemokine receptor binding and chemokine activity (NES = 1.49, q = 0.07). When compared to non-responders, intermediate responders showed significantly reduced serine/threonine kinase activity (q \u0026lt; 0.05) and stress-responsive signaling pathways involving MAPK, NF-κB, and TGF-β cascades (q \u0026lt; 0.05).\u003c/p\u003e\n\u003cp\u003eProteomic GSEA showed that extreme responders demonstrated significantly greater innate immunity pathway activity compared to intermediate responders (NES = 1.65, q = 0.03) and trended toward enhanced chemokine signaling pathways (q \u0026lt; 0.1).\u003c/p\u003e\n\u003cp\u003eTime-Dependent Gene Expression Analysis Reveals Immune Modulation Associated with Treatment\u003c/p\u003e\n\u003cp\u003eWe performed time-dependent analysis of immune gene expression across treatment phases (baseline, pembrolizumab monotherapy, IL-2 addition, and post-IL-2; mean: 3.1 timepoints per patient) to capture dynamic patterns underlying sequential administration of pembrolizumab and IL-2.\u003c/p\u003e\n\u003cp\u003eIn a pooled analysis of all response groups, the administration of IL-2 produced the most pronounced transcriptional response, with six genes significantly altered (B2W4 vs. B2W2, n=15 paired samples; q \u0026lt; 0.05, |log\u003csub\u003e2\u003c/sub\u003e[FC]| \u0026gt; 0.6). Five genes were upregulated, including chemokine receptors and ligands (CCR3, CCL3), inflammatory mediators (IL-11), and immune effector molecules (CSF1, HMGB2), while CD40 was downregulated (Fig. 4a). In contrast to the robust IL-2 signature, analyses of pembrolizumab monotherapy (B2W2 vs. B1W1) and post-IL-2 phases (B4W1 vs. B2W4) showed no significant transcriptional changes.\u003c/p\u003e\n\u003cp\u003eCross-sectional transcriptomic analyses revealed distinct gene expression patterns between response groups at sequential correlative timepoints. Following pembrolizumab monotherapy (B2W2), both extreme and intermediate responders exhibited significant C3 upregulation compared to non-responders (2.2-fold, p = 0.019 and p = 0.003, respectively) (Fig. 4b). Extreme responders uniquely displayed enhanced CCL11 expression (2.1-fold, p = 0.012 versus non-responders). IL-2 administration (B2W4) generated marked transcriptional differences (Fig. 4c, 4d). Extreme responders demonstrated coordinated upregulation of pro-inflammatory chemokines compared to non-responders, including IL-8 (5.6-fold, p = 0.007), IL-1β (3.1-fold, p = 0.012), CXCL1/2/3 (2.3-2.8-fold, p \u0026lt; 0.03), and CCL7 (2.1-fold, p = 0.0004). Treatment responders exhibited significant CXCL9 upregulation (1.7-fold in extreme and 2.0-fold in intermediate responders, p \u0026lt; 0.04). Intermediate responders showed greater expression of IL-8 (9.8-fold, p = 0.006), FOS (4.5-fold, p = 0.026), CSF1 (1.6-fold, p = 0.01) and IL-1α (2.2-fold, p = 0.026) compared to extreme responders. Durable responses may depend not just on peak inflammation, but on sustained innate/adaptive immune engagement and complement activation.\u003c/p\u003e\n\u003cp\u003eTo identify differential treatment response pathways between groups, we performed difference-in-differences analyses comparing gene expression changes during sequential treatment phases. During the IL-2 treatment phase, extreme responders showed significantly enhanced upregulation of PKC isoforms (PRKCA: Δlog₂[FC] = 0.72, p = 0.005; PRKCB: Δlog₂[FC] = 0.40, p = 0.007), BCL6 (Δlog₂[FC] = 0.74, p = 0.014), MEF2A (Δlog₂[FC] = 0.44, p = 0.003), and TGFB1 (Δlog₂[FC] = 0.79, p = 0.016) compared to non-responders. Extreme responders also demonstrated enhanced TGFBR1 upregulation versus intermediate responders (Δlog₂[FC] = 0.56, p = 0.019). Following IL-2 discontinuation, extreme responders showed upregulation of inflammatory mediators (IL1A: Δlog₂[FC] = 1.56, p = 0.026; IL17A: Δlog₂[FC] = 1.29, p = 0.022) and complement factors (C5, C6, CFB: Δlog₂[FC] = 0.55-1.73, p \u0026lt; 0.03) compared to intermediate responders. Extreme responders exhibited both early and sustained transcriptional programs marked by innate immune activation, chemokine signaling, PKC and TGF-β pathway engagement, and complement activation, which differentiated them from intermediate responders. These findings suggest IL-2–induced immune remodeling as a critical determinant of therapeutic depth and durability. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003ePeripheral Blood Immune Profiling Identifies Patterns Associated with Response Groups\u003c/p\u003e\n\u003cp\u003eFlow cytometry analysis of peripheral blood immune cell populations was performed to identify immune states associated with durable treatment-free survival. Pooled analysis across all timepoints revealed that the frequency of CD16+ natural killer (NK) cells within the total circulating immune cell population was markedly elevated in treatment responders compared to non-responders, with intermediate responders demonstrating a 6.05-fold higher frequency (q = 1.77×10\u003csup\u003e-4\u003c/sup\u003e) and extreme responders showing a 4.20-fold higher frequency (q = 1.86×10\u003csup\u003e-2\u003c/sup\u003e) (Fig. 5a). Additionally, circulating CD4+ helper T-cell and CD19+ B-cell frequencies were elevated in extreme compared to intermediate responders (1.38-fold, q = 1.11×10\u003csup\u003e-4\u003c/sup\u003e; 1.93-fold, q = 4.85×10\u003csup\u003e-4\u003c/sup\u003e, respectively; Fig. 5b,c).\u003c/p\u003e\n\u003cp\u003eLongitudinal analysis revealed treatment-induced immune dynamics that varied by response group. Extreme responders demonstrated significant increases in CD15+ myeloid-derived suppressor cell (MDSC) populations from baseline to post-IL-2 (B1W1 to B2W4: 5.03-fold, p = 0.0391). Intermediate and non-responders exhibited no statistically significant temporal changes in any immune population analyzed.\u003c/p\u003e\n\u003cp\u003eAmong 25/26 patients (96.1%) with evaluable PD-1 expression data across the four collection timepoints, we observed distinct patterns that correlated with clinical outcomes. Longitudinal analysis revealed that patients achieving extreme response demonstrated suppression of PD-1+ cell frequencies across B1W1 through B4W1. In contrast, non-responders demonstrated more variable PD-1 expression levels throughout treatment (Fig. 5d). Furthermore, baseline PD-1+ expression levels demonstrated elevated median expression in non-responders compared to extreme and intermediate responders across T-cell populations, including live PD-1+ T-cells, FoxP3+ regulatory T-cells (Treg/PD-1+), helper T-cells (Th/PD-1+) and cytotoxic T-cells (Tc/PD-1+) (Fig. 5e). While no individual results achieved statistical significance, these patterns were consistently observed across T-cell subsets.\u003c/p\u003e"},{"header":"3. Discussion","content":"\u003cp\u003eDespite improved outcomes with the advent of immunotherapy, advanced renal cell carcinoma represents a significant therapeutic challenge. Current standard-of-care treatment approaches, including dual checkpoint inhibition or checkpoint inhibitor-tyrosine kinase inhibitor combinations, often require prolonged or indefinite therapy resulting in cumulative treatment-related morbidity, economic burden, and impact on quality of life. The implications of prolonged therapy are substantial, with contemporary studies demonstrating that immunotherapy combinations incur significant annual healthcare costs and increase emergency department visits and hospitalizations.\u003csup\u003e16,26–28\u003c/sup\u003e Moreover, a recent systematic review and meta-analysis identified patient-reported outcomes as an independent prognostic predictor of overall survival, highlighting the importance of treatment approaches that preserve quality of life.\u003csup\u003e29\u003c/sup\u003e These findings underscore the need for treatment strategies that can achieve durable disease control without requiring indefinite therapy administration. Consequently, treatment-free survival has emerged as a patient-centered endpoint that incorporates dual goals of durable disease control and freedom from ongoing treatment burden.\u003csup\u003e17,18,30,31\u003c/sup\u003e The present study evaluated survival outcomes following a novel short-course regimen of combination ICI-cytokine therapy, in the form of pembrolizumab plus IL-2, in patients with advanced ccRCC. This study highlights long-term follow up beyond a median of 6 years, allowing for longitudinal assessment of the durable response in extreme responders clinically.\u003c/p\u003e\n\u003cp\u003eOur results demonstrate that a finite course of pembrolizumab plus IL-2 can achieve exceptional treatment-free survival outcomes, with 42.3% (95% CI: 23.5-60.0%) of patients remaining free of both disease progression and subsequent therapy administration after over 6 years of follow-up. Notably, these patients experienced no persistent grade ≥2 toxicities, suggesting that durable disease control can be achieved without the chronic morbidity associated with continuous immunotherapy approaches.\u003csup\u003e32,33\u003c/sup\u003e After a median follow-up duration of 76.4 months, 11/26 (42.3%) patients achieved CR, 8/26 (30.8%) achieved PR, and 5/26 (19.2%) maintained SD, contributing to a DCR of 92.3%. The pembrolizumab plus IL-2 regimen demonstrated a complete response rate substantially higher than other contemporary combination immunotherapy regimens, which have reported CR rates ranging from 8-16%.\u003csup\u003e7,34,35\u003c/sup\u003e Moreover, TFS rates were durable over long-term observation, with 76.9% (95% CI: 55.7-88.9%) of patients remaining treatment-free at 1 year from therapy completion, and 42.3% (95% CI: 23.5-60.0%) maintaining treatment-free status at both the 3- and 5-year follow-up timepoints.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWhile direct cross-trial comparisons require caution due to differences in patient populations and study designs, our treatment-free survival outcomes compare favorably to contemporary standards, despite our study’s more limited sample size. The 5-year landmark analysis of CheckMate 214 demonstrated that, following two years of ipilimumab plus nivolumab treatment, 18% of patients maintained treatment-free survival at 5 years, with a mean allocation of 16.0 months on protocol therapy, 11.1 months treatment-free, and 13.7 months surviving after subsequent therapy initiation.\u003csup\u003e18\u003c/sup\u003e The 43-month follow-up analysis of the KEYNOTE-426 trial of pembrolizumab plus axitinib versus sunitinib reported a median of 20 pembrolizumab administrations and 15 (range: 0.03–48) months of axitinib.\u003csup\u003e36\u003c/sup\u003e While TFS was not a systematically reported endpoint from the trial, 204/432 patients (47.2%) in the combination arm received subsequent therapy at any time following study treatment discontinuation.\u003csup\u003e36\u003c/sup\u003e In contrast, the pembrolizumab plus IL-2 regimen required substantially less treatment exposure with longer treatment-free intervals. Patients received a median of 12 pembrolizumab doses and 25.5 IL-2 doses in this study. Following discontinuation of treatment, the median TFS time was 23.8 months with 42.3% of patients remaining treatment-free after 5 years.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eMoreover, compared to previous trials of high-dose IL-2 monotherapy, our trial demonstrates superior response rates. A study of 155 mRCC patients treated with high dose IL-2 monotherapy found a CR rate of 7% and ORR of 21%, while a contemporary retrospective analysis of 810 mRCC patients from the PROCLAIM registry, most of whom were treated with the 14-doses-in-a-row schedule, found a CR rate of 5% and ORR of 24.7%.\u003csup\u003e37,38\u003c/sup\u003e In contrast, our pembrolizumab plus IL-2 combination regimen achieved a CR rate of 42.3% and ORR of 73.1%, representing approximately 5-fold and 3-fold improvements, respectively. Nevertheless, these encouraging signals must be interpreted within the context of our single-arm design and limited sample size. Prospective randomized comparison would be required to definitively establish superiority over alternative treatment approaches.\u003c/p\u003e\n\u003cp\u003eOur findings of durable response with pembrolizumab plus high-dose IL-2 are particularly notable when contextualized with data from the phase 3 PIVOT-09 trial of BEMPEG plus nivolumab (BEMPEG+nivo) versus TKI in advanced ccRCC.\u003csup\u003e24\u003c/sup\u003e Despite both approaches targeting the IL-2 pathway with PD-1 inhibition, BEMPEG plus nivolumab achieved a 23.0% ORR and 2.0% complete response rate in intermediate/poor-risk patients, while our sequential approach delivered substantially improved response and survival outcomes across IMDC risk categories.\u003csup\u003e24\u003c/sup\u003e These divergent outcomes likely result from differences in immune receptor engagement and treatment administration strategy. While BEMPEG demonstrates selective CD122/CD132 binding to preferentially activate CD8+ T-cells and NK-cells over Tregs, aldesleukin engages the trimeric IL-2 receptor (CD25/CD122/CD132).\u003csup\u003e39,40\u003c/sup\u003e This broader receptor engagement may provide more potent stimulation to tumor-reactive CD8+ T cells that transiently upregulate CD25 during activation, supporting their optimal expansion, function, and memory formation.\u003csup\u003e40–43\u003c/sup\u003e Interestingly, Eisner and colleagues demonstrated that responders to IL-2 therapy paradoxically exhibited higher baseline expression of immunosuppressive markers (such as MDSCs, Tregs, and neutrophils) that typically predict poor response to PD-1 inhibition.\u003csup\u003e44\u003c/sup\u003e Our sequential approach with pembrolizumab priming followed by high-dose IL-2 may more effectively leverage the distinct and complementary mechanisms of ICI and cytokine therapies than the simultaneous administration of BEMPEG+nivo in PIVOT-09.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eOur correlative analyses, enabled by over six years of follow-up on this phase 2 trial, provide hypothesis-generating insights into mechanisms underlying durable treatment-free survival through characterization of extreme responders and their distinct biomarker profiles. Although only a portion of individual biomarkers survived multiple testing correction, convergent evidence from flow cytometry, transcriptomic, and proteomic platforms suggests the presence of biologically coherent patterns warranting further validation. Longitudinal flow cytometry revealed that extreme responders demonstrated suppression of PD-1+ cell frequencies in peripheral blood across T-cell subsets from baseline through completion of treatment, while non-responders demonstrated variable PD-1 population frequencies. This PD-1 suppression may suggest successful immune reconditioning, with chronically exhausted T-cells transitioning toward functional competence. Notably, both extreme and intermediate responders demonstrated marked enrichment of circulating CD16+ NK-cells which may serve as effector mechanisms for sustained antitumor immunity. Recent single-cell analyses in ccRCC have demonstrated that advanced disease progression is characterized by depletion of functional, cytotoxic NK-cells and enrichment of dysfunctional, tissue-resident NK-cell populations with impaired degranulation capacity.\u003csup\u003e45\u003c/sup\u003e We highlight that the preservation or enhancement of circulating CD16+ NK-cells in extreme responders may suggest maintenance of cytotoxic immune surveillance capabilities that could contribute to durable disease control. While the expansion of CD15+ MDSCs may traditionally be associated with immune suppression, their increase exclusively in extreme responders during IL-2 therapy may reflect transient neutrophilic activation or immune homeostasis, rather than functional suppression. Our findings support a mechanistic model of controlled immune reconditioning, whereby PD-1 blockade facilitates immune reinvigoration without overstimulation or clearance of exhausted lymphocyte populations, allowing for IL-2–mediated immune amplification in a biologically contained and clinically tolerable fashion.\u003csup\u003e46\u003c/sup\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThese immune cell population changes aligned with molecular signatures in extreme responders. Temporal analysis revealed distinct gene expression signatures that differentiated response groups, with extreme responders demonstrating coordinated upregulation of pro-inflammatory chemokines following IL-2 administration, suggesting enhanced T-cell trafficking. Moreover, time-independent comparisons of response groups identified elevated immune recruitment chemokines (such as IL-8 and CXCL1/2) combined with reduced cellular stress markers (such as MIC-A/B and GMZA) in extreme responders, potentially representing optimal immune homeostasis for sustained antitumor surveillance.\u003csup\u003e47\u003c/sup\u003e A study of ipilimumab in hematologic cancers similarly demonstrated chemokine upregulation (CXCL2, IL-8) and reduced soluble MIC-A in treatment responders, supporting conserved mechanisms of immune cell recruitment and cellular stress modulation.\u003csup\u003e48\u003c/sup\u003e Furthermore, a pharmacodynamic study of PD-1 blockade demonstrated that nivolumab induces CXCL9 and CXCL10 production in mRCC tumors.\u003csup\u003e23\u003c/sup\u003e The sequential treatment strategy of this study may play a role in achieving immune homeostasis, with pembrolizumab initially rescuing exhausted T-cells from PD-1-mediated suppression, creating permissive conditions for subsequent IL-2-mediated activation without overwhelming the overall system. This controlled immune reconditioning could explain both the durable treatment-free survival and the favorable toxicity profile compared to high-dose IL-2 monotherapy or prolonged checkpoint inhibition. This immune remodeling was supported by multi-omic signals including innate immune priming via CD16⁺ NK-cells, sustained chemokine and cytokine induction, and enhanced expression of PKC and TGF-β pathway genes. The reduction in cellular stress ligands such as MIC-A/B and downregulation of PD-1 expression further suggest an immunologically poised but restrained state, enabling effective tumor control without immune toxicity. While further validation in independent cohorts is needed, these findings suggest a biological framework for finite combination immunotherapy through immune-checkpoint inhibition and cytokine chemotaxis and identify potential biomarkers for patient selection and treatment optimization. Recent work has highlighted the potential of IL2RG-related gene signatures as prognostic and predictive biomarkers in ccRCC, which could be further explored in the context of pembrolizumab plus IL-2 therapy to identify patients most likely to achieve durable response.\u003csup\u003e49\u003c/sup\u003e\u003c/p\u003e\n\u003cp\u003eSeveral important limitations must be acknowledged. The non-randomized, single-arm design of this study prevents definitive attribution of outcomes to the combination regimen versus individual components or patient selection factors. Our study population of 26 patients is appropriate for initial statistical exploration of biomarkers, rather than generalizability to broader populations. While we had a limited number of patients with poor-risk disease, the distribution between favorable and intermediate risk groups is reflective of the real-world IMDC risk distribution observed in clinical practice for RCC patients.\u003csup\u003e50\u003c/sup\u003e Our single-center experience may not reflect outcomes achievable in broader community practice, particularly given the specialized requirements for IL-2 administration.\u003csup\u003e51\u003c/sup\u003e However, for appropriately selected patients, the potential for durable treatment-free survival may justify additional upfront resource intensity, particularly when weighed against the cumulative burden and costs of indefinite therapy. Future studies could explore less toxic IL-2 receptor-targeted cytokines such as IL-2 fusion proteins or monoclonal antibody complexes that may maintain beneficial immunologic effects while reducing specialized administration requirements.\u003csup\u003e52\u003c/sup\u003e\u003c/p\u003e\n\u003cp\u003eIn summary, our findings demonstrate that time-limited pembrolizumab plus IL-2 therapy is not only feasible and tolerable, but also capable of delivering durable clinical benefit, with high objective and complete response rates, prolonged overall survival, and extended treatment-free survival. These results establish a clinical and mechanistic rationale for time-limited immunotherapy that is capable of inducing sustained tumor control. Future studies should validate biomarker signatures in independent cohorts, develop immune profiling–based patient stratification tools, and incorporate randomized comparisons with standard-of-care regimens. Importantly, integration of patient-reported outcomes and health economic analyses will be essential to fully characterize the clinical and societal value of finite immunotherapy strategies with long-term disease control.\u003c/p\u003e"},{"header":"4.\tMethods","content":"\u003cp\u003e\u003cstrong\u003ePatients\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eEligible patients were aged 18 years or older with a diagnosis of metastatic renal cell carcinoma containing a clear cell component and with measurable disease per Response Evaluation Criteria in Solid Tumors (RECIST; version 1.1). Patients were required to have an Eastern Cooperative Oncology Group (ECOG) performance status of 0 or 1 and adequate organ and marrow function. Additionally, patients were required to undergo exercise or thallium stress testing (unless deemed exempt after evaluation by a cardiologist) and pulmonary function testing within 3 months of treatment initiation. Patients were excluded from participation in the study if they had active brain metastases, active autoimmune disease requiring systemic treatment within the past 2 years, received more than 1 systemic therapy in the past year, IL-2 therapy within 1 year, or anti-PD-1/PD-L1 agents at any time. Comprehensive inclusion and exclusion criteria are detailed in the study protocol in Supplementary Information. All patients provided written informed consent before any screening or study procedures took place.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTrial Design and Treatments\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis was a single-arm, single-institution phase 2 trial (ClinicalTrials.gov identifier: NCT02964078) investigating pembrolizumab and high dose IL-2 for the treatment of advanced clear cell renal cell carcinoma. The trial was conducted at the H. Lee Moffitt Cancer Center (Tampa, FL) with the approval of the Moffitt Cancer Center institutional review board (protocol: MCC-18536).\u003c/p\u003e\n\u003cp\u003eThe study architecture involved a maximum of four treatment blocks, each comprised of three 200 mg doses of pembrolizumab administered intravenously at 21 (\u0026plusmn;3) -day intervals. During the first and fourth blocks, patients were treated with pembrolizumab monotherapy. During the second and third blocks, two cycles of 5-in-a row high-dose (HD) IL-2 were administered between consecutive pembrolizumab doses. Each cycle of HD IL-2 consisted of up to five 600,000 IU/kg doses of IL-2 given every 8 hours for a total of 20 doses planned for block 2 and block 3, respectively. The protocol did not permit replacement of missed treatment doses. Computed tomography (CT) radiographic assessment of disease was obtained at baseline prior to treatment, and subsequent assessments were obtained at 9-week intervals while on treatment and at 2\u0026ndash;3-month intervals for at least two years or until progression of disease after discontinuing treatment. Tumor-response assessments were performed at each imaging timepoint according to Response Evaluation Criteria in Solid Tumors (RECIST) 1.1 criteria. Patients achieving a complete response (CR) at the first RECIST assessment timepoint were eligible to continue with pembrolizumab monotherapy. Patients with stable or progressive disease at this timepoint proceeded to treatment block 2. After completing the first treatment block, patients were permitted to proceed to subsequent treatment blocks in the absence of disease progression or intolerable toxicity.\u003c/p\u003e\n\u003cp\u003ePeripheral blood samples for correlative analyses were collected at baseline (B1W1), after four cycles of pembrolizumab (B2W2), after 10 doses of IL-2 (B2W4), and after IL-2 was completed (B4W1).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTrial Oversight\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA Protocol Monitoring Committee (PMC) was responsible for monitoring the safety data from this study as per the standard Moffitt Cancer Center Data Safety Monitoring Plan for investigator-initiated studies. The PMC monitors adverse event reporting, data and safety monitoring, and internal audit findings. Upon review, the PMC may approve a study for continuation, recommend revisions, or suspend or close the protocol. Pembrolizumab was provided by Merck Sharp \u0026amp; Dohme Corp., a subsidiary of Merck \u0026amp; Co., Inc., Kenilworth, NJ, USA. The conduct of this investigation abided by all applicable local regulatory requirements and was performed in accordance with the Good Clinical Practice (GCP) Guidelines of the International Council for Harmonization of Technical Requirements for Pharmaceuticals for Human Use (ICH) and the Declaration of Helsinki. All authors confirm that the trial was conducted with respect to the standards of Good Clinical Practice. All authors had access to the data, participated in the writing and reviewing of this manuscript, approved the submitted manuscript, and confirmed the accuracy and completeness of the data.\u003c/p\u003e\n\u003cp\u003eStatistical Analyses\u003c/p\u003e\n\u003cp\u003eFollowing a Simon 2-stage minimax design, the study evaluated the first 15 patients for objective response, with early termination specified if \u0026le; 3 responses were observed (stage 1 futility boundary). Upon passing this threshold, accrual continued to a total of 24 evaluable patients. The null hypothesis (objective response rate \u0026le; 20%) would be rejected in favor of the alternative hypothesis (response rate \u0026ge; 45%) if \u0026ge; 8 cumulative responses were observed across both stages (alpha=0.10, power=0.90).\u003c/p\u003e\n\u003cp\u003eDose-limiting toxicities were monitored using a sequential boundary method that accommodated variable follow-up durations as enrollment progressed.\u003csup\u003e53\u003c/sup\u003e The design implemented a Pocock-type stopping boundary with maximum 5% probability of boundary crossing when the true dose-limiting toxicity rate was at the pre-specified acceptable threshold of 15%. All patients who received at least one dose of pembrolizumab were included in the safety analysis.\u003c/p\u003e\n\u003cp\u003eOutcomes and Assessments\u003c/p\u003e\n\u003cp\u003eThe primary objective of this study was twofold: to evaluate anti-tumor activity, targeting an objective response rate (ORR) \u0026ge;45% in the ever-treated population, and to assess treatment safety using a sequential toxicity boundary method. Secondary objectives were survival endpoints (assessed using Kaplan-Meier analysis) including overall survival (OS), measured from enrollment to death from any cause; progression-free survival (PFS), defined as time from enrollment to first documented disease progression per RECIST 1.1 criteria or death from any cause; and treatment-free survival (TFS), calculated from completion of study treatment to initiation of next therapy, death, or last follow-up (in patients not receiving subsequent therapy). Exploratory endpoints included correlative biomarker analyses.\u003c/p\u003e\n\u003cp\u003eAll tumor-response assessments were performed in accordance with RECIST 1.1 criteria. Complete response was defined as disappearance of all target and non-target lesions. Partial response was defined as a 30% or more decrease in the sum of diameters of target lesions, taking as a reference the baseline sum diameters. Progressive disease was defined as a \u0026ge;20% increase in the sum of diameters of target lesions and unequivocal progression of existing non-target lesions or appearance of a new lesion. Stable disease was defined as insufficient increase or decrease to qualify as complete or partial response or disease progression.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCorrelative Analyses\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFlow cytometry staining and analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePeripheral blood was collected from clinical trial participants in up to three 10 mL sodium heparin vacutainers. The blood was then transferred and centrifuged for 5 minutes at 500xg to collect plasma. This plasma was stored at -80\u0026deg;C. Peripheral blood mononuclear cells (PBMCs) were then isolated from the cell pellet using Ficoll-Paque density gradient centrifugation, washed twice in phosphate-buffered saline (PBS) with 2% fetal bovine serum (FBS), and then cryopreserved using Recovery Freeze Media (Thermo Fisher Scientific, Waltham, Massachusetts), overnight controlled rate freezing and then LN2 storage. For flow batch runs, cells were rapidly thawed at 37\u0026deg;C, washed twice with Fc blocking buffer (5% Human Serum inactivated) and resuspended at 1 \u0026times; 10\u003csup\u003e6\u003c/sup\u003e cells/mL in staining buffer. A multicolor flow cytometry panel was designed to assess immune cell populations, using titrated antibodies against CD19 (SJ25C1, BUV496), HLA-DR (G46-6, BV650), CD14 (M5E2, BUV805), CD15 (W6D3, BV711), CD45 (HI30, BV786), CD11b (D12, BB700), CD33 (WM53, PE), CD279/PD-1 (J43, BUV737), CD274/PD-L1 (MIH1, PE-Cy7), CD56/NCAM (NCAM16.2, BV605), CD16 (B73.1, APC), CD3 (UCHT1, BUV395), CD8 (HIT8\u0026alpha;, R718), CD4 (SK3, BV480), CD127/IL-7R\u0026alpha; (HIL-7R-M21, BV421), CD25/IL-2R\u0026alpha; (M-A251, BB515), FOXP3 (259D/C7, PE-CF594), and Fixable Viability Stain 780 (all from BD Biosciences, Franklin Lakes, New Jersey). Antibody concentrations were optimized via serial dilutions on PBMCs from normal donor leukopaks (purchased from AllCells, Huntsville, Alabama) to maximize signal-to-noise ratio. Cells were stained with antibody cocktails for 20 minutes at 4\u0026deg;C, with FOXP3 staining performed post-fixation/permeabilization using the BD Cytofix/Cytoperm kit, and viability staining applied prior to fixation.\u003c/p\u003e\n\u003cp\u003eSingle-color and fluorescence-minus-one (FMO) controls, prepared using leukopak-derived PBMCs, were used to establish compensation matrices and gating boundaries, respectively, with rainbow beads (Spherotech, Lake Forest, Illinois) ensuring instrument standardization. Samples were acquired on a BD FACSSymphony A5 (LSRFortessa) with BD FACSDiva Software (Version 8.0.1), collecting at least 50,000 events per sample. Data were analyzed in FCS Express 7 (De Novo Software, Pasadena, California), using FMO-guided hierarchical gating to identify T cells (CD3+), B cells (CD19+), monocytes (CD14+), neutrophils (CD15+), and NK cells (CD56+CD16+), with dead cells excluded via viability staining. Expression of activation (HLA-DR, PD-1, PD-L1) and regulatory markers (CD25, FOXP3) was quantified as mean fluorescence intensity or percentage of positive cells. See Extended Data Fig.\u0026nbsp;1\u0026nbsp;for flow cytometry sequential gating/sorting strategies.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTranscriptomic and Proteomic Assays\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTranscriptomic (NanoString nCounter\u0026reg; PanCancer Immune Profiling Panel for Gene Expression) and proteomic (Olink\u0026reg; Target 96 Cardiometabolic Analysis Service USD, Target 96 Immuno-Oncology Analysis Service USD and Target 96 Oncology II Analysis Service) assays were performed. NanoString gene expression data were analyzed using nSolver 4.0 software, with normalization performed against six housekeeping genes (PGK1, CLTC, HPRT1, GUSB, GAPDH, TUBB). \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eClassification of Response Groups by TFS\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo classify patient response, we established discrete categories based on duration of treatment-free survival. Upon examination of the TFS distribution in our cohort, three distinct clusters of patients emerged with natural separation points at approximately 6 months and 5 years. Based on these observed groupings, we classified patients as extreme responders (TFS \u0026gt;5 years), intermediate responders (TFS between 6 months and 5 years), and non-responders (TFS \u0026lt;6 months). This classification approach was determined through inspection of the TFS distribution, which revealed clear multi-modal separation. These groupings were established prior to molecular analyses to avoid bias in the interpretation of gene expression patterns.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTime Independent Molecular Analyses\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDifferential gene and protein expression analysis was performed using the Mann-Whitney U test for non-parametric comparisons or Welch\u0026apos;s t-test when statistical assumptions were met (normality via Shapiro-Wilk test, equal variance via Levene\u0026apos;s test, n \u0026ge; 15).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eGene Set Enrichment Analysis (GSEA)\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eComprehensive pathway enrichment analysis was performed using the GSEApy implementation of Gene Set Enrichment Analysis.\u003csup\u003e54,55\u003c/sup\u003e For each platform and comparison, ranked gene/protein lists were generated based on log2 fold change values. This ranking approach maintains the directionality of expression changes while being compatible with the non-parametric differential expression analysis.\u003c/p\u003e\n\u003cp\u003ePre-ranked GSEA was conducted using GSEApy version 1.1.8 with the following parameters:\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003eGene ranking: Log2 fold change of difference in group means\u003c/li\u003e\n \u003cli\u003ePermutation number: 5,000\u003c/li\u003e\n \u003cli\u003eMinimum gene set size: 15\u003c/li\u003e\n \u003cli\u003eMaximum gene set size: 500\u003c/li\u003e\n \u003cli\u003eWeighted score type: 1\u003c/li\u003e\n \u003cli\u003eRandom seed: 42 for reproducibility\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eGene set databases interrogated included Gene Ontology Biological Process 2021, Gene Ontology Molecular Function 2021, KEGG 2021 Human, WikiPathway 2021 Human, and Reactome 2022. Pathway significance was assessed using nominal p-values and tiered FDR q-values, with pathways considered significantly enriched at FDR \u0026lt; 0.05, nominally enriched at FDR \u0026lt; 0.10, and exploratory enrichment at FDR \u0026lt; 0.25.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTime-Dependent Molecular Analyses\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo characterize dynamic patterns of immune gene expression across treatment phases, we performed comprehensive time-dependent analyses using multiple complementary approaches. Control genes (e.g., NEG_, POS_) and housekeeping genes (PGK1, CLTC, HPRT1, GUSB, GAPDH, TUBB) were systematically excluded from analyses.\u003c/p\u003e\n\u003cp\u003eData were pooled across all patients regardless of response category to identify general treatment effects. For each treatment phase (pembrolizumab monotherapy: B2W2 vs B1W1; IL-2 addition: B2W4 vs B2W2; post-IL-2: B4W1 vs B2W4), paired t-tests were conducted for genes with sufficient paired expression data (\u0026ge;3 paired samples per timepoint comparison). Multiple testing correction was applied using the Benjamini-Hochberg false discovery rate (FDR) method. Genes were considered significantly altered if they met both statistical (FDR \u0026lt; 0.05) and biological (|log₂FC| \u0026gt; 0.6) significance thresholds. This pooled approach allowed identification of consistent transcriptional changes induced by each treatment phase across the entire cohort.\u003c/p\u003e\n\u003cp\u003eWe performed cross-sectional transcriptomic analyses to reveal distinct gene expression patterns between response groups at specific timepoints. At each timepoint (B1W1, B2W2, B2W4, B4W1), expression levels were compared between extreme, intermediate, and non-responders using either Mann-Whitney U tests (non-parametric) or Welch\u0026apos;s t-tests based on statistical assumption testing (normality via Shapiro-Wilk test, equal variance via Levene\u0026apos;s test). A minimum of 3 samples per group was required for analysis. Following pembrolizumab monotherapy (B2W2), comparisons identified early molecular signatures associated with treatment outcomes. Similarly, after IL-2 administration (B2W4), analyses determined whether response groups exhibited different gene expression profiles potentially associated with IL-2 responsiveness. Expression differences were considered significant at p \u0026lt; 0.05, with FDR correction applied for multiple testing across genes within each comparison.\u003c/p\u003e\n\u003cp\u003eTo investigate whether temporal expression patterns differed between response groups, we performed difference-in-differences analyses comparing the magnitude of expression change between response groups. For each treatment transition (pembrolizumab effect: B2W2 vs B1W1; IL-2 effects: B2W4 vs B2W2; post-IL-2 effect: B2W4 vs B4W1; overall effect: B4W1 vs B1W1), we calculated individual patient treatment effects as the change in gene expression between timepoints. Treatment effects were then compared between response group pairs (extreme vs non-responders, intermediate vs non-responders, extreme vs intermediate responders) using Welch\u0026apos;s t-tests. This approach identified genes where the treatment response magnitude differed significantly between response categories, requiring a minimum of 2 patients per group with paired timepoint data.\u003c/p\u003e\n\u003cp\u003eMissing data were handled using available case analysis, including all patients with data at each timepoint rather than restricting to complete cases. For patients with multiple samples at the same timepoint, duplicate measurements were aggregated using median values. Longitudinal analyses only included patients with paired data at both relevant timepoints for each comparison.\u003c/p\u003e\n\u003cp\u003eFor all time-dependent analyses, statistical significance was assessed at \u0026alpha; = 0.05. Effect sizes were calculated as Cohen\u0026apos;s d for t-tests and fold change metrics for biological interpretation. Individual patient trajectories were tracked across timepoints with group means and standard errors calculated. Results were stratified into significance tiers: highly significant (FDR \u0026lt; 0.05), nominally significant (FDR \u0026lt; 0.10), and exploratory (FDR \u0026lt; 0.25) given the nature of this biomarker discovery study.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFlow Cytometry Analyses\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNine immune markers (CD4, CD8, CD19, CD16, CD25, CD15, CD33, CD14, CD11b) were analyzed using four measurement types: cell frequency as percentage of parent population, percentage of total cells, median fluorescence intensity (MFI), and geometric mean fluorescence. Log₂ transformation was applied to fluorescence measurements to normalize distributions. Between-group comparisons used the Mann-Whitney U test (for n \u0026lt; 15 or non-normal distributions) or Welch\u0026apos;s t-test following statistical assumption testing (Shapiro-Wilk, Levene\u0026apos;s tests). A minimum of 5 samples per group was required. Multiple testing correction used the Benjamini-Hochberg FDR for between-group comparisons and Bonferroni correction for pairwise temporal comparisons. Effect sizes included fold change, Cohen\u0026apos;s d, Glass\u0026apos;s delta, and Cliff\u0026apos;s delta.\u003c/p\u003e\n\u003cp\u003eFor baseline comparisons of PD-1 across treatment response groups, Kruskal-Wallis tests assessed overall group differences, followed by pairwise Mann-Whitney U tests when significant (p \u0026lt; 0.1 exploratory threshold). For longitudinal analysis of subjects with complete four-timepoint data (B1W1, B2W2, B2W4, B4W1), Friedman\u0026apos;s non-parametric repeated measures ANOVA tested overall temporal changes. Duplicate measurements were aggregated using median values. Pairwise temporal comparisons used Wilcoxon signed-rank tests for all timepoint combinations (B1W1 vs. B2W2, B2W2 vs. B2W4, B2W4 vs. B4W1, B2W2 vs. B4W1, and B4W1 vs. B1W1). Effect sizes were calculated as r = |Z|/\u0026radic;N, where Z is the standardized test statistic.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFor all flow cytometry analyses, statistical significance was assessed at \u0026alpha; = 0.05. Individual patient trajectories were tracked across timepoints with group means and standard errors calculated. Effect sizes and confidence intervals accompanied p-values for clinical interpretation.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSoftware Specifications, Data Processing, and Quality Control\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eStatistical analyses and figure generation were performed with R version 4.0.2 and Python version 3.11.5 using the pandas (v2.1.4), numpy (v1.24.3), scipy.stats (v1.15.2), statsmodels (v0.14.0), matplotlib (v3.7.2), seaborn (v0.13.2), scikit-learn (v1.3.0), and GSEApy (v1.1.8) modules. Platform-specific quality control included: (1) assessment of data completeness and missing value patterns, (2) parameter availability across timepoints, (3) sample size distribution validation, and (4) control gene identification and exclusion. The code used for data analysis is available at https://github.com/MCC18536/Pem-IL2/.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eData Availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAggregate patient-related information has been made available on ClinicalTrials.gov (https://clinicaltrials.gov/ct2/show/results/NCT02964078). Additional requests for raw and analyzed data can be referred to the corresponding author (J.C.) and will be reviewed promptly as part of a data transfer agreement per H. Lee Moffitt Cancer Center institutional policies, to determine whether the request is subject to any clinical trial patient confidentiality or intellectual property requirements.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFinancial support by contract with Clinigen, Inc. Pembrolizumab furnished by Merck Sharp \u0026amp; Dohme Corp., a subsidiary of Merck \u0026amp; Co., Inc., Kenilworth, NJ, USA. IND and regulatory support through Moffitt Cancer Center. \u0026nbsp;\u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRole of the Funding Source\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe funder of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eReporting Summary\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFurther information on research design is available in the Nature Portfolio Reporting Summary linked to this article.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe thank the patients and their families, investigators, coinvestigators, and the study team for their support and contributions to this research. We thank Michael Holman (Prometheus) for his advocacy for the project at its early stages.\u0026nbsp;\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eSiegel, R. L., Kratzer, T. B., Giaquinto, A. N., Sung, H. \u0026amp; Jemal, A. Cancer statistics, 2025. \u003cem\u003eCA Cancer J Clin\u003c/em\u003e (2025) doi:10.3322/caac.21871.\u003c/li\u003e\n\u003cli\u003eZnaor, A., Lortet-Tieulent, J., Laversanne, M., Jemal, A. \u0026amp; Bray, F. International variations and trends in renal cell carcinoma incidence and mortality. \u003cem\u003eEur Urol\u003c/em\u003e \u003cstrong\u003e67\u003c/strong\u003e, 519–530 (2015).\u003c/li\u003e\n\u003cli\u003eRathmell, W. K. \u003cem\u003eet al.\u003c/em\u003e Management of Metastatic Clear Cell Renal Cell Carcinoma: ASCO Guideline. \u003cem\u003eJ Clin Oncol\u003c/em\u003e \u003cstrong\u003e40\u003c/strong\u003e, 2957–2995 (2022).\u003c/li\u003e\n\u003cli\u003eTannir, N. M. \u003cem\u003eet al.\u003c/em\u003e Nivolumab plus ipilimumab versus sunitinib for first-line treatment of advanced renal cell carcinoma: extended 8-year follow-up results of efficacy and safety from the phase III CheckMate 214 trial. \u003cem\u003eAnn Oncol\u003c/em\u003e \u003cstrong\u003e35\u003c/strong\u003e, 1026–1038 (2024).\u003c/li\u003e\n\u003cli\u003eRini, B. I. \u003cem\u003eet al.\u003c/em\u003e Pembrolizumab plus Axitinib versus Sunitinib for Advanced Renal-Cell Carcinoma. \u003cem\u003eNew England Journal of Medicine\u003c/em\u003e \u003cstrong\u003e380\u003c/strong\u003e, 1116–1127 (2019).\u003c/li\u003e\n\u003cli\u003eChoueiri, T. K. \u003cem\u003eet al.\u003c/em\u003e Nivolumab plus Cabozantinib versus Sunitinib for Advanced Renal-Cell Carcinoma. \u003cem\u003eNew England Journal of Medicine\u003c/em\u003e \u003cstrong\u003e384\u003c/strong\u003e, 829–841 (2021).\u003c/li\u003e\n\u003cli\u003eMotzer, R. J. \u003cem\u003eet al.\u003c/em\u003e Nivolumab plus Ipilimumab versus Sunitinib in Advanced Renal-Cell Carcinoma. \u003cem\u003eNew England Journal of Medicine\u003c/em\u003e \u003cstrong\u003e378\u003c/strong\u003e, 1277–1290 (2018).\u003c/li\u003e\n\u003cli\u003eRini, B. I. \u003cem\u003eet al.\u003c/em\u003e Pembrolizumab plus axitinib versus sunitinib for advanced clear cell renal cell carcinoma: 5-year survival and biomarker analyses of the phase 3 KEYNOTE-426 trial. \u003cem\u003eNat Med\u003c/em\u003e 1–10 (2025) doi:10.1038/s41591-025-03867-5.\u003c/li\u003e\n\u003cli\u003eMotzer, R. J. \u003cem\u003eet al.\u003c/em\u003e Nivolumab plus cabozantinib versus sunitinib in first-line treatment for advanced renal cell carcinoma (CheckMate 9ER): long-term follow-up results from an open label, randomised, phase 3 trial. \u003cem\u003eLancet Oncol\u003c/em\u003e \u003cstrong\u003e23\u003c/strong\u003e, 888–898 (2022).\u003c/li\u003e\n\u003cli\u003eMotzer, R. J. \u003cem\u003eet al.\u003c/em\u003e Characterization of Responses to Lenvatinib plus Pembrolizumab in Patients with Advanced Renal Cell Carcinoma at the Final Prespecified Survival Analysis of the Phase 3 CLEAR Study. \u003cem\u003eEur Urol\u003c/em\u003e \u003cstrong\u003e86\u003c/strong\u003e, 4–9 (2024).\u003c/li\u003e\n\u003cli\u003ede Groot, S. \u003cem\u003eet al.\u003c/em\u003e Health-related quality of life and its determinants in patients with metastatic renal cell carcinoma. \u003cem\u003eQual Life Res\u003c/em\u003e \u003cstrong\u003e27\u003c/strong\u003e, 115–124 (2018).\u003c/li\u003e\n\u003cli\u003eCronin, C., O’Sullivan, A. \u0026amp; Naidoo, J. Long-Term Toxicity of Immune Checkpoint Inhibitors—Time to Widen the Lens. \u003cem\u003eJAMA Oncology\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 371–372 (2025).\u003c/li\u003e\n\u003cli\u003eKhan, H. M., Ramsey, S. \u0026amp; Shankaran, V. Financial Toxicity in Cancer Care: Implications for Clinical Care and Potential Practice Solutions. \u003cem\u003eJCO\u003c/em\u003e \u003cstrong\u003e41\u003c/strong\u003e, 3051–3058 (2023).\u003c/li\u003e\n\u003cli\u003eGupta, A., Eisenhauer, E. A. \u0026amp; Booth, C. M. The Time Toxicity of Cancer Treatment. \u003cem\u003eJCO\u003c/em\u003e \u003cstrong\u003e40\u003c/strong\u003e, 1611–1615 (2022).\u003c/li\u003e\n\u003cli\u003eGeorge, S. \u003cem\u003eet al.\u003c/em\u003e The Impact of Adverse Events on Health Care Resource Utilization, Costs, and Mortality Among Patients Treated with Immune Checkpoint Inhibitors. \u003cem\u003eOncologist\u003c/em\u003e \u003cstrong\u003e26\u003c/strong\u003e, e1205–e1215 (2021).\u003c/li\u003e\n\u003cli\u003eShah, N. J. \u003cem\u003eet al.\u003c/em\u003e Health Care Resource Use for Modern First-Line Treatments in Metastatic Renal Cell Carcinoma. \u003cem\u003eJAMA Network Open\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, e2422674 (2024).\u003c/li\u003e\n\u003cli\u003eRegan, M. M. \u003cem\u003eet al.\u003c/em\u003e Treatment-Free Survival: A Novel Outcome Measure of the Effects of Immune Checkpoint Inhibition-A Pooled Analysis of Patients With Advanced Melanoma. \u003cem\u003eJ Clin Oncol\u003c/em\u003e \u003cstrong\u003e37\u003c/strong\u003e, 3350–3358 (2019).\u003c/li\u003e\n\u003cli\u003eMantia, C. M. \u003cem\u003eet al.\u003c/em\u003e Treatment-free survival and partitioned survival analysis of patients with advanced renal cell carcinoma treated with nivolumab plus ipilimumab versus sunitinib: 5-year update of CheckMate 214. \u003cem\u003eJ Immunother Cancer\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, e009495 (2024).\u003c/li\u003e\n\u003cli\u003eKlapper, J. A. \u003cem\u003eet al.\u003c/em\u003e High-dose interleukin-2 for the treatment of metastatic renal cell carcinoma. \u003cem\u003eCancer\u003c/em\u003e \u003cstrong\u003e113\u003c/strong\u003e, 293–301 (2008).\u003c/li\u003e\n\u003cli\u003eAtkins, M. B., Regan, M. \u0026amp; McDermott, D. Update on the role of interleukin 2 and other cytokines in the treatment of patients with stage IV renal carcinoma. \u003cem\u003eClin Cancer Res\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 6342S–6S (2004).\u003c/li\u003e\n\u003cli\u003eMuhitch, J. B. \u0026amp; and Schwaab, T. High-dose IL-2 for Metastatic Renal Cell Carcinoma: Can the First Antitumor Immunotherapy be Reinvented? \u003cem\u003eImmunotherapy\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 955–958 (2014).\u003c/li\u003e\n\u003cli\u003eKhoja, L., Butler, M. O., Kang, S. P., Ebbinghaus, S. \u0026amp; Joshua, A. M. Pembrolizumab. \u003cem\u003eJournal for ImmunoTherapy of Cancer\u003c/em\u003e \u003cstrong\u003e3\u003c/strong\u003e, 36 (2015).\u003c/li\u003e\n\u003cli\u003eChoueiri, T. K. \u003cem\u003eet al.\u003c/em\u003e Immunomodulatory Activity of Nivolumab in Metastatic Renal Cell Carcinoma. \u003cem\u003eClin Cancer Res\u003c/em\u003e \u003cstrong\u003e22\u003c/strong\u003e, 5461–5471 (2016).\u003c/li\u003e\n\u003cli\u003eTannir, N. M. \u003cem\u003eet al.\u003c/em\u003e Bempegaldesleukin Plus Nivolumab Versus Sunitinib or Cabozantinib in Previously Untreated Advanced Clear Cell Renal Cell Carcinoma: A Phase III Randomized Study (PIVOT-09). \u003cem\u003eJournal of Clinical Oncology\u003c/em\u003e (2024) doi:10.1200/JCO.23.02082.\u003c/li\u003e\n\u003cli\u003eChatzkel, J. \u003cem\u003eet al.\u003c/em\u003e Coordinated Pembrolizumab and High Dose IL-2 (5-in-a-Row Schedule) for Therapy of Metastatic Clear Cell Renal Cancer. \u003cem\u003eClinical Genitourinary Cancer\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, 252–259 (2022).\u003c/li\u003e\n\u003cli\u003eKalinich, M. \u003cem\u003eet al.\u003c/em\u003e Prediction of severe immune-related adverse events requiring hospital admission in patients on immune checkpoint inhibitors: study of a population level insurance claims database from the USA. \u003cem\u003eJ Immunother Cancer\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, e001935 (2021).\u003c/li\u003e\n\u003cli\u003eZheng, Y. \u003cem\u003eet al.\u003c/em\u003e Real-World Clinical and Economic Outcomes in Selected Immune-Related Adverse Events Among Patients with Cancer Receiving Immune Checkpoint Inhibitors. \u003cem\u003eOncologist\u003c/em\u003e \u003cstrong\u003e26\u003c/strong\u003e, e2002–e2012 (2021).\u003c/li\u003e\n\u003cli\u003eWang, L. X. \u003cem\u003eet al.\u003c/em\u003e Health care utilization and steroid-refractory toxicities from immune checkpoint inhibitors. \u003cem\u003eCancer\u003c/em\u003e \u003cstrong\u003e126\u003c/strong\u003e, 322–328 (2020).\u003c/li\u003e\n\u003cli\u003eHuang, R. S. \u003cem\u003eet al.\u003c/em\u003e Patient-Reported Outcomes as Prognostic Indicators for Overall Survival in Cancer: A Systematic Review and Meta-Analysis. \u003cem\u003eJAMA Oncol\u003c/em\u003e (2025) doi:10.1001/jamaoncol.2025.3153.\u003c/li\u003e\n\u003cli\u003eChang, E. \u003cem\u003eet al.\u003c/em\u003e A Pooled Analysis of Treatment-Free Survival in Advanced Renal Cell Carcinoma. \u003cem\u003eClin Cancer Res\u003c/em\u003e \u003cstrong\u003e30\u003c/strong\u003e, 3282–3286 (2024).\u003c/li\u003e\n\u003cli\u003eGupta, M. \u003cem\u003eet al.\u003c/em\u003e Treatment-Free Survival After Nivolumab vs Pembrolizumab vs Nivolumab-Ipilimumab for Advanced Melanoma. \u003cem\u003eJAMA Netw Open\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, e2319607 (2023).\u003c/li\u003e\n\u003cli\u003eFletcher, K. \u0026amp; Johnson, D. B. Chronic immune-related adverse events arising from immune checkpoint inhibitors: an update. \u003cem\u003eJ Immunother Cancer\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, e008591 (2024).\u003c/li\u003e\n\u003cli\u003eH, S. \u003cem\u003eet al.\u003c/em\u003e Immune-related adverse events and kidney function decline in patients with genitourinary cancers treated with immune checkpoint inhibitors. \u003cem\u003eEuropean journal of cancer (Oxford, England : 1990)\u003c/em\u003e \u003cstrong\u003e157\u003c/strong\u003e, (2021).\u003c/li\u003e\n\u003cli\u003eChoueiri, T. K. \u003cem\u003eet al.\u003c/em\u003e Lenvatinib plus pembrolizumab versus sunitinib as first-line treatment of patients with advanced renal cell carcinoma (CLEAR): extended follow-up from the phase 3, randomised, open-label study. \u003cem\u003eLancet Oncol\u003c/em\u003e \u003cstrong\u003e24\u003c/strong\u003e, 228–238 (2023).\u003c/li\u003e\n\u003cli\u003eYanagisawa, T. \u003cem\u003eet al.\u003c/em\u003e Updated systematic review and network meta-analysis of first-line treatments for metastatic renal cell carcinoma with extended follow-up data. \u003cem\u003eCancer Immunol Immunother\u003c/em\u003e \u003cstrong\u003e73\u003c/strong\u003e, 38 (2024).\u003c/li\u003e\n\u003cli\u003ePlimack, E. R. \u003cem\u003eet al.\u003c/em\u003e Pembrolizumab Plus Axitinib Versus Sunitinib as First-line Treatment of Advanced Renal Cell Carcinoma: 43-month Follow-up of the Phase 3 KEYNOTE-426 Study. \u003cem\u003eEuropean Urology\u003c/em\u003e \u003cstrong\u003e84\u003c/strong\u003e, 449–454 (2023).\u003c/li\u003e\n\u003cli\u003eYang, J. C. \u003cem\u003eet al.\u003c/em\u003e Randomized Study of High - Dose and Low - Dose Inter leukin - 2 in Patients With Metastatic Renal Cancer. \u003cem\u003eJ Clin Oncol\u003c/em\u003e \u003cstrong\u003e21\u003c/strong\u003e, 3127–3132 (2003).\u003c/li\u003e\n\u003cli\u003eFishman, M. \u003cem\u003eet al.\u003c/em\u003e Overall survival by clinical risk category for high dose interleukin-2 (HD IL-2) treated patients with metastatic renal cell cancer (mRCC): data from the PROCLAIMSM registry. \u003cem\u003ej. immunotherapy cancer\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, 84 (2019).\u003c/li\u003e\n\u003cli\u003eSharma, M. \u003cem\u003eet al.\u003c/em\u003e Bempegaldesleukin selectively depletes intratumoral Tregs and potentiates T cell-mediated cancer therapy. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 661 (2020).\u003c/li\u003e\n\u003cli\u003eCharych, D. \u003cem\u003eet al.\u003c/em\u003e Modeling the receptor pharmacology, pharmacokinetics, and pharmacodynamics of NKTR-214, a kinetically-controlled interleukin-2 (IL2) receptor agonist for cancer immunotherapy. \u003cem\u003ePLoS One\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, e0179431 (2017).\u003c/li\u003e\n\u003cli\u003eBeebe, K. D. \u003cem\u003eet al.\u003c/em\u003e The Immunogenomic Landscape of Peripheral High-Dose IL-2 Pharmacodynamics in Patients with Metastatic Renal Cell Carcinoma: A Benchmark for Next-Generation IL-2-Based Immunotherapies. \u003cem\u003eJ Immunol\u003c/em\u003e \u003cstrong\u003e213\u003c/strong\u003e, 29–39 (2024).\u003c/li\u003e\n\u003cli\u003eLaPorte, K. M., Hernandez, R., Santos Savio, A. \u0026amp; Malek, T. R. Robust IL-2-dependent antitumor immunotherapy requires targeting the high-affinity IL-2R on tumor-specific CD8+ T cells. \u003cem\u003eJ Immunother Cancer\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, e006611 (2023).\u003c/li\u003e\n\u003cli\u003eWu, W. \u003cem\u003eet al.\u003c/em\u003e IL-2Rα-biased agonist enhances antitumor immunity by invigorating tumor-infiltrating CD25+CD8+ T cells. \u003cem\u003eNat Cancer\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, 1309–1325 (2023).\u003c/li\u003e\n\u003cli\u003eEisner, J. R. \u003cem\u003eet al.\u003c/em\u003e Distinct Predictive Immunogenomic Profiles of Response to Immune Checkpoint Inhibitors and IL2: A Real-world Evidence Study of Patients with Advanced Renal Cancer. \u003cem\u003eCancer Research Communications\u003c/em\u003e \u003cstrong\u003e2\u003c/strong\u003e, 894–903 (2022).\u003c/li\u003e\n\u003cli\u003eXu, W. \u003cem\u003eet al.\u003c/em\u003e Progressive natural killer cell dysfunction in advanced-stage clear-cell renal cell carcinoma and association with clinical outcomes. \u003cem\u003eESMO Open\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 104105 (2025).\u003c/li\u003e\n\u003cli\u003eAndo, S. \u0026amp; Araki, K. CD8 T-cell heterogeneity during T-cell exhaustion and PD-1-targeted immunotherapy. \u003cem\u003eInt Immunol\u003c/em\u003e \u003cstrong\u003e34\u003c/strong\u003e, 571–577 (2022).\u003c/li\u003e\n\u003cli\u003eHa, H., Debnath, B. \u0026amp; Neamati, N. Role of the CXCL8-CXCR1/2 Axis in Cancer and Inflammatory Diseases. \u003cem\u003eTheranostics\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, 1543–1588 (2017).\u003c/li\u003e\n\u003cli\u003eDavids, M. S. \u003cem\u003eet al.\u003c/em\u003e Ipilimumab for Patients with Relapse after Allogeneic Transplantation. \u003cem\u003eNew England Journal of Medicine\u003c/em\u003e \u003cstrong\u003e375\u003c/strong\u003e, 143–153 (2016).\u003c/li\u003e\n\u003cli\u003eHu, W. \u003cem\u003eet al.\u003c/em\u003e Prognostic and therapeutic relevance of IL2RG-related LncRNAs in clear cell renal cell carcinoma. \u003cem\u003eSci Rep\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 29651 (2025).\u003c/li\u003e\n\u003cli\u003eErnst, M. S. \u003cem\u003eet al.\u003c/em\u003e Outcomes for International Metastatic Renal Cell Carcinoma Database Consortium Prognostic Groups in Contemporary First-line Combination Therapies for Metastatic Renal Cell Carcinoma. \u003cem\u003eEuropean Urology\u003c/em\u003e \u003cstrong\u003e84\u003c/strong\u003e, 109–116 (2023).\u003c/li\u003e\n\u003cli\u003eDutcher, J. P. \u003cem\u003eet al.\u003c/em\u003e High dose interleukin-2 (Aldesleukin) - expert consensus on best management practices-2014. \u003cem\u003eJ Immunother Cancer\u003c/em\u003e \u003cstrong\u003e2\u003c/strong\u003e, 26 (2014).\u003c/li\u003e\n\u003cli\u003eTanigawa, K. \u0026amp; Redmond, W. L. Current landscape and future prospects of interleukin-2 receptor (IL-2R) agonists in cancer immunotherapy. \u003cem\u003eOncoimmunology\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 2452654.\u003c/li\u003e\n\u003cli\u003eIvanova, A., Qaqish, B. F. \u0026amp; Schell, M. J. Continuous Toxicity Monitoring in Phase II Trials in Oncology. \u003cem\u003eBiometrics\u003c/em\u003e \u003cstrong\u003e61\u003c/strong\u003e, 540–545 (2005).\u003c/li\u003e\n\u003cli\u003eSubramanian, A. \u003cem\u003eet al.\u003c/em\u003e Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. \u003cem\u003eProc Natl Acad Sci U S A\u003c/em\u003e \u003cstrong\u003e102\u003c/strong\u003e, 15545–15550 (2005).\u003c/li\u003e\n\u003cli\u003eFang, Z., Liu, X. \u0026amp; Peltz, G. GSEApy: a comprehensive package for performing gene set enrichment analysis in Python. \u003cem\u003eBioinformatics\u003c/em\u003e \u003cstrong\u003e39\u003c/strong\u003e, btac757 (2023).\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"RCC, immunotherapy, cytokine, treatment-free survival, pembrolizumab, interleukin-2, combination immunotherapy, durable survival, durable response, long-term follow up, multi-omics, proteomics, transcriptomics","lastPublishedDoi":"10.21203/rs.3.rs-7698261/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7698261/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Prolonged or indefinite systemic therapy remains standard for advanced clear cell renal cell carcinoma (ccRCC), often resulting in cumulative toxicities and treatment burden. We conducted a single-arm phase 2 trial of a fixed-duration regimen of pembrolizumab plus high-dose interleukin-2 in treatment-naive advanced ccRCC. Among 26 patients treated, after a median follow-up of 76.4 months, the objective response rate was 73%, with complete responses in 42%. Median overall survival was not reached, with 5-year survival of 73%; median progression-free survival was 19.3 months; and median treatment-free survival was 23.8 months, with 42% of patients remaining treatment-free at 5 years. No grade 5 adverse events occurred, and no patient with durable disease control experienced persistent grade ≥2 toxicities. Correlative analyses identified immune patterns associated with durable benefit, including enrichment of CD16⁺ natural killer cells, suppression of PD-1⁺ T-cell frequencies, and coordinated chemokine, complement, and PKC/TGF-β pathway activation. ClinicalTrials.gov identifier: NCT02964078.","manuscriptTitle":"Pembrolizumab plus High-Dose IL-2 in Advanced Clear Cell Renal Cell Carcinoma: Six-Year Survival Outcomes and Molecular Signatures from a Phase 2 Trial","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-08 09:55:46","doi":"10.21203/rs.3.rs-7698261/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-communications","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"NCOMMS","sideBox":"Learn more about [Nature Communications](http://www.nature.com/ncomms/)","snPcode":"","submissionUrl":"https://mts-ncomms.nature.com/","title":"Nature Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Communications","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"e953a0d1-8e40-433f-a96e-15f178e56cd7","owner":[],"postedDate":"October 8th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":55914172,"name":"Health sciences/Medical research/Clinical trial design/Clinical trials/Phase II trials"},{"id":55914173,"name":"Biological sciences/Cancer/Cancer therapy/Cancer immunotherapy"},{"id":55914174,"name":"Biological sciences/Cancer/Tumour biomarkers"},{"id":55914175,"name":"Biological sciences/Cancer/Tumour immunology"}],"tags":[],"updatedAt":"2026-02-03T18:40:57+00:00","versionOfRecord":[],"versionCreatedAt":"2025-10-08 09:55:46","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7698261","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7698261","identity":"rs-7698261","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}