First-in-Human Safety, Tolerability, Efficacy, and Pharmacokinetics of Pegfosimer Manganese (SN132D) for Contrast-Enhanced MRI of Breast Cancer.

The FIH phase I study was designed as an open-label, nonrandomized, non–placebo-controlled study of the safety, tolerability, pharmacokinetics (PK), and preliminary efficacy of the contrast agent pegfosimer manganese in newly diagnosed breast cancer patients. The study was conducted at Sahlgrenska University Hospital, Gothenburg, Sweden, and Uppsala Academic Hospital, Uppsala, Sweden, from September 2019 to November 2022 ( NCT04080024 ). A written informed consent was obtained from each study participant before any study protocol activities took place. The trial was approved by the Swedish Independent Ethics Committee and conducted following Good Clinical Practice (GCP) guidelines of the International Conference on Harmonization and the 2013 World Medical Association Declaration of Helsinki 13 and subsequent updates. Female patients at least 18 years of age with histological or cytological diagnosis of breast cancer (tumor >5 mm; cT1c-cT4; cN0-cN3; Mx) and signed informed consent were considered eligible to participate in the study. In total, 16 female breast cancer patients were included, and 14 of those fulfilled the inclusion/exclusion criteria for participation in the study. Histologically or cytologically confirmed breast cancer patients, aged 18 or above, exhibiting an Eastern Cooperative Oncology Group 14 performance status of maximum 1, adequate renal and hepatic function at screening, who were willing to undertake 3 preoperative MRI investigations in 1 day were eligible to participate in the study. Patients who had any condition that was contraindicated with the MRI examination, including but not limited to body mass index (BMI) >40 kg/m 2 , claustrophobia, history of brain or heart surgery, metallic implant (eg, pacemaker, cochlear implant), permanent make-up, work as metalworkers or welders, or inability to stay in prone position for 45 minutes, were excluded from the study. Pegfosimer manganese (supplied as a 65-mM Mn sterile solution manufactured by Basic Pharma Manufacturing PV, Geleen, the Netherlands) was thawed overnight in a refrigerator and brought to room temperature at least 1 hour before the administration of 10, or 20 μmol Mn/kg, given as a single intravenous infusion over 1 hour at a rate of 5 mL/h, in combination with a saline carrier. The total administered volume varied based on the body weight. As a precaution for acute hypersensitivity reactions related to the infusion, pegfosimer manganese was administered using a slow intravenous infusion for 1 hour. To warrant any potential infusion-related reactions in response to the polyethylene glycol–coating of pegfosimer manganese and toxicity-related reactions from Mn, the protocol allowed for enough time to detect potential acute and long-term toxicities; the first 3 participants in each cohort were dosed at least 1 week apart, and the remaining participants were at least 24 hours apart. The study was designed as a dose escalation study of 3 dose levels (10, 20, and 30 μmol Mn/kg) administered to 3 sequential cohorts, allowing for improvement of MRI quality. The formal stopping rule for dose escalation (Table 1 ) was strictly followed as per the Common Terminology Criteria for Adverse Events (CTCAE) v5.0, 15 considering the recommendations published by Sibille et al, 16 an adaptation to FIH studies of the grading system proposed by the World Health Organization, 17 National Institutes of Health, 18 and Food and Drug Administration. 19 To proceed with any dose escalations, the safety, PK, and MRI scan quality was reviewed by an independent safety review committee (iSRC). Applied Dose Escalation Stopping Rules AEs indicates adverse events; AUC, area under the curve; C max , maximum plasma concentration; iSRC, independent safety review committee, Mn, manganese; SADR, serious adverse drug reaction; SAE, serious adverse event; SOC, system organ class. The study duration for each participant was approximately 24 to 30 days, including a 10-day screening period and an approximate 2- to 3-week follow-up period. Participants attended at least 3 visits related to the clinical study. At screening, demographics, physical examination, prior and concomitant medications, and eligibility were recorded. Screening took place within 10 days of the planned dosing day. On the day of dosing, eligible participants were admitted to the FIH unit for assessment of baseline parameters, including a baseline unenhanced MRI scan to determine baseline signals in target tissues, following study drug administration as a single intravenous infusion over 1 hour together with a saline carrier. Postdose MRI scans were performed on 2 separate occasions, 1 or 2 hours and 4 hours after the end of the infusion, respectively. After the infusion, the participants were observed at the FIH unit for up to 25 hours with regular registration of safety and PK blood samplings, see below. The visit for safety follow-up evaluation was scheduled 14 days from the day of study drug administration. The overall safety, including AEs and serious AEs (SAEs), was assessed from the start of the pegfosimer manganese single intravenous infusion until the end-of-study visit by medical clinicians. The grading of the severity/intensity (grade 1 to grade 5) of AEs followed the CTCAE v5.0. 15 AEs were assessed as unlikely, possibly, or probably related to the study drug. Injection site reactions (redness, itching, pain, swelling, bruising, burning) were reported as AEs. A single 12-lead electrocardiogram (ECG) was recorded in a supine position after 10 minutes of rest using an ECG machine predosing, 30 minutes, 1 hour, 2 hours, 4 hours, 7 hours, and 9 hours postdosing start, and at the safety follow-up visit to assess heart rate, PQ/PR, QRS, QT, and QTcF intervals. All ECG values were reviewed and interpreted on site by the investigator. A complete physical examination included assessments of the head, eyes, ears, nose, throat, skin, thyroid, neurological, lungs, cardiovascular, abdomen, lymph nodes, and extremities and was conducted predosing, and 9 hours postdosing. Vital signs, comprising, systolic and diastolic blood pressure, pulse, respiratory rate, saturation, and body temperature were measured in a supine position after 5 minutes of rest predosing, 30 minutes, 1 hour, 2 hours, 4 hours, 7 hours, 9 hours postdosing start, and at the safety follow-up visit. Blood samples for analysis of clinical chemistry and hematology parameters were collected predosing, 9 hours postdosing start, and at the safety follow-up visit through venipuncture or a peripheral venous catheter. Plasma concentrations of Mn were determined by sampling approximately 5 mL venous blood, collected through a peripheral venous catheter of the study drug at the following time points postdosing start; 60 minutes before (baseline), 0 minutes, 10 minutes, 20 minutes, 60 minutes, 65 minutes, 75 minutes, 90 minutes, 105 minutes, 2 hours, 3 hours, 5 hours, 9 hours postdosing start, and lastly 1 sample between day 3 to day 14. The time point of PK sampling did not deviate more than ±10% from the planned time. The blood samples were collected in prelabeled lithium heparin tubes. The collected blood samples were kept at approximately 4°C and were, within 1 hour, centrifuged in a refrigerated centrifuge at 1800 g for 10 minutes to separate the plasma and frozen at −20°C. Analysis was based on the actual sampling times recorded during the study. To calculate PK parameters, concentrations below the lower limit of quantification (LLOQ) occurring before C max was defined as zero. Concentrations below LLOQ occurring after C max were omitted from the analysis. C max , T max , and T last were derived from the observed plasma concentration data. AUC was derived by integration of the plasma concentration versus time curve using linear interpolation for increasing plasma levels and logarithmic interpolation for decreasing plasma levels (linear up-log down method). AUC 0-t was calculated from time 0 to the time t of the last detectable plasma concentration. For AUC 0-inf , the area was calculated to the last time point with a measurable plasma concentration and then extrapolated to infinity using the concentration in the last quantifiable sample and lambda z . The plasma elimination half-life (T ½z ) was calculated by ln 2/lambda z . The initial plasma half-life (T 1/2initial ) was estimated by linear-logarithmic regression based on the observed concentration at the end of infusion and the observed concentrations at 65 and 75 minutes after the end of infusion. The imaging protocol was optimized on both sites via test scanning at the clinical sites of phantoms, calibrated with 0.5 mmol/mL Gd Dotarem, and healthy volunteer test persons. The test scan sequences comprised THRIVE Noise from the phantom, and Survey, THRIVE, THRIVE Noise, STIR, SPAIR, DWI, MOLLI, and VFA from the test persons. Breast scanning was performed on a 1.5 T MRI Philips Achieva dStream scanner with the dedicated Philips breast-coil BREAST 7CH, or, if participants were too large to fit the scan setup, flex-coil wrapping around the participant lying flat on the scanner bed (Uppsala Academic Hospital), or a 3 T MRI Philips Ingenia scanner with the dedicated Philips breast-coil BREAST DCI 10CH (Sahlgrenska University Hospital). On each scanner, an anterior-posterior phased-array coil (Philips) was used for the field of view. The imaging settings agreed for the trial breast and abdominal MRI acquisitions were agreed to consist of the following sequences: STIR, THRIVE spoiled gradient echo, dynamic THRIVE, T1-weighted Modified Look-Locker inversion recovery (MOLLI), 2-point water Dixon. Before the first MRI acquisition, the participants were prepared according to the respective clinical site's standard requirements for MRI investigations, including the removal of metal objects such as eyeglasses and jewelry. The participants were instructed to wear ear protection and to perform breath-holding during the scanning. Each participant underwent the 30-minute long MRI scan 3 times: 1 or 2 hours, and 4 hours after the end of pegfosimer manganese infusion. The MOLLI served as the T1 quantification sequence. The quantitative image analysis was performed using a validated software, Core Lab, following an imaging processing pipeline developed by the imaging CRO (Antaros Medical, Mölndal). An external qualified radiologist evaluated the pseudonymized images. The radiologist was instructed to use all available information to increase their confidence in identifying the location of the primary tumor, liver, pancreas, and metastatic lesions. This included accessing all 3 time points, but also x-ray mammography images and noncontrast/contrast MRI examinations, if available. The measurements were produced by annotating the regions of interest to obtain the signal intensity and contrast intensity. The radiologist was instructed to place the region of interest (ROI) conservatively, to avoid regions with partial volume effects from large vessels or ducts. Upon identifying a structure, the radiologist marked it with an arrow or a circle. The annotated image sequences were then transferred to Antaros Medical AB for technical quality control and quantification of enhancement in the ROI. The images were individually windowed for black background and graded contrast by acquiring scans without any excitation pulses to generate images with no signal except for noise. Briefly, the resolution and spatial coverage were reduced to make acquisition times reasonable. This required those sequence to be acquired as the last scan in a dynamic series of a minimum of 2 images, an extra low-resolution signal scan was acquired to obtain a total of 3 images. The scanner type employed in the study only allowed this technique when image acceleration was turned off (geometry parameter “Uniformity” set to “Classic”). The implementation of acquiring the noise level from a scan with modified sequence settings enabled the noise to serve as a reference signal between imaging sessions. To assess the signal-to-noise ratio (SNR), the signal was determined from a scan with high resolution, whereas the noise was determined from the faster noise scan. Special consideration to the way image signal scaling factors (DICOM tag slope) employed by Philips scanners was taken. The contrast-to-noise ratio (CNR) was determined by subtracting the signal from a reference region from the analyzed tissue before dividing with the noise level of the noise scan. The chest and back muscle tissue were selected as the reference region for breast and abdominal scans, respectively. SNR and CNR were calculated as the absolute ROI contrast intensity changes from baseline (precontrast time point) versus each postcontrast time point relative to the reference tissues. All statistical analyses and descriptive summaries were performed using SAS version 9.4 (SAS Institute, Cary, NC). PK parameters were calculated by noncompartmental analysis using the software Phoenix WinNonlin version 8.3 (Certara, USA). Generally, no imputation of data was performed except for concentrations under LLOQ, which were replaced with LLOQ/2 if more than 50% of the values for a given time point were above LLOQ.

The study was conducted during the period between September 2019 and November 2022. The study population comprised 14 White females with breast cancer, with a mean age of 63 years and a mean BMI of 27.75 kg/m 2 . Six participants received 10 μmol Mn/kg (cohort 1), and 8 participants received 20 μmol Mn/kg (cohort 2) of pegfosimer manganese, respectively. The demographic and baseline characteristics are summarized in Table 2 . Overall, the data between cohorts 1 and 2 were comparable, except for a difference in the mean age (56.8 vs 67.6 years), and BMI was higher (29.6 vs 26.4 kg/m 2 ). Demographics and Baseline Characteristics BMI indicates body mass index; ECOG, Eastern Cooperative Oncology Group; n, number of participants; SD, standard deviation. Following the dosing of cohort 1 (n = 6) with 10 μmol Mn/kg pegfosimer manganese, a recommendation to escalate the dose was issued by the iSRC. The dosing of n = 4 participants in cohort 2 revealed that 20 μmol Mn/kg was sufficient to not proceed with escalation, and this cohort was expanded to a total of n = 8 participants, a number judged to be sufficient for assessment of the imaging clinical relevance. Pegfosimer manganese was generally tolerated by the participants in the study as assessed by the results from AEs, ECG, vital signs, safety laboratory parameters, and physical examinations. There were no deaths, SAEs, or AEs leading to withdrawal from the study. Twelve of the 14 included participants (n = 4 out of 6 in cohort 1, n = 8 out of 8 in cohort 2) reported a total of 29 AEs (Table 3 ). In general, AEs were more frequently reported by participants in cohort 2. Most AEs (25 out of 29) were assessed as possibly (10 events) or probably (15 events) related to pegfosimer manganese. The AEs were assessed as mild (18 events) or moderate (10 events) in intensity, except for 1 AE of increased transaminases experienced by a participant in cohort 2, which was assessed as severe and probably related to the study drug. All AEs of moderate to severe intensity were reported by participants in this cohort. Adverse Events Classified by the MedDRA System Organ Class and Preferred Term AEs indicates adverse events; E, number of events; MedDRA, Medical Dictionary for Regulatory Authorities; n, number of participants. Linear-logarithmic (lin-log) plot of the geometric mean manganese plasma concentration versus time in cohort 1 (10 μmol Mn/kg, n = 6, data points expressed as circles and curve-fit as solid line) and cohort 2 (20 μmol Mn/kg, n = 8, data points expressed as squares and curve-fit as dashed line). The most commonly reported AEs were “Flushing/hot flush” (5 participants; n = 3 in cohort 1, n = 2 in cohort 2), “Feeling hot” (3 participants; n = 3 in cohort 2), increased transaminase levels (4 participants; n = 4 in cohort 2), and nausea/vomiting (2 participants; n = 2 in cohort 2). There were no clinically relevant changes to baseline in clinical chemistry, hematology, urinalysis, or safety laboratory parameters over time, except for increased liver transaminases. The elevated transaminase levels, judged to be dose-dependent, were asymptomatic and not evident until 25 hours after the end of infusion. These elevations were transient, being assessed at the safety follow-up visit by the investigator as “Recovered/resolved.” All ECGs were interpreted as “Normal” or “Abnormal, not clinically significant” by the investigator at screening, predose, and all postdose assessments of the measurements. The PK parameters were calculated from a noncompartmental analysis using linear interpolation for increasing plasma levels and logarithmic interpolation for decreasing plasma levels (linear up-log-down method) over time. The peak Mn plasma concentrations over time occurred by the end of the 1-hour infusion and showed a C max of 0.9145 mg/L and 1.985 mg/L, for cohorts 1 and 2, respectively. After the end of infusion, the plasma concentration rapidly declined and was below LLOQ (0.1 mg/L) in most participants within 1 hour after the end of infusion (Fig. 1 ). The mean AUC 0-inf plasma exposure to pegfosimer manganese was approximately 2-fold as high in cohort 2 (1.913 h·mg/L), as compared with cohort 1 (1.008 h·mg/L). The mean initial half-life was approximately 7 minutes in both groups. The mean plasma CL was 0.5498 L/h/kg in cohort 1 and 0.6070 L/h/kg in cohort 2. Individual PK data are listed in Table 4 . Pharmacokinetics Parameters of Pegfosimer Manganese Plasma Mn-Levels AUC 0-inf indicates area under the plasma concentration-time curve from time 0 extrapolated to infinite time; AUC 0-last , area under the plasma concentration-time curve from time 0 extrapolated to last measurable time; CL, total body clearance; C max , maximum plasma concentration; n, number of participants; NC, not calculated; SD, standard deviation; T last , time of last observed plasma concentration; T max , time to C max ; T ½ , plasma elimination half-life; T 1/2initial , initial elimination half-life; T 1/2z , terminal plasma elimination half-life; V ss , volume of distribution at steady state; V z , volume of distribution associated with the terminal elimination phase. Contrast enhancement efficacy was evaluated based on the SNR and CNR changes from the baseline precontrast MRI as well as the ability to produce images of clinical relevance. Individual changes of increased absolute CNR and SNR in the primary breast tumor could be observed in 11 out of 14 participants (Table 5 ). Individual changes of increased absolute CNR and SNR in the liver and pancreas were also observed in all participants and cohorts (Supplementary Information S1, http://links.lww.com/RLI/B3 ). One participant in cohort 1 (1/6, 17%) and 1 participant in cohort 2 (1/8, 12.5%) did not obtain SNR and CNR calculations for the primary breast tumors, as the lesion was missed by the radiologist. One participant in cohort 2 (1/8, 12.5%, participant 9) did not complete the postcontrast scans due to discomfort. All participants (6/6, 100%) in cohort 1 were images with a 1.5 T field strength. In cohort 2, 2 participants (2/8, 25%) were imaged with a 1.5 T field strength, and 6 participants (6/8, 75%, remaining participants) were imaged with a field strength of 3 T. The reason for using 1.5 or 3 T field strength depended on which site the participants were recruited to. Individual Mean Contrast-to-Noise Ratios and Signal-to-Noise Ratios From Participants in Cohort 1 (10 μmol Mn/kg) and Cohort 2 (20 μmol Mn/kg) in the Primary Breast Tumor at Baseline Precontrast MRI, 2 or 3 Hours MRI Poststart of Infusion, and 5 Hours MRI Poststart of Infusion With Pegfosimer Manganese Bolded values denote positive changes from baseline MRI. NE indicates not evaluated. Pegfosimer manganese was subjectively evaluated for its ability to produce clinically relevant low-background MRI scans for clinical visualization of primary breast tumor (Fig. 2 ), liver, and pancreas (Fig. 3 ), as well as quantification of the SNR and CNR in ROIs. At the dose level of 20 μmol Mn/kg, the iSRC judged to expand the cohort to assess the efficacy of the study drug. Four representative T1-weighted MOLLI 3 T (A–C) and 1.5 T (D) MRI scan series of primary breast cancer tumor (arrows) baseline precontrast MRI (left column), 2 to 3 hours postdosing (post-dose 1, center column), and 5 hours postdosing (postdose 2, right column) after the start of the 1-hour infusion with 20 μmol Mn/kg pegfosimer manganese. The noise was individually windowed for background and graded contrast. Representative abdominal MRI T1-weighted 3 T scans before (left), 3 hours after (center), and 5 hours after (right) the 1-hour infusion of 20 μmol Mn/kg pegfosimer manganese.

The study was an FIH, open-label, phase 1 trial designed as a single intravenous infusion, dose-escalation study of the safety and tolerability of pegfosimer manganese, a novel macromolecular Mn-based contrast agent. Additionally, PKs and preliminary MRI contrast enhancement were explored. While several tumor types were feasible in the FIH study, based on the proposed passive targeting mechanism for pegfosimer manganese, the study objectives were first explored in participants with breast cancer. Once a pegfosimer manganese dose level that was safe and sufficient for MRI had been established, a separate pancreatic cancer patient expansion arm was included. However, no pancreatic cancer patients were included in the study. Pegfosimer manganese was tolerated at the dosage of 20 μmol Mn/kg, and given the predefined dose escalation stopping rule, no further dose escalation was made from this point, and the cohort was expanded to n = 8. Therefore, this dose level should not be considered as a maximum tolerated dose. Pegfosimer manganese was generally considered safe and well-tolerated by the participants in the study. There were no deaths, SAEs, or AEs leading to withdrawal of a participant from the study. Participants in the higher dose group tended to report more AEs. The most common AEs were flushing/hot flush, feeling hot, transaminases increased, and nausea/vomiting. Nausea and vomiting have commonly been reported as AEs in other studies with Mn-based contrast agents. 20 The investigator judged pegfosimer manganese to be emetic, suggesting that a prophylactic antiemetic treatment in further studies could benefit the patients. A further study has since been completed in a different population of younger females suffering from endometriosis ( NCT05664828 ), including coadministration of the antispasmodic, Buscopan, a standard clinical practice at the site for pelvic imaging. A transient nonclinically significant elevation in plasma transaminase levels was also observed in 2 participants. This elevation is suggestive that these adverse effects are likely related to a release of manganese ions from the nanoparticles, a profile similar to other manganese-based contrast agents. 21 The independent radiologist judged the pegfosimer manganese-mediated MRI contrast enhancement to be of clinical relevance to visualizing malignant breast tumors. The second post-MRI time point shows sufficient contrast enhancement in the tumor for at least 4 hours postend of infusion; this will allow for extensive MRI scanning, and high-resolution protocols should they be indicated. A contrast enhancement in the liver and pancreas is common for manganese-based contrast agents. 22 Pegfosimer manganese showed this enhancement of the liver and pancreas, with clear delineation of several anatomical details such as the pancreatic ducts. A deeper understanding of the clinical application of pegfosimer manganese in enhancing the liver, and pancreas will require further consideration. As the proposed EPR mechanism of action is not unique to breast tumors, and given the level of contrast enhancement shown in the study, this may indicate broader applicability in the diagnosis and characterization of other solid tumor types. 23 , 24 This would need to be strengthened by performing larger, statistically powered clinical studies in the future. Additional studies may also include imaging of lymph nodes, where the mechanism of action is compatible with enhancement of metastases. Moreover, there may be potential for the product to be used for early detection and monitoring of tumor response to therapy. Additionally, there may also be further potential for using pegfosimer manganese as a screening tool to predict the prospects of success for other drugs relying on the EPR effect. As a phase 1 study, the primary objective was to ensure safety. Although Mn-based contrast agents have previously been successfully used for MRI, exposure to excess Mn could be a safety risk. Careful blood-sampling and elementary analysis for Mn was therefore incorporated in the current study to allow PK-modeling in support of any safety observations that could potentially be attributed to Mn. The present study applied the Linear-up/Logarithmic-down model to analyze plasma Mn PK, aligning with International Conference on Harmonization/GCP Guidelines for Statistics in Clinical Trials. The model accounts for the infusion phase (linear trapezoidal method for increasing concentrations) and clearance phase (logarithmic calculation for blood clearance). However, this approach may not optimally model pegfosimer manganese due to nanoparticle-related factors such as particle size distribution, and renal clearance thus may lead to inaccuracies in AUC calculations. The model specified in the clinical study protocol was chosen based on best assumptions for support of any safety signal potentially attributed to Mn at the time of study start. It does not however comprehensively describe PK for the polymeric carrier particles of pegfosimer manganese. This was explored in a later, separate exploratory study 25 published before completion of this manuscript. This subsequent, exploratory study, using the surrogate analyte Si in the same blood samples, developed in-house/non-GxP, nontraditional PK model to address these challenges. Unlike the Lin-up/Log-down analysis showing a 7-minute initial half-life for plasma Mn, the custom model suggested a 42-minute half-life for Si, reflecting the postrenal filtrated portion of pegfosimer manganese rather than maximum blood Mn concentration. While the results from Mn and Si analyses are not directly comparable due to the different models, the results are indicative of Mn-ion dissociation from the carrier particles. In the paper by Axelsson et al, it was considered that this is mainly due to competitive binding to albumin. The clinical results shown here suggest that this process is not instantaneous and that there is sufficient association for delivery to the tumors through the remaining manganese-carrying particles leading to clinically relevant tumor contrast with minimal background enhancement. Future steps include applying the custom model to the plasma Mn-data, which would account for nanoparticle-specific behavior, to be able to correlate Mn with Si to address potential Mn-dissociation post-infusion. There were several limitations to the study. First, the subjective classification of whether an image was of diagnostic quality, as judged by a single radiologist. Applying central imaging training before the inclusion of image reading and agreement by several experienced individuals would be of value in follow-up studies to reduce radiologist reader bias. The study had few participants to conclude the proof of concept of the contrast enhancement. All individuals were White, and future studies should look to include a more diverse population. In addition, other solid tumor types should be investigated to understand the proof of concept and pegfosimer manganese's wider applicability. Lastly, further studies need to build on the small sample size in this study with larger, statistically powered investigations. Overall, the results of this study indicate that pegfosimer manganese has an acceptable safety profile providing sufficient contrast enhancement at a clinically relevant dose for assessment of appropriate MRI sequences in participants with primary breast tumors. Further investigation is warranted to fully explore the potential of pegfosimer manganese and its ability to meet the rising need for improved tumor characterization, diagnosis, and follow-up.