A Phase Ib First-In-Patient Study Assessing the Safety, Tolerability, Pharmacokinetics, and Pharmacodynamics of Ponsegromab in Participants with Cancer and Cachexia

This article is featured in Highlights of This Issue, p. 469

Patients with advanced cancer often develop debilitating cachexia characterized by unintended weight loss, lack of appetite, fatigue, muscle atrophy, impaired physical function, and increased risk of death (1, 2). It is thought that cachexia is mediated by imbalanced signaling of the metabolic and immune systems (1, 2). As a result, there are several potential therapeutic targets for the management of cancer-related cachexia (1, 2).

One promising target is cytokine growth differentiation factor 15 (GDF-15), a member of the TGFβ superfamily. GDF-15 is a global regulator of the body's response to stress, and markedly elevated GDF-15 concentrations have been reported in patients with various types of advanced cancer (3–6). Furthermore, elevated GDF-15 concentrations are associated with weight loss and poor outcomes in patients with certain types of cancer (6–9). In mouse tumor models GDF-15 promotes cachexia, while inhibition of GDF-15 reverses weight loss and improves survival (6, 10, 11).

In rodents and non-human primates, GDF-15 induces anorexia and weight loss by acting through the glial cell-derived neurotrophic factor family receptor alpha like (GFRAL; refs. 12–14). GFRAL is a transmembrane protein expressed, for the most part, in specific areas of the brainstem and requires the tyrosine kinase co-receptor RET for signaling (15). Signaling is mediated via dimers of circulating GDF-15 binding to a coreceptor complex composed of two GFRAL molecules and two RET molecules, leading to an activated pentameric receptor complex (15).

Ponsegromab (PF-06946860, Pfizer Inc) is a humanized mAb that is a highly selective and potent inhibitor of GDF-15 (6). Ponsegromab treatment has been shown to improve the cachectic phenotype and increase survival in a GDF-15 secreting mouse non–small cell lung cancer (NSCLC) tumor model (6). Previously, ponsegromab has been evaluated in two phase I single-dose studies in healthy participants (16, 17). Here, we report findings from a phase Ib, first-in-patient study of ponsegromab in participants with advanced cancer and cachexia. The primary objective of this study was to characterize the safety and tolerability of repeated subcutaneous administrations of ponsegromab in this patient population. Additional objectives included assessment of serum ponsegromab concentrations, serum GDF-15 concentrations, and exploratory measures of ponsegromab efficacy.

This study (NCT04299048) was a 24-week (12-week treatment and 12-week follow-up; Supplementary Fig. S1) phase Ib open-label clinical trial conducted at 13 sites in the United States from November 2020 to March 2022. The study protocol was approved by an Institutional Review Board or Independent Ethics Committee for each participating investigational center. All participants provided written informed consent. The study was conducted in compliance with ethical principles of the Declaration of Helsinki and all International Conference on Harmonization Good Clinical Practice Guidelines.

Key inclusion criteria included the following: age ≥18 years; a documented histologic or cytologic diagnosis (per American Joint Committee on Cancer criteria) of advanced metastatic NSCLC, advanced/unresectable pancreatic cancer, or metastatic colorectal cancer; cachexia as defined using a modified version of Fearon criteria [ref. 18; body mass index (BMI) <20 kg/m² with involuntary weight loss >2% within 6 months prior to screening, or involuntary weight loss >5% within 6 months prior to screening irrespective of BMI (patient's report was used to estimate involuntary body weight loss if medical record documentation was unavailable)]; elevated serum GDF-15 concentrations (≥1.5 ng/mL, assessed using the Elecsys® GDF-15 immunoassay; Roche Diagnostics; ref. 19); and entering the study at the first or second cycle of current course of standard-of-care anticancer therapy.

Standard-of-care therapy for enrolled participants, per the study protocol, included a platinum + pemetrexed ± pembrolizumab, a platinum + nab-paclitaxel or paclitaxel ± pembrolizumab, or pembrolizumab alone (for advanced NSCLC); folfirinox, nab-paclitaxel + gemcitabine, or gemcitabine alone (for advanced/unresectable pancreatic cancer); and folfox ± biologic (i.e., bevacizumab or cetuximab/panitumumab), folfiri ± biologic, folfoxiri ± biologic, or pembrolizumab (for metastatic colorectal cancer).

Key exclusion criteria included the following: other forms of cancer unless considered cured (>5 years without evidence of recurrence); cachexia caused by other reasons (e.g., severe chronic obstructive pulmonary disease, New York Heart Association class III-IV heart failure, acquired immunodeficiency syndrome); known symptomatic brain metastases requiring steroids (asymptomatic or previously diagnosed brain metastases were allowed if participants completed treatment and recovered from acute effects of radiotherapy or surgery); planned antitumor radiotherapy; positive test for human immunodeficiency virus; active reversible causes of decreased food intake [e.g., National Cancer Institute Common Terminology Criteria for Adverse Events (CTCAE) grade 3 or 4 oral mucositis or mechanical obstructions making the participant unable to eat]; receiving tube feedings or parenteral nutrition; severe gastrointestinal disease; history of gastrectomy; and pregnancy or breastfeeding at the time of screening.

Participants received open-label 200 mg ponsegromab, administered subcutaneously by site staff, every 3 weeks (i.e., every 21 days) for 12 weeks (five doses total). Participants continued to receive their standard-of-care antitumor therapy (as described in the inclusion criteria above) and could receive supportive standard-of-care interventions as needed.

Safety assessments included monitoring for injection site reactions, physical examinations, vital sign measurements, 12-lead ECG, clinical laboratory assessments, and treatment-emergent adverse event (TEAE) monitoring. TEAEs were summarized descriptively, with severity (CTCAE grade 1, 2, 3, 4, or 5) and relationship to ponsegromab treatment determined by site investigators. TEAEs were coded using Medical Dictionary for Regulatory Activities (MedDRA) v24.1.

Blood samples were obtained at day 1 (baseline) and at weeks 1, 2, 3, 4, 5, 6, 9, 12, 13, 14, 15, 18, and 24 (or early termination) for assessment of serum ponsegromab and GDF-15 concentrations. In addition, blood samples were obtained at day 1 (baseline) and at weeks 2, 3, 6, 9, 12, 15, 18, and 24 (or early termination) to evaluate the presence of anti-drug antibodies (ADA) and neutralizing antibodies (NAb; if necessary). At dosing visits (day 1 and weeks 3, 6, 9, and 12), blood samples were collected prior to ponsegromab administration. In addition, at day 1 and week 12, postdose blood samples (≥3 hours after ponsegromab administration) were collected for serum ponsegromab and GDF-15 analysis. In contrast to GDF-15 assessments at screening, which used Elecsys® GDF-15 immunoassay from Roche Diagnostics, assessments of unbound and total GDF-15 during the baseline, treatment, and follow-up periods used assays developed by Pfizer Inc. See the Supplementary Methods for more detail on the ponsegromab, GDF-15, and ADA assays utilized.

Body weight was measured as an exploratory endpoint on day 1 (baseline; prior to ponsegromab administration) and at weeks 3, 6, 9, 12, 15, 18, and 24. Because Fearon's definition of cachexia is based on body weight, a gain in body weight would constitute reversal of cachexia. In addition, an association between body weight gain and survival has been shown in patients with cancer (20). Body weight was recorded using a calibrated scale with accuracy to the nearest 0.2 lb (or 0.1 kg) set on a flat surface and under the following conditions: (i) after void of urine; (ii) after removal of shoes, bulky layers of clothing, and jackets; and (iii) while remaining still during the measurement. CT scans of the chest, abdomen, and pelvis were performed at screening (baseline), at weeks 6 and 12 (or early termination) to monitor tumor burden following standard RECIST 1.1 criteria (21), and to evaluate (by a trained central reader) Lumbar Skeletal Muscle Index (LSMI). See Supplementary Methods for more information on LSMI assessment.

Physical activity and sleep were monitored via accelerometry [1 week of monitoring via wearable digital sensors at the lumbar region (during the day) and at the nondominant wrist (continuously)] during screening (baseline) and at weeks 6 and 12. Weekly mean values were calculated for sedentary, light, and moderate activity times (per day), mean activity level during the maximum 6 minutes of daily activity, total sleep time (per day), and gait speed.

The Functional Assessment of Anorexia-Cachexia Therapy (FAACT; ref. 22) was completed by participants on day 1 (baseline) and at weeks 6, 12, and 18. The FAACT is a combination of (i) the 27-item Functional Assessment of Cancer Therapy-General (FACT-G) core instrument that assesses physical well-being, emotional well-being, functional well-being, and social well-being in the past 7 days, and (ii) a 12-item anorexia and cachexia subscale (ACS) that assesses patient perception of appetite and weight. Five items within the ACS comprise an anorexia-related symptoms scale (5IASS). Individual FACT-G and ACS items are rated on a scale of 0–4. As a result, 5IASS, ACS, FACT-G, and FAACT total scores range 0–20, 0–48, 0–108, and 0–156, respectively, with higher scores indicating a higher health-related quality of life. Typically, a score ≤ 37 on the ACS is indicative of anorexia in patients with cancer (23).

A sample size of approximately 8 participants (assuming a minimum of 6 participants would complete the study) was chosen on the basis of the need to minimize first exposure of a new chemical entity to patients with advanced cancer and cachexia and the requirement to provide adequate safety, tolerability, and pharmacokinetic/pharmacodynamic assessments.

A mixed-model repeated-measures analysis was used to analyze change from baseline body weight at all postdose timepoints. The model included participant as a random term and baseline, time, and baseline-by-time interaction as fixed terms. A compound symmetry covariance matrix was fitted to the repeated times within participants. Baseline weight was defined as the last predose measurement on day 1. The least squares (LS) mean change at week 12 was compared with an external historical control [normally distributed with a mean (SE) change of –0.57 (0.51) kg] derived from a meta-analysis of historical data in 543 placebo-treated patients with cancer cachexia. The meta-analysis included three sources: non-Pfizer placebo controlled clinical trials in patients with cancer cachexia (NSCLC n = 4; NSCLC/colorectal n = 1); a subset of patients with NSCLC and > 5% body weight loss in Pfizer-sponsored cancer trials (n = 3); and a subset of patients with lung cancer with cachexia identified in the Optum database. The effective n of the historical control was 120, representing the number of study participants that the historical control is equivalent to and derived using the standard errors of both the external historical control and the ponsegromab LS mean.

All other efficacy, safety, pharmacokinetic, GDF-15, and immunogenicity endpoints were summarized without additional statistical analysis and were based on observed data only (i.e., no imputation for missing data).

Upon request, and subject to review, Pfizer will provide the data that support the findings of this study. Pursuant to certain criteria, conditions, and exceptions, Pfizer may also provide access to the related individual de-identified participant data. See https://www.pfizer.com/science/clinical-trials/trial-data-and-results for more information. Qualified researchers’ proposals are submitted via the Vivli platform at https://vivli.org/members/enquiries-about-studies-not-listed-on-the-vivli-platform/.

A total of 20 participants were screened; 11 were enrolled in the study, and 10 received at least one dose of study treatment (Supplementary Fig. S2). Of the 10 participants receiving treatment, 8 (80.0%) completed treatment (2 participants discontinued due to global deterioration of health status). Of the 8 that entered the follow-up period, 7 (70.0%) completed the study (one participant died during follow-up). Participant demographics and clinical characteristics are shown in Table 1. A majority of participants were White (70%) and male (70%). Median (range) age was 69.5 (47–81) years, and median (range) BMI was 22.8 (17.3–30.9) kg/m². Primary cancer types were colorectal (n = 4; 40.0%), NSCLC (n = 3; 30.0%), and pancreatic (n = 3; 30.0%). Supplementary Table S1 summarizes the representativeness of the study population in relation to cancer cachexia.

All participants had received medications in the 28 days prior to the first dose of study treatment, with ondansetron (n = 5; 50%), dexamethasone (n = 4; 40%), and atorvastatin, lorazepam, pantoprazole, and prochlorperazine (n = 3; 30% each) being reported most frequently. All participants also received concomitant medications, with ondansetron and pantoprazole (n = 6; 60% each), dexamethasone (n = 5; 50%), and lorazepam, prochlorperazine, and sodium chloride (n = 4; 40% each) being reported most frequently.

No TEAEs related to ponsegromab treatment or injection site reactions were reported. Likewise, no adverse trends in clinical laboratory tests, vital signs, or ECG parameters attributable to ponsegromab dosing were evident.

A total of 92 all-causality TEAEs, all deemed unrelated to ponsegromab treatment, were reported among the 10 (100.0%) participants (Table 2). Among these TEAEs unrelated to ponsegromab treatment, the most common were anemia (n = 5; 50%), diarrhea (n = 5; 50%), nausea (n = 4; 40%), and fatigue (n = 4; 40%). A majority of these TEAEs were mild (grade 1; n = 54; 58.7%) or moderate (grade 2; n = 26; 28.3%) in intensity. Overall, TEAEs of maximum grade 1, 2, 3, 4, and 5 were reported in 0 (0%), 4 (40.0%), 3 (30.0%), 2 (20.0%), and 1 (10.0%) participant(s), respectively. Five serious TEAEs (unrelated to ponsegromab treatment) were reported in 4 (40%) participants. One participant experienced acute cholecystitis (grade 4 intensity) on day 156 (during the follow-up period) that resolved on day 160. One participant experienced sepsis (grade 3 intensity) on day 12 that resulted in permanent discontinuation from study treatment. One participant experienced an increase in transaminases (began on day 22; resolved with sequelae on day 25; resulted in temporary discontinuation of study treatment) and anemia (began on day 39; resolved on day 45; resulted in permanent discontinuation from study treatment). Finally, one participant experienced neoplasm progression on day 131 (during the follow-up period) that resulted in death.

Median unbound and total ponsegromab concentrations over time are shown in Fig. 1. Unbound ponsegromab concentration was detected as early as 3 hours after the first dose on day 1, with a median concentration of 86.3 ng/mL, and ranged from 2,169–16,780 ng/mL from weeks 1–15 with large inter-individual variability observed across all visits. Median unbound ponsegromab C_trough ranged from 2,169–6,898 ng/mL across the dosing period (weeks 3–15; Supplementary Table S2). Median total ponsegromab concentrations were higher than unbound ponsegromab concentrations at all-time points; 228 ng/mL at 3 hours postdose on day 1 and week 1–24 values ranging from 4,352–32,760 ng/mL. Median total ponsegromab C_trough values ranged from 10,080–26,270 ng/mL from weeks 3 to 15.

Median (range) serum unbound GDF-15 concentration was 2.269 (0.915–3.670) ng/mL at baseline (assessed prior to dosing on day 1). Following subcutaneous administration of 200 mg ponsegromab every 3 weeks, median unbound GDF-15 concentrations were below the lower limit of quantification (0.0424 ng/mL) from 3 hours post-dose on day 1 to week 15 (3 weeks after final dose), returned to approximately baseline concentration by week 18 (6 weeks after final dose), and was slightly above baseline concentration on week 24 (12 weeks after final dose; Fig. 2A). Median (range) serum total GDF-15 concentration was 4.260 (2.24–8.91) ng/mL at baseline and increased over time following ponsegromab administration (from day 1 to week 15). The maximal observed median total GDF-15 concentration was 3,230 ng/mL at week 15 (Fig. 2B).

All ponsegromab-treated participants were ADA negative at baseline (n = 10), and all remained ADA negative after receiving up to five subcutaneous doses of ponsegromab every 3 weeks (n = 9).

Mean increases from baseline in body weight were observed at all time points during the treatment (weeks 3, 6, 9, and 12) and follow-up (weeks 15, 18, and 24) periods (Fig. 3). Mean (SD) body weight at baseline (assessed prior to dosing on day 1) was 70.49 (16.97) kg, and an LS mean (SE) increase of 4.63 (1.98) kg was observed at week 12, representing an increase of approximately 6.6% relative to baseline. On the basis of a mean change in body weight of –0.57 kg for the external historical control, the adjusted LS mean (90% CI) increase from baseline in body weight among subcutaneous 200 mg ponsegromab every 3 weeks–treated participants was 5.20 (1.67–8.73) kg at week 12 (one-sided P = 0.01). Maximal increase in body weight among ponsegromab-treated participants occurred at week 15; LS mean (SE) increase of 5.89 (1.98) kg corresponded to an increase of approximately 8.4% relative to baseline. In addition to increases in body weight, mean (SD) increases from baseline in LSMI were observed at week 6 [1.30 (2.97) cm²/m²] and at week 12 [0.44 (2.95) cm²/m²]. Mean (SD) LSMI at baseline was 45.17 (8.80) cm²/m².

Mean (SD) FAACT 5IASS, ACS, FACT-G total, and FAACT total scores at baseline were 11.9 (4.7), 29.3 (9.3), 72.0 (14.6), and 101.3 (19.6), respectively. Increases in 5IASS, ACS, FACT-G total, and FAACT total scores over baseline were observed during the treatment period (weeks 6 and 12), with maximal increases occurring at Week 12 (Table 3). At week 12, mean (SD) change from baseline was 4.5 (3.0) for 5IASS, 9.0 (4.3) for ACS, 9.0 (11.1) for FACT-G total, and 18.0 (13.3) for FAACT total score.

Mean (SD) sedentary activity time at baseline was 707.7 (191.4) min/day and accounted for 49.7% (13.7%) of the total day. Mean reductions from baseline in sedentary activity were observed at weeks 6 and 12, with maximal mean (SD) reductions of –146.4 (66.5) minutes/day and –10.2% (4.2%) of the total day at week 12 (Supplementary Fig. S3A). Mean (SD) light activity time at baseline was 44.28 (19.48) minutes/day and accounted for 3.11% (1.37%) of the day. Mean reductions from baseline in light activity of –5.65 (18.94) minutes/day and –0.32% (1.37%) of the total day were observed at week 6, followed by mean increases of 3.85 (11.89) minutes/day and 0.27% (0.83%) of the total day at week 12 (Supplementary Fig. S3B). Mean (SD) moderate activity time at baseline was 20.68 (12.26) minutes/day and accounted for 1.45% (0.86%) of the total day. Mean increases from baseline in moderate activity were observed at weeks 6 and 12, with maximal mean (SD) increases of 14.01 (12.33) minutes/day and 0.97% (0.86%) of the total day at week 12 (Supplementary Fig. S3C). At baseline, the mean (SD) average activity level during the maximum 6 minutes of daily activity was 388.3 (132.5). Mean increases from baseline were observed at weeks 6 and 12, with a maximal increase of 73.1 (63.9) at week 12 (Supplementary Fig. S3D).

Mean (SD) total sleep time at baseline was 357.3 (120.8) minutes/day and accounted for 25.1% (8.5%) of the day. Mean increases from baseline in total sleep time were observed at weeks 6 and 12, with a maximal increase of 57.4 (223.9) minutes/day and 4.1% (15.9%) of the total day at week 12 (Supplementary Fig. S3E). Mean (SD) gait speed at baseline was 0.782 (0.091) m/s. Mean increases from baseline in gait speed were observed at weeks 6 and 12 during the treatment, with a maximal increase of 0.049 (0.028) m/s at week 12 (Supplementary Fig. S3F). Mean (SD) 95th percentile of gait speed at baseline was 0.976 (0.131) m/s. Mean increases from baseline in 95th percentile were observed at weeks 6 and 12, with a maximal increase of 0.080 (0.041) m/s at week 12.

To assess how participants responded to their standard-of-care antitumor therapy during the study, tumor burden was characterized as per RECIST 1.1 criteria. Of the 8 participants evaluated at week 6, 3 (37.5%) had a partial response to treatment and 5 (62.5%) had stable disease. At week 12, 2 (25%) had a partial response to treatment, 3 (37.5%) had stable disease, and 3 (37.5%) had progressive disease.

Overall, this first-in-patient study demonstrates that repeated (every 3 weeks) subcutaneous administration of 200 mg ponsegromab was generally safe and well tolerated in participants with cachexia and advanced NSCLC, pancreatic cancer, or colorectal cancer. This conclusion is based on the observation that no TEAEs were deemed related to ponsegromab treatment by site investigators; no injection site reactions were reported; there were no apparent adverse trends in safety laboratory tests, vital signs, or ECG parameters attributed to ponsegromab dosing; and no participants were positive for ADA during the study. The study also demonstrated that median unbound and total serum ponsegromab concentrations were detected soon (3 hours) after administration of the first subcutaneous dose, increased substantially over the first week, and were maintained throughout the treatment period. Median unbound ponsegromab concentrations were maintained above 2,169 ng/mL through week 15. This increase in serum unbound ponsegromab was associated with a decrease in median serum unbound GDF-15 concentrations to below the limit of detection over the course of the treatment period.

Participants were required to have elevated serum concentrations of GDF-15, based on the Elecsys® GDF-15 assay from Roche Diagnostics, prior to study entry; typical serum concentrations in healthy individuals range from approximately 0.4–1.1 ng/mL depending on age (24). Median unbound serum GDF-15 concentration at baseline was 2.269 ng/mL, based on the assay developed by Pfizer Inc. Following initiation of treatment with subcutaneous ponsegromab every three weeks, median unbound serum GDF-15 concentrations were immediately suppressed to below the lower limit of quantification (< 0.042 ng/mL) from day 1 (3 hours after initial dose) through week 15 (3 weeks after cessation of treatment). In contrast, total GDF-15 levels increased over time following treatment with subcutaneous 200 mg ponsegromab every three weeks and reached maximal level at week 15 (3 weeks after the final dose). At weeks 18 and 24, median unbound ponsegromab was below the limit of detection. Median unbound serum GDF-15 concentration was at approximately baseline level at week 18 and was slightly above baseline at week 24 (approximate 69.20% change from baseline), whereas median total GDF-15 remained considerably elevated from baseline (approximately 78,900% and 22,900% change from baseline at weeks 18 and 24, respectively) and tracked with the total ponsegromab concentration (Supplementary Fig. S4). Following ponsegromab dosing, the immediate suppression of unbound GDF-15 and rise in total GDF-15 observed are consistent with the expected mechanism of action for ponsegromab, which binds unbound GDF-15 with high affinity and neutralizes the ability of GDF-15 to signal through GFRAL (6). GDF-15 is a dimer with two binding sites for ponsegromab (one binding site/subunit). Three species of GDF-15 may exist depending on the relative amount of GDF-15 versus ponsegromab present: fully bound GDF-15 (both binding sites are ponsegromab-bound), partially bound GDF-15 (one binding site occupied by ponsegromab and the other site remains unbound), and unbound GDF-15 (both binding sites remain unbound). The total GDF-15 assay was designed to measure all species of GDF-15, regardless of ponsegromab binding status by using a polyclonal anti-GDF-15 antibody for immunoprecipitation prior to LC/MS-MS detection. The unbound GDF-15 assay could potentially detect both unbound GDF-15 and partially bound GDF-15; therefore, a sample pretreatment step was used to remove ponsegromab-bound GDF-15 (both fully and partially bound) as described in the Methods. However, the efficiency of this pretreatment step is not 100% and a residual amount of ponsegromab-bound GDF-15 may remain (unpublished data). At weeks 18 and 24, unbound ponsegromab concentration was below the limit of quantitation while total GDF-15 was still considerably elevated from baseline (Fig. 2B). As such, the observed apparent increase in unbound GDF-15 above baseline level at week 24 likely is due to nonspecific detection of the partially bound GDF-15. However, it is possible that the detected signal could be a combination of unbound and partially bound GDF-15 given the assay format.

Suppression of unbound GDF-15 was accompanied by preliminary exploratory evidence of increased appetite and weight gain in study participants. LS mean increases in bodyweight from baseline exceeded 2.5% at all assessments from weeks 6 to 24 and exceeded 5% at all assessments from weeks 9 to 18 (week 18 representing 6 weeks after last dose of ponsegromab). At the end of treatment (week 12), the mean change in body weight was approximately 6.6% above baseline. Compared with a loss of 0.57 kg in the external historical control, derived from a meta-analysis of historical data in placebo-treated cancer cachexia populations, the adjusted LS mean increase in body weight among ponsegromab-treated participants at week 12 was 5.20 kg (7.4%; 1-sided P = 0.01). In patients with NSCLC who did not respond to standard-of-care anti-neoplastic treatments, an increase in body weight was associated with reduced risk of disease progression and death, independent of tumor response (20). The survival benefit was comparable for weight gains >0%, >2.5%, and >5% over baseline (20). Thus, ponsegromab treatment was associated with a potentially clinically meaningful weight gain during the study. Modest increases in LSMI were also observed during the study. Mean (SD) LSMI at baseline was 45.17 (8.80) cm²/m². Cancer cachexia may be defined as an LSMI of <55 cm²/m² in males and of <39 cm²/m² in females (18).Because the study population was 70% male, this baseline LSMI value is indicative of a cachectic phenotype in a majority of study participants.

The progressive loss of skeletal muscle associated with cancer cachexia results in diminished physical function, which directly impacts quality of life (1, 2). In this study, improved physical activity was generally observed, on average, using both wrist and lumbar digital sensors during the treatment period. Overall, participants demonstrated reduced sedentary time, increased activity levels, faster gait, and increased maximum gait speed. Specifically, improvements in physical activity at week 12 included a nearly 2.5-hour mean decrease in the amount of time participants were sedentary, a 14-minute mean increase in the amount of time participants participated in moderate activity, and a 0.049 m/s mean increase in participant gait speed (a small-to-medium, but clinically meaningful, improvement; refs. 25, 26). This increase in gait speed was accompanied by a mean increase of 0.080 (0.041) m/s in maximum gait speed (e.g., the 95th percentile of gait speed) at week 12. Finally, a nearly 1-hour mean increase in total sleep time was also observed at week 12.

In addition to objective improvements in body weight and physical activity, subjective improvements in symptoms and concerns related to anorexia were observed during the treatment period, based on changes in FAACT ACS score. Mean ACS score at baseline (29.3) indicated an anorexic study population. A mean improvement in ACS score of 9 points was observed at week 12 among ponsegromab-treated participants.

These exploratory efficacy findings are consistent with the role of elevated GDF-15 in appetite loss and cachexia among participants with advanced cancer, with inhibition of GDF-15 improving appetite and attenuating weight loss among this population. These conclusions, however, are limited by the small study sample size. Efficacy findings are also limited by the open-label design, which may introduce bias into subjective evaluations, such as patient-reported outcomes, because both participants and clinical staff knew participants had been treated with the investigational drug. The small sample size and open-label design were based on the need to minimize first exposure of patients with advanced cancer and cachexia to this new chemical entity and to transparently establish an initial assessment of clinical safety, pharmacokinetics, pharmacodynamics, and immunogenicity. Finally, the lack of an internal placebo control was a limitation and necessitated the use of an external historical control (derived from historical data in placebo-treated cancer cachexia populations considered similar to this phase Ib study population) for the analysis of body weight. This analysis was conducted to provide additional context around the body weight increases observed in this study by adjusting for unintended weight loss that could, potentially, occur in participants not receiving investigational drug.

Overall, the primary (assessment of ponsegromab safety and tolerability) and secondary (assessment of serum ponsegromab concentrations) study objectives were achieved, and there was preliminary, exploratory evidence of ponsegromab efficacy (weight gain along with increased appetite and activity) in participants with advanced cancer and cachexia. These findings support the continued development of ponsegromab for the treatment of cachexia and are the basis for expanded testing in larger patient populations to confirm safety/tolerability findings and more thoroughly assess efficacy. A phase II study (NCT05546476) is currently being conducted to confirm the findings of this phase Ib study.

J. Crawford reports grants from Pfizer during the conduct of the study, as well as other support from Pfizer and personal fees from Actimed outside the submitted work. R.A. Calle reports Pfizer employment at the time of study execution; in addition, R.A. Calle is currently affiliated with Regeneron Pharmaceuticals, and is a Regeneron and Pfizer shareholder. S.M. Collins reports personal fees from Pfizer during the conduct of the study. Y. Weng reports being a full-time employee and shareholder of Pfizer. S.L. Lubaczewski reports other support from Pfizer during the conduct of the study. C. Buckeridge reports personal fees from Pfizer Inc during the conduct of the study. E.Q. Wang is a full-time employee of Pfizer Inc., and shareholder of Pfizer stock. M.A. Harrington reports personal fees and other support from Pfizer during the conduct of the study. M.I. Rossulek reports other support from Pfizer Inc outside the submitted work, and is an employee of Pfizer Inc. J.H. Revkin reports other support from Pfizer Inc during the conduct of the study. No other disclosures were reported.

J. Crawford: Investigation, writing–review and editing. R.A. Calle: Conceptualization, methodology, writing–review and editing. S.M. Collins: Formal analysis, writing–review and editing. Y. Weng: Formal analysis, writing–review and editing. S.L. Lubaczewski: Conceptualization, methodology, writing–review and editing. C. Buckeridge: Conceptualization, methodology, writing–review and editing. E.Q. Wang: Conceptualization, formal analysis, methodology, writing–review and editing. M.A. Harrington: Conceptualization, writing–review and editing. A. Tarachandani: Conceptualization, writing–review and editing. M.I. Rossulek: Conceptualization, supervision, methodology, writing–review and editing. J.H. Revkin: Conceptualization, supervision, methodology, writing–review and editing.

The authors would like to thank the patients, families, and research sites and their teams who participated in the trial. The authors would also like to thank the principal investigators (Sunil Babu, Alicia Swink Swin, Karen Russell, Karng Log, Arash Gabayan, and Jose Garcia) of the study who recruited participants that were randomized and received treatment. The authors would also like to thank Meghana Deshpande, Frederick McCush, Parya Nouri, Lindsey Ewan King, Matthew Blatnik, Katherine Wright, Marcela Araya, and Ying Wang for their support on ponsegromab, GDF-15, and ADA/Nab bioanalysis, as well as Hui Ding for assistance with data generation (synthesis of data tables). This study was sponsored by Pfizer Inc. Medical writing support was provided by Matt Soulsby, PhD, CMPP, of Engage Scientific and was funded by Pfizer.

The publication costs of this article were defrayed in part by the payment of publication fees. Therefore, and solely to indicate this fact, this article is hereby marked “advertisement” in accordance with 18 USC section 1734.