Mesonephric and mesonephric-like adenocarcinomas of gynecologic origin: A single-center experience with molecular characterization, treatment, and oncologic outcomes.

Patients with MA (cervix) or MLA (uterus, ovary, other) treated at Memorial Sloan Kettering Cancer Center (MSK) from January 1, 2008, to December 31, 2021, were retrospectively identified using a natural language search in the institutional pathology database, as well as a review of the institutional clinical database using the keywords “mesonephric” and “mesonephric-like.” Cases were only included if the tumors had typical mesonephric morphology with supporting immunohistochemical expression patterns (ie, hormone receptor-negative status, or at most focal/weak, and positive GATA3 or TTF1 status). Slides or digitally scanned images underwent histologic re-review and confirmation of pathology by an expert gynecologic pathologist (KP), and clinical characteristics, treatment regimens, and oncologic outcomes were obtained from medical records. The prior Pors et al publication [ 3 ] included 48 cases from MSK, and this analysis included 39 cases not previously published. Somatic mutation and copy number alteration status were obtained from the clinical US Food and Drug Administration (FDA)-authorized tumor-normal sequencing panel, MSK-Integrated Mutation Profiling of Actionable Cancer Targets (MSK-IMPACT) [ 16 ]. Computed tomography images were retrospectively reviewed by an expert radiologist (JG) using Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1 to determine response to therapy for patients with available radiology images [ 17 ]. Waterfall plots of percent change in tumor size following first-line treatment for recurrence were generated for each origin of MA and MLA (uterus, cervix, ovary). For the entire cohort (including patients who received initial treatment at MSK), progression-free survival (PFS1) was calculated as the time from initial surgery to progression (as determined by the treating physician) or death. Second PFS (PFS2) was calculated as the time from start of treatment for recurrence to subsequent progression (as determined by the treating physician) or death. Overall survival (OS) was calculated as the time from surgery to the date of last follow-up or death. Median PFS1, PFS2, and OS were estimated using the Kaplan-Meier method, and the 95% confidence intervals were reported. Median follow-up was estimated using the reverse Kaplan-Meier method, which is calculated by reversing the censor and event status of patients. All analyses were performed using R version 4.2.1 (R Foundation for Statistical Computing, Vienna, Austria).

A total of 81 patients were confirmed to have gynecologic MA or MLA, 36 (44%) of whom received initial surgical management and treatment at our institution. Table 1 describes the demographic and clinicopathologic characteristics of the entire cohort. For the overall cohort, median age at MA/MLA diagnosis was 60 years (IQR, 53–67 years). MA in the cervix was diagnosed in 9 patients (11%) and MLA in 72 patients (89%). Gynecologic site of origin for MLA was most commonly the uterus (n=42, 52%), followed by the ovary (n=28, 35%), and other (Mullerian not otherwise specified; n=2, 2%). Among the 42 cases of uterine MLA, 29 had mixed histology noted on the pathology report. Of these mixed histology tumors, 20 included endometrioid, 8 carcinosarcoma, and 1 clear cell. Among the 28 cases of ovarian MLA, 20 had mixed histology noted on the pathology report. Of these mixed histology tumors, 10 included endometrioid, 4 mucinous features or mucinous borderline, 3 carcinosarcoma, 2 serous borderline, and 1 low-grade serous. Among the entire cohort, 58 patients (72%) had recurrent disease, most commonly in the lung (n=28, 48%) ( Table 2 ). Among the 36 patients who were initially treated at our institution, 20 (56%) recurred ( Table 2 ). Initial recurrence rates (PFS1) were high regardless of stage at presentation. Median PFS1 rate at 24 months for the 36 patients who underwent initial treatment at MSK, by stage, was 71% (95% CI: 52%–96%) for stage I, 67% (95% CI: 30%–100%) for stage II, 27% (95% CI: 8.9%–80%) for stage III, and 50% (95% CI: 19%–100%) for stage IV disease ( Figure 2 ). For the 20 patients with recurrent disease who received initial treatment at MSK, 11 (55%) received platinum-based systemic treatment, 2 (10%) received liposomal doxorubicin with or without bevacizumab, 2 (10%) underwent secondary cytoreductive surgery, 1 (5%) received immunotherapy, 1 (5%) enrolled on a clinical trial, 1 (5%) received hormonal treatment, 1 (5%) received targeted therapy (temsirolimus), and 1 (5%) received unknown treatment ( Table 2 ). One patient received an MEK inhibitor. Median follow-up for the entire cohort was 64.9 months (95% CI: 56.5–98.5 months). Among all 81 patients, median OS was 83.8 months (95% CI: 73-not evaluable [NE]). For the 36 patients who received initial treatment at MSK, median follow-up was 55.8 months (95% CI: 46.8–73.3 months). Median PFS1 and PFS2 were 33 months (95% CI: 17-NE) and 8.3 months (95% CI: 6.9–14 months), respectively ( Figures 3A and 3B ). Median OS was 87 months (95% CI: 58-NE) ( Figure 3C ). Response to systemic therapy at time of first recurrence is displayed in the waterfall plots shown in Supplementary Figure S1 . Fourteen patients with uterine MLA were evaluable; 9 received platinum-doublet chemotherapy, 3 received immunotherapy, 1 received hormonal therapy, and 1 received targeted therapy. The response rate among the 9 patients treated with platinum-doublet chemotherapy was 77.8% (n=7; 1 complete and 6 partial responses). Patients who received both targeted (temsirolimus) and hormonal therapies had a best response of stable disease. The 3 patients who received immunotherapy demonstrated a best response of progression of disease (POD), though notably all had mismatch repair-proficient disease and/or had a low tumor mutational burden. For the patients with uterine MLA and initial treatment at MSK, median PFS2 was 8.4 months (95% CI: 6.9-NE). Among the 7 patients with ovarian MLA who were evaluable, 4 received platinum-doublet chemotherapy, 1 received platinum-doublet therapy with bevacizumab, 1 received chemotherapy with bevacizumab, and 1 received immunotherapy. No patients had a CR. Of the 4 patients who received platinum-doublet chemotherapy, 1 (25%) had a PR, 2 (50%) had SD, and 1 (25%) had POD. The remaining 3 patients had SD on platinum-doublet therapy with bevacizumab treatment, chemotherapy and bevacizumab, and immunotherapy, respectively. For the patients with ovarian MLA who received initial treatment at MSK, the median PFS2 was 7.9 months (95% CI: 3.9-NE). Three patients with cervical MA were evaluable. No patients had a CR or PR. The patient who received platinum-doublet therapy and the patient who received immunotherapy (and notably had programmed death-ligand 1 [PD-L1]-positive immunohistochemistry [IHC]) had SD. The patient on platinum-doublet therapy progressed after 5 months on treatment, and the patient on immunotherapy progressed after 22 months on treatment. The patient who enrolled on a clinical trial demonstrated a best response of POD. Twenty-six (72%) of the 36 patients who were initially treated at our institution underwent clinical molecular profiling via MSK-IMPACT. All but 1 MA/MLA harbored MAPK pathway alterations (n=25, 96%), including KRAS (n=20, 77%) or NRAS (n=3, 11%) hotspot missense mutations, HRAS amplification (n=1, 4%), or a NF1 truncating mutation (n=1, 4%) ( Figure 4 ). MAPK pathway alterations were generally mutually exclusive, except for 1 ovarian MLA that harbored both a hotspot KRAS and a hotspot NRAS mutation. PIK3CA hotspot mutations were found in 5 (30%) of 15 uterine MLAs and in 3 (30%) of 9 ovarian MLAs ( Figure 4 ).

Here, we report on the molecular features, recurrent treatment paradigms, and oncologic outcomes of patients with MA or MLA of gynecologic origin from a single institution. Our data demonstrate that 96% of these tumors harbored an MAPK pathway alteration. Recurrent disease most commonly occurred in the lung (48%), and once recurrent, median PFS2 was 8.3 months. A large retrospective study of 99 patients with MA or MLA of gynecologic origin found similarly poor oncologic outcomes as our study [ 3 ]. While our institution contributed cases to this large retrospective study, in the current study, we focused on patients who underwent initial treatment at MSK and expanded upon systemic therapy response and molecular findings. Consistent with our results, Pors et al demonstrated a high rate of patients who presented with advanced-stage disease (II-IV); 60% of patients had cervical MA, 58% had uterine MLA, and 39% had ovarian MLA [ 3 ]. Pors et al also demonstrated a high recurrence rate of 52% among the entire cohort, which is consistent with our finding of a 56% recurrence rate in the cohort of patients who received initial treatment at MSK. Importantly, congruent with Pors et al and other retrospective studies, our data demonstrate a propensity of recurrences in the lung [ 1 – 3 , 9 ]. Our finding of a median PFS2 of 8.3 months for patients who underwent initial treatment at MSK supports prior analyses showing PFS is limited in the setting of recurrence [ 3 , 10 – 12 ]. Beyond clinicopathologic characteristics and oncologic outcomes, we assessed response to systemic treatment in the recurrent setting for this rare histology. Immunotherapy, even in 1 patient with cervical MA and positive PD-L1 IHC, did not result in tumor regression. In a case report, Montagut et al reported on a patient with recurrent uterine MLA as well as a review of the available literature on systemic treatment response in mesonephric gynecologic cancer [ 18 ]. The patient was diagnosed with recurrence (in the lungs and peritoneum) 10 months after initial surgical management and demonstrated a durable response to carboplatin and paclitaxel. Though limited by a small sample, our data support the use of platinum-based chemotherapy for patients with uterine MLA. The limited PFS observed, however, indicates a need for improved therapeutics in this rare subset of gynecologic cancers. MA and MLA of gynecologic origin demonstrate nearly ubiquitous somatic alterations in the MAPK signaling pathway. In our study, 96% of patients with MA or MLA of gynecologic origin had a somatic alteration in the MAPK signaling pathway, with most having a somatic KRAS mutation. Similarly, da Silva et al found a high prevalence of somatic alterations in the MAPK signaling pathway in patients with MA or MLA of gynecologic origin; 89% of patients had alterations in KRAS , 7% in NRAS , and 4% in BRAF [ 2 ]. Importantly, da Silva et al demonstrated that mutations affecting the RAS/MAPK family of genes were clonal throughout the metastatic process when comparing primary and paired metastatic samples [ 2 ]. This evidence, which is concordant with our results demonstrating poor PFS in the recurrent setting, suggests patients with MA or MLA may benefit from therapy targeting the MAPK signaling pathway. Low-grade serous ovarian carcinoma (LGSOC) is a rare type of ovarian cancer characterized by a high prevalence of somatic MAPK pathway alterations, most commonly in KRAS , which is altered in one-third of LGSOC tumors [ 19 – 21 ]. Multiple recent studies of single-agent MEK inhibitors in LGSOC have shown promising results, particularly in patients with a KRAS mutation, in whom the response rate to treatment with binimetinib was 44% [ 22 , 23 ]. A similar strategy may be beneficial for patients with MA or MLA. Combination strategies have also shown promising results in patients with LGSOC, particularly in those harboring a KRAS mutation [ 24 ]. The combination of VS-6766 and defactinib showed encouraging response with durability across patients with LGSOC, with response rates of 46% in all patients and 64% in patients with KRAS -mutated LGSOC [ 24 ]. Strengths of this study include the large cohort of 81 patients, including 36 patients who underwent initial treatment at our institution. Despite this being one of the largest retrospective studies of this rare histology, the sample included in this analysis was small, which makes it difficult to interpret oncologic outcomes and tumor responses to systemic treatments. This study is also limited by its retrospective nature, specifically the retrospective radiologic image review to determine tumor response to treatment. Moreover, the treatment regimen was at the discretion of the treating physician, and therefore it was likely focused on the gynecologic organ of origin rather than mesonephric histology. Ultimately, prospective data on response to systemic therapy for patients with MA or MLA of gynecologic origin are limited. In our study, 56% of patients who received initial treatment at our institution demonstrated recurrence, most commonly involving the lung. Importantly, there appears to be a high prevalence of MAPK pathway alterations in MA and MLA tumors. Therapies targeting the MAPK pathway recently have shown promising results in other rare gynecologic cancers with a high prevalence of KRAS mutations and are worthy of further study in MA and MLA of gynecologic origin. Toward this end, a single-institution study ( NCT05787561 ) of the dual MEK/RAF inhibitor avutometinib in combination with the FAK inhibitor defactinib in patients with recurrent MA or MLA of gynecologic origin is currently accruing patients.

Mesonephric adenocarcinoma (MA) and mesonephric-like adenocarcinoma (MLA) of the gynecologic tract are rare cancers that frequently harbor alterations in mitogen-activated protein kinase (MAPK) pathway-related genes, most commonly in KRAS [ 1 – 4 ]. MA occurs in the uterine cervix, is not associated with human papillomavirus, and arises from mesonephric (Wolffian) duct remnants within the deep cervical wall [ 5 – 8 ]. MLA arises in the uterine corpus, ovary, para-adnexal soft tissue, or vagina [ 9 – 14 ], typically does not have identifiable Wolffian duct remnant origins, and is thought to arise from transdifferentiation of Mullerian epithelium, including endometriosis [ 15 ]. MA and MLA are histologically similar, often express GATA3, thyroid transcription factor-1 (TTF-1; particularly MLA), calretinin, and luminal CD10, and are generally estrogen receptor and progesterone receptor negative, or at most focally positive ( Figure 1 ) [ 3 ]. To date, no prospective therapeutic studies have been completed in patients with MA or MLA of gynecologic origin. We aim to describe the molecular features, response to treatment, and oncologic outcomes for patients with MA or MLA of gynecologic origin who were treated at a single institution.