Pimicotinib versus placebo for tenosynovial giant cell tumour (MANEUVER): an international, randomised, placebo-controlled, phase 3 trial.

# Pimicotinib versus placebo for tenosynovial giant cell tumour (MANEUVER): an international, randomised, placebo-controlled, phase 3 trial *The Lancet 2026* Articles Pimicotinib versus placebo for tenosynovial giant cell tumour (MANEUVER): an international, randomised, placebo-controlled, phase 3 trial Hairong Xu, Xiaohui Niu, Vinod Ravi, Javier Martin-Broto, Albiruni Abdul Razak, Ramy Saleh, Yong Zhou, Jingnan Shen, Tang Liu, Kamlesh Kumar Sankhala, César Serrano, Silvia Stacchiotti, Jing Wang, Giacomo G Baldi, Yi Feng, Yingqi Hua, Tao Li, Piotr Rutkowski, Xiaojing Zhang, Gabriel Tinoco, Qingping Zou, Boyao Shan, Xiangyu Zhu, Hans Gelderblom Summary Lancet 2026; 407: 1072–83 Background Tenosynovial giant cell tumour (TGCT) is a rare, locally aggressive neoplasm that aects otherwise Published Online healthy adults. There are few systemic treatment options, highlighting an unmet need. We report the results of part 1 March 5, 2026 of the MANEUVER trial, which aimed to evaluate the ecacy and safety of pimicotinib, a highly selective, potent, https://doi.org/10.1016/ colony-stimulating factor-1 receptor inhibitor, in patients with TGCT. S0140-6736(25)02602-9 See Comment page 1032 Methods MANEUVER is a randomised, placebo-controlled, phase 3 study done in 40 specialised hospitals in Asia, For the plain language summary Europe, and North America. Patients aged 18 years and older with unresectable, symptomatic TGCT (patient-reported see Online for appendix 1 worse stiness or worst pain of at least 4 on a scale of 0–10) were randomly assigned (2:1, double-blind) to oral, once- Department of Orthopedic daily pimicotinib 50 mg or placebo for 24 weeks (part 1). An independent statistician used a central interactive web Oncology Surgery, Beijing Jishuitan Hospital, Capital response system to generate the randomisation schedule; stratification was by region (China vs non-China). Masking Medical University, Beijing, was achieved by using placebo identical in appearance to pimicotinib. In part 1, patients, all investigators, and study China (Prof H Xu MD, funders were masked to treatment assignments. All patients who completed part 1 were allowed to continue to open- Prof X Niu MD); The University label part 2: pimicotinib-treated patients could continue the same dosage and placebo-treated patients could cross of Texas MD Anderson Cancer Center, Houston, TX, USA over to receive pimicotinib for 24 weeks. Eligible patients who completed part 2 were allowed to continue once-daily (Prof V Ravi MD); Medical pimicotinib long-term in part 3. The primary endpoint was objective response rate (ORR) at week 25 by blinded Oncology Department, independent review committee per Response Evaluation Criteria in Solid Tumors version 1.1 in the intention-to-treat Fundación Jiménez Diaz population (all randomised patients). Safety was analysed in patients who received at least one dose of study drug. University Hospital; University Hospital General de Villalba, Missing data were not imputed and only observed data were analysed. Trial enrolment is complete; the study is and Instituto de Investigación registered at ClinicalTrials.gov (NCT05804045) and is ongoing. Sanitaria Fundacion Jiménez Diaz, Madrid, Spain Findings Between April 27, 2023, and March 29, 2024, 126 patients were screened and 94 patients (China [n=45], non- (J Martin-Broto MD); Princess Margaret Cancer Center, China [n=49]) were randomly assigned to, and received, pimicotinib (n=63) or placebo (n=31). 30 (32%) patients were Toronto, ON, Canada male and 64 (68%) were female. ORR at week 25 was 54% (34 of 63) in the pimicotinib group and 3% (one of 31) in the (A A Razak MD); Cedars Cancer placebo group (absolute dierence 51% [95% CI 33–63], p<0·0001). Pimicotinib was associated with mainly mild Center, McGill University Health Center, Montreal, QC, treatment-emergent adverse events, including mostly manageable asymptomatic laboratory abnormalities and clinical Canada (Prof R Saleh MD); West events, such as pruritus, facial oedema, rash, periorbital oedema, and fatigue. The only grade 3 or 4 treatment-emergent China Hospital, Sichuan adverse event occurring in more than 10% of pimicotinib-treated patients was increase in blood creatine phosphokinase, University, Chengdu, China in eight (13%) of 63 patients. The most common treatment-emergent adverse events in the placebo group were fatigue (Y Zhou MD); The First Affiliated Hospital of Sun Yat- and arthralgia. Dose reductions occurred in five (8%) of 63 pimicotinib-treated patients and treatment discontinuations sen University, Guangzhou, in one (2%) of 63 pimicotinib-treated patients. There was no cholestatic hepatotoxicity, drug-induced liver injury, or China (Prof J Shen PhD); The hypopigmentation of skin or hair. Second Xiangya Hospital of Central South University, Changsha, China Interpretation Pimicotinib showed robust antitumour activity with clinically meaningful improvements in TGCT- (Prof T Liu MD); Precision related functional limitations and symptom burden, oering an eective treatment option with a manageable safety NextGen Oncology and profile for this underserved condition. Research Center, Los Angeles, CA, USA (K K Sankhala MD); Vall d’Hebron Institute of Funding Abbisko Therapeutics. Oncology, Barcelona, Spain (C Serrano MD); Fondazione Copyright © 2026 The Author(s). Published by Elsevier Ltd. This is an Open Access article under the CC BY-NC-ND IRCCS Istituto Nazionale dei 4.0 license. Tumori, Milan, Italy (S Stacchiotti MD); Hunan Provincial People’s Hospital, Introduction villonodular synovitis, TGCT is a rare disease with an Changsha, China (J Wang MD); Tenosynovial giant cell tumour (TGCT) is a locally estimated annual global incidence of 43 cases per Hospital of Prato, Azienda USL aggressive, soft tissue tumour originating in the joint 1 million person-years.1,2 TGCT is driven by a small Toscana Centro, Prato, Italy (G G Baldi MD); The Second synovium, bursae, or tendon sheaths. Previously known number of neoplastic cells overexpressing growth factor Hospital of Shanxi Medical as giant cell tumour of the tendon sheath or pigmented colony-stimulating factor-1 (CSF-1), due to genetic 1072 Articles University, Taiyuan, China Research in context (Y Feng MD); Shanghai General Hospital, Shanghai, China Evidence before this study in China, Europe, and North America, pimicotinib demonstrated (Y Hua MD); Zhejiang Cancer For patients with unresectable tenosynovial giant cell tumour robust tumour shrinkage, early and durable symptomatic benefit, Hospital, Hangzhou, China (TGCT), systemic treatment might be the only suitable and improved quality of life for patients. These improvements (T Li MD); Maria Sklodowska- Curie National Research treatment modality. Globally, these patients have few were also seen in prespecified subgroups, including age, sex, Institute of Oncology Warsaw, treatment options, emphasising a large unmet need for more Eastern Cooperative Oncology Group performance status, Warsaw, Poland effective and tolerable systemic treatments. Less than half of geographical region (China vs non-China), disease location, and (Prof P Rutkowski MD); Liaoning patients treated with pexidartinib in the ENLIVEN trial or number of previous surgeries, suggesting that pimicotinib Cancer Hospital and Institute, Shenyang, China vimseltinib in the MOTION trial had an objective tumour delivers broad efficacy across diverse populations. Pimicotinib (X Zhang PhD); The Ohio State response (per Response Evaluation Criteria in Solid Tumours had a manageable safety profile and was well tolerated, with low University Comprehensive version 1.1). Both trials almost entirely enrolled patients from incidences of dose reduction and treatment discontinuation, and Cancer Center, Columbus, OH, the USA, Canada, Europe, and Australia; pexidartinib is there was no evidence of drug-induced liver injury, cholestatic USA (G Tinoco MD); Abbisko Therapeutics, Shanghai, China registered in the USA, Taiwan, and Korea, and vimseltinib is hepatotoxicity, or skin or hair hypopigmentation. (Q Zou PhD, B Shan MA, registered in the USA and Europe. In addition, pexidartinib is a X Zhu BS); Leiden University Implications of all the available evidence multikinase inhibitor with significant risk for serious cholestatic Medical Center, Leiden, Pimicotinib expands treatment options in TGCT and provides an hepatotoxicity or mixed liver injury, or both. We searched Netherlands additional, targeted, systemic treatment for patients who are not (Prof H Gelderblom MD) PubMed and Embase for original articles and reviews published amenable to surgery. In MANEUVER, pimicotinib as a once-daily, Correspondence to: between Jan 1, 2019, and June 17, 2024, using key terms oral treatment showed early and robust efficacy and safety in Prof Xiaohui Niu, Department of (“tenosynovial giant cell tumours”, “TGCT”, “villonodular patients with TGCT. The diverse patient population Orthopedic Oncology Surgery, synovitis”, “CSF-1R”). There were few clinical trials using growth Beijing Ji Shui Tan Hospital, (approximately equal China vs non-China sites) suggests that the factor colony-stimulating factor-1 receptor (CSF-1R) inhibitors Capital Medical University, study findings would be broadly generalisable. Pimicotinib has Beijing 100035, China with robust efficacy (including quality-of-life domains), well potential to address the critical unmet needs of patients with niuxiaohui@263.net tolerated safety profiles, and generalisability to a diverse global TGCT by offering a highly selective, potent, systemic treatment population. There remains a need for highly selective, systemic with broad efficacy and a manageable safety profile, while therapies for a broad population of patients with TGCT, alleviating pain and improving physical function to enhance especially given the relatively young age of patients facing this quality of life in this life-limiting condition. There are unanswered chronic disease with potentially devastating morbidity. questions about the use of systemic treatments in TGCT that Added value of this study could be addressed with real-world usage. These include duration MANEUVER is the first international, randomised, placebo- of treatment, optimal start and restart times (stop-and-go controlled, double-blind, phase 3 trial designed to evaluate the approach), dosing strategies—such as dose modifications or dose efficacy and safety of the CSF-1R inhibitor pimicotinib in patients intervals that balance efficacy, safety, and tolerability—treatment with TGCT. In adult patients with TGCT who were enrolled at sites sequencing, and impact of CSF-1R inhibitors on family planning. abnormalities.2 CSF-1 binds to CSF-1 receptor (CSF-1R), TGCT and 34% with localised TGCT had at least resulting in local recruitment, proliferation, and one recurrence).3,4,7 For patients with un resectable TGCT, accumulation of CSF-1R-expressing non-neoplastic systemic treatment might be the only suitable treatment inflammatory cells, which comprise most of the tumour.2 modality. TGCT aects an otherwise healthy adult population CSF-1R is a validated therapeutic target for systemic (median age 38 years) with potentially devastating TGCT treatment.4,9,10 CSF-1R tyrosine kinase inhibitors morbidity.3–5 The high symptom burden of TGCT is have transformed the treatment landscape of TGCT. In characterised by mild to extremely debilitating the ENLIVEN and MOTION pivotal trials, the CSF-1R symptoms.4,5 In a study from the international TGCT inhibitors pexidartinib and vimseltinib showed a Support Registry, published in 2025, patients indicated significant objective response rate (ORR) per Response pain (92%), swelling (86%), stiness (83%), and reduced Evaluation Criteria in Solid Tumours (RECIST) version 1.1 range of motion (85%) as common symptoms.4 of 39% for pexidartinib and 40% for vimseltinib Dierential diagnosis of TGCT is challenging, often (both p<0·0001 vs placebo).11,12 However, there has been leading to delays in treatment.4,6 Although TGCT is not concern over the unfavourable risk–benefit profile of life-threatening, it can cause irreversible joint and bone pexidartinib, particularly the risk for severe damage if not treated adequately.3,7,8 hepatotoxicity.9,11,13,14 As of 2026, vimseltinib is registered in Surgery has been the primary treatment for controlling the USA and Europe, and pexidartinib is registered in TGCT to date. However, in some cases it has resulted in the USA, South Korea, and Taiwan.9,10,15 Given the absence substantial morbidity. There is a risk of postoperative of consensus for optimal surgical treatment and the complications, delayed recovery, and disease recurrence limitations with current systemic therapies, there remains (among patients with previous surgery, 72% with diffuse a need for the wider availability of highly selective systemic Articles therapies with broader ecacy and an acceptable safety were identified based on clinical assessments. Eligible profile.4,12 patients were aged 18 years or older, and had histologically Pimicotinib is an oral, once-daily, highly selective, and confirmed, symptomatic TGCT. Symptomatic disease was potent tyrosine kinase receptor inhibitor that prevents defined as patient-reported worst stiness or worst pain of CSF-1R phosphorylation and subsequent accumulation of at least 4 (scale of 0 to 10, with 10 being stiness or pain as non-neoplastic CSF-1R-expressing cells in the tumour.16–18 bad as imaginable) in the 2 weeks preceding In preclinical studies, pimicotinib was shown to be over randomisation. Patients were required to have confirmed, 40 times more selective for CSF-1R than for other tyrosine unresectable disease, defined as a TGCT lesion that was kinases in the CSF-1R family (KIT, FLT3, PDGFRα, extensively invasive, located in a complex anatomical site, VEGFR2).18 This selectivity might limit o-target activity and could not be completely resected, or when surgery that is often associated with unwanted side-eects.11,19 In a might have caused dysfunction or serious complications, phase 1 trial of patients with TGCT not amenable to as confirmed by two clinical experts. All patients were also surgery, pimicotinib 50 mg once daily had an ORR required to have measurable disease per RECIST of 67·5% at week 25 and 85% at 2-year follow-up (both by version 1.1, with at least one lesion measuring at least RECIST version 1.1) over a median treatment duration of 2 cm by MRI, and to provide a tumour tissue specimen or 20·7 months.20 Patients had durable improvements in histologically confirmed TGCT for central pathological joint pain and stiness, with generally mild treatment- diagnosis. Patients were excluded if they had been treated emergent adverse events and no signs of serious liver previously with CSF-1 or CSF-1R inhibitors before injuries or hair hypopigmentation.20 We report here the randomisation. Previous treatment with multikinase primary analysis of the double-blind, part 1 phase of the inhibitors that include the CSF-1 or CSF-1R pathway, such MANEUVER trial, which compared the ecacy and safety as imatinib and nilotinib, was allowed. Full study eligibility of pimicotinib versus placebo in patients with TGCT. criteria are listed in appendix 2 (pp 5–6). Patients self- reported sex (male or female) and race or ethnicity. All Methods patients provided written informed consent to participate Study design in any study-related activity. MANEUVER is an international, multicentre, randomised, placebo-controlled, double-blind, phase 3 Randomisation and masking trial conducted at 23 specialised hospitals across China, Eligible patients were randomly assigned (2:1) to seven in Europe, and ten in North America. All sites pimicotinib or matching placebo via a central interactive were required to have a medical specialist with web response system, a service system for masked knowledge about TGCT, sucient resources to conduct randomisation and trial supply management that ensured the study, and comply with the study protocol. Part 1 the integrity of study procedures. An independent (presented here) is the 24-week double-blind study, statistician provided the randomisation schedule. Each part 2 (ongoing) is the 24-week open-label treatment patient was assigned a unique randomisation number. phase, and part 3 (ongoing) is the long-term open-label Patients were stratified by region (China vs non-China). In extension phase. The trial was conducted in accordance part 1, patients, study investigators and personnel, central with the principles of the Declaration of Helsinki and imaging readers, reviewers, and the study funder were Good Clinical Practice guidelines. The protocol, masked to assigned treatment. Un masking for the funder amendments, and informed consent forms were only was allowed after all patients completed part 1 and approved by the local institutional review board or reached the primary ecacy analysis timepoint at week 25. independent ethics committee at participating sites Data analysis was done after data cleaning and database See Online for appendix 2 (appendix 2 pp 3–4) and by the appropriate regulatory lock. The appearance of placebo was identical to pimicotinib authorities. A more detailed description of MANEUVER capsules and contained similar excipients but without the is published;16 the protocol and statistical analysis plan active pharmaceutical ingredient. are available in appendix 2 (pp 20–226). A key protocol amendment was made during the study, namely to add Procedures an open-label extended treatment period (ie, part 3) that Patients received either pimicotinib 50 mg once daily provides a longer treatment option for patients who (with no food restrictions) or matching placebo completed part 2. Neither patients nor members of the administered orally in 28-day cycles for 24 weeks public were involved directly in the initial design, (six cycles). Patients who had received pimicotinib, conduct, or reporting of this trial. The study is registered completed part 1 (24-week, double-blind treatment period), at ClinicalTrials.gov (NCT05804045) and is ongoing. and still met the study eligibility criteria could continue to part 2 at the same dose they received in part 1; placebo- Patients treated patients could cross over to receive open-label Patients were recruited to the study through a multicentre pimicotinib 50 mg once daily for 24 weeks. Patients who approach, using specialised hospitals that provided completed part 2 and still met the study eligibility criteria comprehensive care for TGCT, where eligible individuals could continue to part 3 (open-label treatment extension) 1074 Articles at the same dose they received in part 2. Patients attended approximately 30% reduction on an 11-point NRS for pain two clinic visits (day 1 and day 15) during cycles 1 and 2, using the self-reported Patient Global Impression of and one clinic visit (day 1) during subsequent cycles. Change scale, a commonly used validated measure for Tumour imaging (MRI) was done at screening (baseline), assessing clinically important dierences in pain.23 Health- week 13, and week 25, or the end of treatment within related quality of life was another secondary ecacy 25 weeks, whichever occurred first. Clinical outcome analysis assessed at week 25 as mean change from baseline assessments were conducted at baseline (before using the EQ-5D-5L visual analogue scale score. Analysis randomisation), week 13, and week 25 for relative range of of data for the additional secondary endpoints (duration of motion (assessed by the treating physician, under masked response by BIRC per RECIST version 1.1, duration of conditions, using a goniometer per the American Medical response by BIRC per tumour volume score, ORR by Association reference standard), and at baseline, week 1, investigator per RECIST version 1.1 at week 25, duration of and every 4 weeks for worst stiness numerical rating response by investigator per RECIST version 1.1, NCI scale (NRS) score, Brief Pain Inventory (BPI) worst pain Patient-Reported Outcomes version of the CTCAE, and NRS score, physical function (TGCT-specific Patient- pharmacokinetics) and exploratory endpoints listed in the Reported Outcomes Measurement Information protocol is ongoing and will be reported separately. Safety System–Physical Function [PROMIS-PF]),21 and health- endpoints including treatment-emergent adverse events related quality of life using the 5-level EQ-5D questionnaire were evaluated with NCI CTCAE version 5.0 and included (EQ-5D-5L) visual analogue scale score. Clinical events leading to dose modification, dose interruptions, evaluations, including vital signs, physical examination, and treatment discontinuation. laboratory evaluations, and review of medications (study drug, analgesic use, concomitant medications) were done Statistical analysis at each clinic visit. Adverse event reporting was continuous The first planned analysis was the primary analysis of this from signed informed consent until, and including, trial and included randomly assigned patients who 30 days after the last dose of study drug, and each event completed part 1 with a follow-up MRI scan at week 25, was graded according to the National Cancer Institute withdrew consent, or died. For the primary endpoint, (NCI) Common Terminology Criteria for Adverse Events assuming an ORR at week 25 of 40% in the pimicotinib (CTCAE) version 5.0. Investigator discretion for dose group and 6% in the placebo group, a sample size of interruption, dose reduction, clinically indicated 75 evaluable patients provided 90% power to detect a supportive care for disease-related symptoms, and dierence between the two groups (Fisher’s exact test with medication for symptom relief (eg, antidiarrhoeals, a two-sided significance level of α=0·05). The primary antiemetics) was permitted. endpoint ORR target for pimicotinib was based on 85% of the lower limit of the 95% CI for interim ORR in the Outcomes The primary endpoint was ORR at week 25 by blinded independent review committee (BIRC) per RECIST 126 patients assessed for eligibility version 1.1 in the intention-to-treat (ITT) population. The BIRC comprised three industry radiologists, who had 32 excluded after screening due to extensive experience with TGCT and tumour volume not meeting eligibility criteria score. Two radiologists assessed each timepoint for all patients. If there was discordance between them, a third 94 randomly assigned radiologist who was masked to the identity of the two primary readers, was not provided with clinical data, and not involved in the primary radiology review did an independent adjudication review. Key secondary 63 assigned to pimicotinib 31 assigned to placebo 63 included in intention-to-treat 31 included in intention-to-treat endpoints at week 25 were analysed in the following analysis analysis prespecified hierarchical order: (1) BIRC-assessed ORR based on tumour volume score;22 and clinical outcome assessments for mean change from baseline in: (2) relative 63 received pimicotinib 31 received placebo range of motion of the aected joint (relative to reference criteria for the same joint); (3) worst stiness NRS score; 4 discontinued treatment 2 crossed over to part 2 in advance, (4) BPI worst pain NRS score; and (5) PROMIS-PF score. 3 withdrawal by patient based on physician decision A supportive analysis of key secondary endpoints was 1 physician decision the proportion of patients achieving a decrease in BPI worst pain NRS score of at least 30%, without an increase 59 completed double-blind treatment 29 completed double-blind treatment in narcotic analgesic use of at least 30% (BPI-30 period (part 1) period (part 1) responders) at week 25. The threshold to establish a clinically important dierence in pain is based on Figure 1: Trial profile Articles phase 1 study (data not published); the target for the by treatment group, and exact two-sided 95% CIs were placebo group was the upper 95% CI of the placebo group calculated using the Clopper–Pearson method. The in ENLIVEN.11 Primary and key secondary ecacy primary estimand and estimand for ORR per tumour endpoints were analysed for all randomly assigned patients volume score also accounted for intercurrent events and (intention-to-treat population), and safety was analysed in included any response (RECIST version 1.1) at week 25 all patients who received at least one dose of the assigned regardless of modifications due to toxicity, discontinuation treatment (safety analysis population). ORRs at week 25 in of treatment without disease progression, or before the ITT population and across subgroups were summarised receiving an alternative anticancer therapy (including surgery). For clinical outcomes including range of motion, Pimicotinib (n=63) Placebo (n=31) Total (N=94) stiness, pain, and physical function, intercurrent events were similarly handled as the primary estimand, except Age, years 41·0 (31–52) 36·0 (27–47) 40·0 (30–52) when patients received an alternative antitumour therapy Sex (including surgery) in which case data were collected and Male 18 (29%) 12 (39%) 30 (32%) used regardless (treatment policy strategy). Female 45 (71%) 19 (61%) 64 (68%) Key secondary endpoints were tested sequentially in the ECOG PS prespecified order (1–5 listed above) using a hierarchical 0 21 (33%) 12 (39%) 33 (35%) gatekeeping procedure, whereby they were tested only if 1 42 (67%) 19 (61%) 61 (65%) the preceding endpoint showed a statistically significant Liver function treatment eect. Analysis of the key secondary endpoints, Normal 59 (94%) 29 (94%) 88 (94%) except ORR per tumour volume score, was done using Mild dysfunction 3 (5%) 2 (7%) 5 (5%) mixed models for repeated measurements, which Moderate dysfunction 1 (2%) 0 1 (1%) included change from baseline as the dependent variable, Race and treatment, baseline, visit, stratification factor of Asian 32 (51%) 16 (52%) 48 (51%) China versus non-China, and treatment by visit White 26 (41%) 11 (36%) 37 (39%) interaction, and the baseline by visit interaction as fixed Black or African American 3 (5%) 2 (7%) 5 (5%) eects; relative range of motion also included joint Other* 2 (3%) 2 (7%) 4 (4%)* category as a fixed term. Clinically meaningful changes Geographical region from baseline to week 25 were established at the China 31 (49%) 14 (45%) 45 (48%) individual patient level based on psychometric analyses Europe 18 (29%) 10 (32%) 28 (30%) and exit interviews. The proportion of BPI-30 responders North America 14 (22%) 7 (23%) 21 (22%) at week 25 was compared between treatment groups. All TGCT type between-group statistical tests were compared at a Diffuse 53 (84%) 27 (87%) 80 (85%) two-sided α=0·05 (Fisher’s exact test). Safety data were Localised 8 (13%) 3 (10%) 11 (12%) summarised descriptively. Missing data for all outcomes Unknown 2 (3%) 1 (3%) 3 (3%) were not imputed and only observed data were analysed. Tumour location There was no planned formal interim analysis. Knee 33 (52%) 14 (45%) 47 (50%) As per the statistical analysis plan, any patients Ankle 9 (14%) 5 (16%) 14 (15%) regarded as having protocol deviations were included in Hip 7 (11%) 6 (19%) 13 (14%) the primary and key secondary ecacy endpoint Foot 4 (6%) 4 (13%) 8 (9%) analyses, and in the safety analysis. Statistical analyses Wrist 5 (8%) 2 (7%) 7 (7%) were done using SAS software version 9.4. An Other† 5 (8%) 0 5 (5%) independent data monitoring committee ensured continuous monitoring of the safety profile and oversaw Lower or upper extremity the conduct of the trial. Lower 54 (86%) 29 (94%) 83 (88%) Upper 9 (14%) 2 (7%) 11 (12%) Role of the funding source Previous surgery for TGCT Abbisko Therapeutics designed and funded the study, 0 26 (41%) 12 (39%) 38 (40%) provided study drug, and was involved in the collection, ≥1 37 (59%) 19 (61%) 56 (60%) analyses, and reporting of data. Previous systemic treatment for TGCT Yes (imatinib) 2 (3%) 4 (13%) 6 (6%) Results No 61 (97%) 27 (87%) 88 (94%) Between April 27, 2023, and March 29, 2024, 126 patients Data are n (%) or median (IQR). ECOG PS=Eastern Cooperative Oncology Group performance status. with TGCT were screened and 94 patients were randomly TGCT=tenosynovial giant cell tumour. *Includes one patient with mixed races of “Black or African American” and assigned to oral pimicotinib 50 mg once daily (n=63) or “White”, and one each who reported as “Colombian”, “Caribbean”, and “South American”. †Includes two patients reported as ‘right foot (big thumb)’, ‘left jaw’. matching placebo (n=31). 88 (94%) of 94 patients completed part 1 treatment; four (6%) of 63 patients Table 1: Baseline demographic and disease characteristics (intention-to-treat population) discontinued pimicotinib treatment (three [5%] patients 1076 Articles Pimicotinib (n=63) Placebo (n=31) Absolute difference, p value pimicotinib–placebo (95% CI*) Primary endpoint Objective response rate at week 25: BIRC based on RECIST 54% (41 to 67) 3% (0·1 to 17) 51% (33 to 63) <0·0001‡ version 1.1† Complete response 1 (2%) 0 ·· ·· Partial response 33 (52%) 1 (3%)§ ·· ·· Stable disease 20 (32%) 28 (90%) ·· ·· Progressive disease 2 (3%)¶ 0 ·· ·· Not evaluable|| 7 (11%) 2 (7%) ·· ·· Key secondary endpoints Objective response rate at week 25: BIRC based on TVS† 62% (49 to 74) 3% (0·1 to 17) 59% (41 to 70) <0·0001‡ Complete response 1 (2%) 0 ·· ·· Partial response 38 (60%) 1 (3%)§ ·· ·· Stable disease 16 (25%) 28 (90%) ·· ·· Progressive disease 1 (2%) 0 ·· ·· Not evaluable 7 (11%) 2 (7%) ·· ·· Relative ROM, at baseline; at week 25** 59 (94%); 56 (89%) 31 (100%); 28 (90%) ·· ·· Baseline, mean 70·6 (20·7) 66·1 (21·8) ·· ·· Week 25, mean 84·9 (21·4) 67·6 (23·6) ·· ·· LS mean change from baseline at week 25††‡‡ 15·6 (10·4 to 20·9) −0·1 (−7·0 to 6·9) 15·7 (7·3 to 24·1) 0·0003 Worst stiffness NRS score, at baseline; at week 25** 63 (100%); 56 (89%) 31 (100%); 29 (94%) ·· ·· Baseline, mean 5·3 (1·9) 5·6 (1·7) ·· ·· Week 25, mean 2·3 (1·8) 4·8 (2·3) ·· ·· LS mean change from baseline at week 25††‡‡ −3·0 (−3·5 to −2·6) −0·6 (−1·2 to 0·1) −2·4 (−3·2 to −1·7) <0·0001 BPI worst pain NRS score, at baseline; at week 25** 63 (100%); 56 (89%) 31 (100%); 29 (94%) ·· ·· Baseline, mean 4·8 (2·1) 4·8 (2·6) ·· ·· Week 25, mean 2·5 (1·9) 4·4 (2·5) ·· ·· LS mean change from baseline at week 25††‡‡ −2·3 (−2·7 to −1·9) −0·2 (−0·8 to 0·3) −2·1 (−2·8 to −1·4) <0·0001 PROMIS-PF score§§, at baseline; at week 25** 62 (98%); 57 (90%) 31 (100%); 29 (94%) ·· ·· Baseline, mean 41·2 (7·1) 39·8 (6·0) ·· ·· Week 25, mean 46·6 (7·8) 42·6 (7·8) ·· ·· LS mean change from baseline at week 25††‡‡ 5·6 (4·2 to 7·1) 2·2 (0·2 to 4·2) 3·4 (0·9 to 5·9) 0·0074 Other secondary endpoints EQ-5D-5L VAS, at baseline; at week 25** 63 (100%); 58 (92%) 31 (100%); 29 (94%) ·· ·· Baseline, mean 71·7 (14·6) 69·0 (14·3) ·· ·· Week 25, mean 79·4 (11·6) 70·9 (16·9) ·· ·· LS mean change from baseline at week 25†† 8·1 (5·8 to 10·5) 0·4 (-3·0 to 3·8) 7·7 (3·6 to 11·9) 0·0004 BPI-30 response rate¶¶ at week 25‡ 64% (50 to 75) 16% (5 to 34) 47% (27 to 61) <0·0001 Data are n (%), % (95% CI), mean (SD), or LS mean (95% CI) unless stated otherwise. BIRC=blinded independent review committee. BPI=Brief Pain Inventory. EQ-5D-5L=5-level EQ-5D questionnaire. LS=least squares. NRS=numerical rating scale. PROMIS-PF=Patient-Reported Outcomes Measurement Information System–Physical Function. RECIST=Response Evaluation Criteria in Solid Tumours. ROM=range of motion. TVS=tumour volume score. VAS=visual analogue scale. *95% CI for ratio difference was derived using the Wilson method. †95% CIs for rates were calculated using the exact Clopper–Pearson method. ‡p values were obtained using Fisher’s exact test. §The single placebo responder, who had been receiving imatinib from May 2022 until discontinuation in February 2024 (4 weeks before study treatment initiation), with stable disease as the best overall response, showed a partial response under placebo; the possibility of a spontaneous regression or a delayed effect from imatinib cannot be ruled out. ¶One patient initially had a 52% decrease in tumour size (partial response) by week 13 and then a subsequent 38% increase (progressive disease) at week 25; however, by week 37 the tumour size had reduced by 62% (partial response), and the patient was still on treatment. ||Patients were non-evaluable in the pimicotinib group due to early discontinuation from part 1 (n=5) or not having an evaluable target lesion (n=2); in the placebo group, patients were non-evaluable due to entering part 2 (open-label pimicotinib) before completing part 1. **Proportion of the total population who completed the outcome assessment at baseline and week 25. ††LS mean and 95% CI values are summarised by treatment groups. Estimated using mixed-model repeated measures with fixed effects for treatment, baseline, visit, stratification factor of China versus non-China sites, and treatment-by-visit interaction, baseline-by-visit interaction; additionally, joint-type category (knee, ankle, and others) was included as a fixed effect for relative ROM. An unstructured variance–covariance matrix was used. ‡‡A hierarchical gatekeeping testing procedure was used to control the family-wise type 1 error rate at a two-sided α level of 0·05. Statistical significance for any subsequent secondary endpoints was not assessed unless the treatment effect on all preceding secondary endpoints was statistically significant. §§The PROMIS-PF scale includes two different sets of topics for upper and lower extremities, which were used for assessment for patients with upper and lower extremity tumours, respectively. ¶¶BPI-30 responder was defined as a patient with a decrease of at least 30% in mean BPI worst pain NRS who did not have an increase of at least 30% in narcotic analgesic use. Table 2: Primary and secondary endpoint data at week 25 Articles withdrew consent, one [2%] withdrew due to physician treatment with pimicotinib in part 2 (figure 1). Major decision); two (7%) of 31 patients discontinued placebo protocol deviations occurred in 14 (22%) patients in the treatment due to symptomatic progression (both due to pimicotinib group and three (10%) patients in the placebo physician decision) and went on to receive open-label group; patients could have had more than one deviation. † 0 –10 –20 –30 –40 § –50 Tumour response ‡ Figure 2: Objective response –60 status at week 25 per patient (A) and by Complete response –70 patient subgroup (B)* Partial response (A) The bars represent best –80 Stable disease Progressive disease percentage change in the sum –90 Not evaluable of diameters of target lesions –100 in individual patients during the double-blind study period (until week 25), and the colour coding indicates the objective response achieved at a single time point: week 25. (B) Objective response rate at week 25 by patient subgroup. ECOG PS=Eastern Cooperative Oncology Group performance status. TGCT=tenosynovial giant cell tumour. *By blinded independent review committee based on Response Evaluation Criteria in Solid Tumours version 1.1. †Tumour size increased continuously during the treatment period in this patient, reaching a 43% increase from baseline at week 25 (progressive disease). ‡This patient initially had a decrease in tumour size of 52% (partial response) by week 13 and then a subsequent increase of 38% (progressive disease) at week 25; however, by week 37 their tumour size had reduced by 62% (partial response), and they were still on treatment. §This patient previously received imatinib treatment (9 months before evaluation) and had a partial response at week 25; the patient had no response before discontinuation on the day of consent and was randomly assigned after the protocol-mandated 4-week washout. ¶Subgroups with one treatment group containing fewer than five patients were not subjected to statistical testing and were reported only descriptively. 1078 fo mus ni enilesab morf egnahc egatnecrep tseB )%( snoisel tegrat fo retemaid eht A Pimicotinib (n=63) Placebo (n=31) Patients Patients B Objective response rate, n/N (%) Absolute difference (pimicotinib–placebo), % (95% CI)¶ Pimicotinib group Placebo group Tumour location Lower extremity 27/54 (50%) 1/29 (3%) 47% (28–60) Upper extremity 7/9 (78%) 0/2 (0%) 78% Disease location Knee 16/33 (48%) 0/14 (0%) 48% (22–65) Others 18/30 (60%) 1/17 (6%) 54% (27–70) Region China 20/31 (65%) 0/14 (0%) 65% (37–79) Not China 14/32 (44%) 1/17 (6%) 38% (12–56) Age group, years <40 16/29 (55%) 1/17 (6%) 49% (22–66) ≥40 18/34 (53%) 0/14 (0%) 53% (26–69) Sex Male 10/18 (56%) 1/12 (8%) 47% (13–68) Female 24/45 (53%) 0/19 (0%) 53% (31–67) Number of previous surgeries 0 12/26 (46%) 1/12 (8%) 38% (6–57) ≥1 22/37 (59%) 0/19 (0%) 59% (36–74) Previous systemic therapy Yes 1/2 (50%) 1/4 (25%) 25% No 33/61 (54%) 0/27 (0%) 54% (37–66) TGCT type Localised 5/8 (63%) 0/3 (0%) 63% Diffuse 28/53 (53%) 1/27 (4%) 49% (30–62) Unknown 1/2 (50%) 0/1 (0%) 50% ECOG PS 0 12/21 (57%) 1/12 (8%) 49% (15–68) 1 22/42 (52%) 0/19 (0%) 52% (30–67) Overall 34/63 (54%) 1/31 (3%) 51% (33–63) –10 0 10 20 30 40 50 60 70 80 90 100 Favours placebo Favours pimicotinib Articles In the pimicotinib group, these protocol deviations were due to missing study procedures in 11 (17%) patients, study treatment or investigational medicinal product non- compliance in two (3%) patients, eligibility deviation in two (3%) patients, and investigational medicinal product dispensing or accountability in one (2%) patient. In the placebo group, these protocol deviations were due to 15 missing study procedures in two (6%) patients and other 10 reasons in one (3%) patient. All other amendments were 5 minor, with no expected eect on patients or study outcomes. –5 The study population was generally representative of patients aected by TGCT (table 1) with baseline –10 demographic and clinical characteristics of patients in both groups being well balanced: median age was 40·0 years (IQR 30–52), 30 (32%) patients were male, 64 (68%) patients were female, 80 (85%) had diuse TGCT, and 56 (60%) had undergone at least one previous surgery. 88 (94%) patients were systemic treatment-naive. 45 (48%) patients were enrolled in the study in China, 28 (30%) in Europe, and 21 (22%) in North America. The data cuto date for the primary analysis was Sept 23, 2024. MANEUVER met its primary endpoint. The ORR at week 25 by BIRC per RECIST version 1.1 was 54% (34 of 63) in the pimicotinib group compared with 3% (one of 31) in the placebo group (absolute dierence 51% [95% CI 33–63], p<0·0001, table 2). In the pimicotinib group, one (2%) patient achieved a complete response and 33 (52%) patients had partial responses. The first scheduled study visit for tumour response evaluation by MRI was at week 13, when 26 (41%) patients in the pimicotinib group had an objective tumour response. The one (3.2%) patient who had a response in the placebo arm as discussed above had the response by week 13. 58 (92%) patients in the pimicotinib group had a reduction in tumour size per RECIST version 1.1 during the double-blind study period (compared with 22 [71%] in the placebo group; figure 2A). This eect on tumour reduction was consistent across prespecified subgroups, including age, sex, Eastern Cooperative Oncology Group performance status, geographical region, disease location, and number of previous surgeries (figure 2B). All key secondary endpoints were met. The ORR at week 25 by BIRC per tumour volume score was 62% (39 of 63) in the pimicotinib group compared with 3% (one of 31) in the placebo group (absolute dierence 59% [95% CI 41–70], p<0·0001), consistent with findings based on RECIST version 1.1 (table 2). Completion rates for worst stiness, worst pain, and physical function were high: 54–63 (86–100%) patients in the pimicotinib group and 28–31 (90–100%) patients in the placebo group (appendix 2 p 7). Statistically significant improvements in change from baseline for all clinical outcome assessments were seen with pimicotinib versus placebo at week 25: relative range of motion (p=0·0003), worst stiness NRS score (p<0·0001), BPI worst pain NRS score (p<0·0001), and PROMIS-PF (p=0·0074; enilesab morf egnahc )ES( naem SL A 1·0 0·5 –0·5 –1·0 –1·5 –2·0 –2·5 –3·0 –3·5 –4·0 enilesab morf egnahc )ES( naem SL B 1·0 0·5 –0·5 –1·0 –1·5 –2·0 –2·5 –3·0 enilesab morf egnahc )ES( naem SL C 8·0 7·0 6·0 5·0 4·0 3·0 2·0 1·0 –1·0 –2·0 Baseline 5 9 13 17 21 25 Time since randomisation (weeks) enilesab morf egnahc )ES( naem SL Placebo group p=0·0003 Pimicotinib group † * p<0·0001 * ‡ * * * * p<0·0001 * ‡ * * * D p=0·0074 † * Figure 3: Change from baseline in clinical outcome assessments in the intention-to-treat population Relative range of motion (A), worst stiffness NRS score (B), BPI worst pain NRS score (C), and physical function on the PROMIS-PF T-score (D). BPI=Brief Pain Inventory. LS=least squares. NRS=numerical rating scale. PROMIS-PF=Patient-Reported Outcomes Measurement Information System–Physical Function *p<0·05 for LS mean group difference at this timepoint; p values are nominal. †p<0·05 for LS mean group difference at week 25; p values are significant (tested in hierarchical order). ‡p<0·0001 for LS mean group difference at week 25; p values are significant (tested in hierarchical order). Articles pimicotinib showed early improvem ents across all clinical Pimicotinib (n=63) Placebo (n=31) outcome assessments, with benefits seen in relative range Any grade Grade 3 or 4 Any grade Grade 3 of motion, worst stiness, worst pain, and physical or 4 function at the first scheduled assessments. Improve- Any treatment-emergent adverse event 63 (100%) 22 (35%) 29 (94%) 1 (3%) ments in clinical outcome assessments were also Any treatment-related adverse event 62 (98%) 18 (29%) 18 (58%) 0 observed at various thresholds (appendix 2 pp 8–9). Serious adverse event 3 (5%) 0 1 (3%) 0 40 (64%) patients in the pimicotinib group compared Treatment related 1 (2%) 0 0 0 with five (16%) patients in the placebo group achieved a Leading to dose modification 36 (57%) 14 (22%) 2 (6%) 1 (3%) BPI-30 response at week 25 (absolute dierence 47% Leading to dose reduction 5 (8%) 0 0 0 [95% CI 27–61], p<0·0001; table 2). Four (6%) patients in Leading to dose interruption 34 (54%) 14 (22%) 2 (6%) 1 (3%) the pimicotinib group used concomitant narcotic Leading to trial discontinuation 1 (2%) 0 0 0 analgesics compared with six (19%) patients in the Treatment related 1 (2%) 0 0 0 placebo group in the BPI-30 analysis. There was Leading to death 0 0 0 0 significant improvement in EQ-5D-5L visual analogue Treatment-emergent adverse event occurring in >10% of patients in either group scale score in the pimicotinib group compared with the Laboratory abnormalities* placebo group at week 25 (absolute dierence 7·7 [95% CI Blood creatine phosphokinase increased 45 (71%) 8 (13%) 5 (16%) 0 3·6–11·9], p=0·0004; table 2). Blood lactate dehydrogenase increased 36 (57%) 0 0 0 The median duration of exposure was 169 days in both Aspartate aminotransferase increased 34 (54%) 0 3 (10%) 0 groups (IQR 167–173 in both groups), with a median Amylase increased 22 (35%) 0 0 0 pimicotinib dose intensity of 94·6% (85·8–99·4). Adverse events, mostly grade 1 or 2, occurred in 63 (100%) α-hydroxybutyrate dehydrogenase increased 16 (25%) 0 0 0 patients treated with pimicotinib and 29 (94%) patients Lipase increased 15 (24%) 2 (3%) 1 (3%) 0 who received placebo (table 3). Most of the common Blood creatine phosphokinase MB increased 12 (19%) 0 1 (3%) 0 (>10% patients) events reported with pimicotinib were Alanine aminotransferase increased 11 (17%) 0 3 (10%) 0 asymptomatic, reversible abnormalities in laboratory HDL increased 10 (16%) 0 0 0 parameters, including increases in blood creatine Blood creatinine increased 10 (16%) 0 0 0 phosphokinase, blood lactate dehydrogenase, aspartate Hypercholesterolaemia 9 (14%) 0 3 (10%) 0 aminotransferase, amylase, α-hydroxybutyrate dehydro- White blood cell count decreased 8 (13%) 0 1 (3%) 0 genase, and lipase. The most common (>10% patients) Apolipoprotein A-I increased 7 (11%) 0 0 0 clinical events with pimicotinib included pruritus, facial Blood cholesterol increased 7 (11%) 0 1 (3%) 0 oedema, rash, periorbital oedema, fatigue, nausea, and Neutrophil count decreased 7 (11%) 1 (2%) 0 0 headache. The most common (>10% patients) treatment- Other clinical adverse events emergent adverse events in the placebo group were Pruritus 33 (52%) 2 (3%) 1 (3%) 0 fatigue, arthralgia, and increases in blood creatine Facial oedema 30 (48%) 0 0 0 phosphokinase (table 3; appendix 2 pp 10–11). Elevations Rash 22 (35%) 2 (3%) 0 0 in liver function tests (aspartate aminotransferase and Periorbital oedema 20 (32%) 0 3 (10%) 0 alanine aminotransferase) were mild and asymptomatic Fatigue 18 (29%) 0 7 (23%) 0 in both groups (appendix 2 pp 12–13). All adverse events Nausea 17 (27%) 0 2 (6%) 0 are shown in appendix 2 (pp 14–18). Headache 13 (21%) 0 2 (6%) 0 Grade 3 or 4 treatment-emergent adverse events were Dizziness 10 (16%) 0 1 (3%) 0 reported in 22 (35%) patients in the pimicotinib group, Insomnia 9 (14%) 0 2 (6%) 0 most commonly increased blood creatine phosphokinase, Hypertension 9 (14%) 2 (3%)† 1 (3%) 0 which occurred in eight (13%) patients (table 3; Rash maculopapular 7 (11%) 1 (2%) 2 (6%) 0 appendix 2 pp 10–11) and increased blood pressure in Arthralgia 4 (6%) 0 7 (23%) 0 four (6%) patients. One serious treatment-related adverse Data are n (%). *Laboratory abnormalities were all asymptomatic and responded well to brief dose interruptions. event of grade 3 increased blood pressure occurred in †Included one patient with a medical history of hypertension, and one patient with abnormal blood pressure at one pimicotinib-treated patient who had a medical baseline. history of hypertension and white-coat syndrome. No Table 3: Summary of safety data (safety analysis set) placebo-treated patients had a serious treatment-related adverse event. There were no deaths in the trial. Five (8%) patients treated with pimicotinib required table 2). Mean scores by clinic visit for all four clinical dose reductions due to adverse events (dermatitis, rash, outcome assessments are shown in figure 3; p values headache, hypersomnia, fatigue, and increased blood before week 25 are nominal. Clinically meaningful creatine phosphokinase each occurred in one [2%] improvements were observed in all clinical outcome patient) versus none in the placebo group (table 3). assessments irrespective of tumour response per RECIST Transient dose interruptions due to adverse events version 1.1 and tumour volume score. In addition, occurred in 34 (54%) patients in the pimicotinib group; 1080 Articles these events were primarily serum enzyme elevations, family life and social activities, resulting in further and the most frequent was increased blood creatine isolation.3–5 These limitations can have substantial phosphokinase in nine (14%) patients. Non-laboratory implications for work and careers, aecting the livelihood adverse events that led to temporary treatment of patients and their families.3–5,7 Furthermore, patients interruptions that occurred in more than 5% of patients might have trouble concentrating due to pain or side- included facial oedema in six (10%) patients, rash in five eects of pain medications. They might be unable to stand, (8%) patients, and periorbital oedema and pruritus each sit, or operate machinery and vehicles. Some patients occurred in four (6%) patients (appendix 2 p 19). might reduce their work hours, change their occupation Dose interruptions due to adverse events occurred in entirely, or retire early, which can create economic burden two (7%) patients in the placebo group (arthralgia and and stress.4,5 The potentially highly debilitating and chronic prostate cancer). The mean percentage of intended nature of TGCT aects the physical, emotional, and cumulative dose received was 90% (SD 12·2) for financial health of patients.4–7 For patients with TGCT, pimicotinib and 98% (5·3) for placebo. Among those who reducing symptom burden, such as range of motion, worst had dose interruptions, 32 (82%) of 39 patients in the stiness, worst pain, and physical function, can pimicotinib group, and two (100%) of two patients in the meaningfully improve quality of life and empower patients placebo group had no more than two treatment to have better control of their disease.4,6,7,24 interruptions. The median duration of dose interruption Currently approved systemic therapies for TGCT due to adverse events was 10 days (IQR 7–15) for the include pexidartinib and vimseltinib, depending on pimicotinib group and 13 days (2–24) for the placebo regional availability.10,12 Pexidartinib is a multitargeted group. One adverse event-related treatment discon- kinase inhibitor that primarily targets CSF-1R.25 tinuation was reported as grade 2 fatigue in one (2%) Treatment with pexidartinib might pose a risk of fatal patient in the pimicotinib group. cholestatic hepatotoxicity and requires registration in a There was no evidence of cholestatic liver toxicity or US Food and Drug Adminstration-mandated Risk drug-induced liver injury, and no reports of skin or hair Evaluation and Mitigation Strategy programme; drug hypopigmentation. labeling informa tion was not readily available for South Korea and Taiwan.9,26 Skin or hair hypopigmentation Discussion and the twice-daily dosing regimen with dietary This randomised, placebo-controlled, double-blind, requirements could aect adherence to pexidartinib, phase 3 trial (MANEUVER) evaluated the ecacy and which is an especially important consideration given the safety of pimicotinib in a geographically diverse young age of patients with TGCT.9,11,13 In fact, patients population of patients with TGCT. Pimicotinib showed receiving pexidartinib have identified convenience and early, robust, statistically significant, and clinically side-eects as domains that limit their overall treatment meaningful antitumour activity in more than half of satisfaction.13 Vimseltinib is a potent CSF-1R inhibitor treated patients with TGCT (ORR 54%).11,12 By the data that is not associated with severe hepatotoxicity; it has a cuto, 58 (90%) of 63 patients in the pimicotinib group twice-weekly regimen that requires at least 72 h between had a decrease in tumour size. Pimicotinib antitumour doses and can be taken without food restrictions.10,12 activity was consistent across prespecified subgroups, There are currently no data directly comparing suggesting broad ecacy. pimicotinib with pexidartinib or vimseltinib in a global Patients treated with pimicotinib had early, durable, randomised study, in part due to the unavailability of robust, clinically meaningful, and statistically significant these agents in some regions, including China.9,10 With improvements in range of motion, worst stiness, worst this caveat, the diverse population of patients treated pain, and physical function, regardless of tumour with pimicotinib in MANEUVER had a higher ORR at response. Notably, the proportion of patients achieving a week 25 than that reported in the ENLIVEN BPI-30 response at week 25 was four times higher with (pexidartinib)11 and MOTION (vimseltinib)12 trials (39% pimicotinib than with placebo. The consistently high [ENLIVEN] and 40% [MOTION] ORR all by BIRC per questionnaire completion rates for patient-reported RECIST version 1.1). The multikinase inhibitors imatinib measures further strengthen these findings. and nilotinib have shown clinical activity in TGCT in a The potential for TGCT as a life-limiting disease might retrospective analysis and a phase 2 study, respectively,27,28 not be fully realised.4–7,24 Physical functioning and quality of but not in a randomised study, and neither has received life domains are especially important given the relatively regulatory approval for TGCT.4 Both imatinib and young age of patients with TGCT.4,5,7,24 Chronic pain and nilotinib are currently prescribed o-label in clinical disability can create heightened emotional and practice (as allowed) due to the large unmet need in this psychological burden.3,4,7 For patients with TGCT, most of disease4,29 and, in many countries, they remain the only whom are otherwise healthy, high symptom burden might available therapeutic options.27,30 Emactuzumab, an impede routine daily activities (eg, cooking, bathing, intravenously administered CSF-1R-targeted monoclonal exercising, lifting children, running errands) and social antibody, is in late-stage clinical investigation for the participation.3–5,7 Patients with TGCT might withdraw from treatment of TGCT.31 Articles In MANEUVER, pimicotinib had a manageable safety TGCT in balanced proportions,11 which might make the profile, with no evidence of cholestatic hepatotoxicity or findings generalisable to geographically diverse popu- drug-induced liver injury, as shown by comprehensive lations. Another strength of the study is the comprehensive evaluation of liver function. Serum enzyme elevations data collection, with high rates of completion of patient- (eg, liver transaminases) were observed with pimicotinib, reported outcomes and patient participation, which as with pexidartinib and vimseltinib. In MANEUVER, the strengthen the validity of the clinical outcome assessments serum enzyme elevations, including increases in blood and the quality of the trial. Limitations of the study include creatine phosphokinase, blood lactate dehydrogenase, a slight imbalance in the tumour location (upper and lower aspartate aminotransferase, amylase, α-hydroxybutyrate limb) between groups: this characteristic was not included dehydrogenase, and lipase, were asymptomatic, reversible, in the stratification factors and could not be perfectly and consistent with the known mechanism of CSF-1 controlled in this small study. In addition, the inflammatory pathway inhibition (depletion of Kuper cells resulting in nature of TGCT might be a limitation when it comes to decreased clearance of these enzymes, but no signs of accuracy of measurement of tumour response, organ damage).32 Consistent with the minimal o-target demonstrating high inter-reader variability. This factor was KIT inhibition with pimicotinib, patients in MANEUVER mitigated by independent central reading, which ensured did not have skin or hair hypopigmentation that was a consistent approach and, when necessary, adjudication. associated with pexidartinib.11 In addition, there were no In conclusion, the pivotal MANEUVER trial supports events of grade 3 oedema in MANEUVER, which were the use of pimicotinib as a highly selective, potent, reported in ENLIVEN and MOTION.11,12 Notably, the safety convenient therapeutic option that provides early and profile of pimicotinib reflects additional laboratory robust tumour reduction, with a manageable safety profile, parameters due to dierences in regional practice in a diverse population of adult patients with TGCT. (α-hydroxybutyrate dehydrogenase was more commonly Patients who received pimicotinib had early, durable, measured at sites in China) and dierences in specific significant, and clinically meaningful improvements in CSF-1R inhibitor study protocols (MANEUVER evaluated range of motion, stiness, pain, and physical function— lipase, amylase, and blood concentrations of lactate quality of life domains that are highly relevant for this dehydrogenase). Adverse events were manageable with socially and professionally active population, and brief dose interruptions that did not aect treatment particularly to those not amenable to surgery. adherence or dose intensity. In MANEUVER, the oral, Contributors once-daily regimen of pimicotinib was well tolerated, as HX, XN, VR, HG, QZ, BS, and XZhu provided the study concept and evidenced by a low rate of dose reductions (five [8%] of design; JM-B, AAR, RS, YZ, JS, TL, KKS, CS, SS, JW, GGB, YF, YH, TL, PR, XZha, and GT accessed, verified, and analysed the data. All authors 63 patients) and treatment discontinuation (one [2%] of had access and contributed to the interpretation of the trial data, 63 patients). Tolerability of pimicotinib contributes to its participated in preparation of the manuscript, and approved the final favourable safety profile, especially in a chronic, potentially draft submitted for publication. All authors vouch for the completeness debilitating disease such as TGCT. In comparison, more and accuracy of the data, and fidelity of the trial to the protocol and statistical analysis plan. than a third of patients had to reduce the dose with vimseltinib in the MOTION trial and treatment with Declaration of interests VR received consulting fees and honoraria from EMD Serono. pexidartinib required discontinuation in more than 10% of JM-B received grants (paid to his institution) from Abbisko Therapeutics, patients in the ENLIVEN study.11,12 Despite the rate of Biosplice, Boehringer Ingelheim, Bristol-Myers Squibb, Cebiotex, treatment interruptions in MANEUVER (34 [54%] of Deciphera, Eli Lilly, Immunome, Inhibrx, Karyopharm, Lixte, Pfizer, 63 patients; median duration 10 days vs 13 days for PharmaMar, Philogen, PTC Therapeutics, Rain Therapeutics, Springworks Therapy, and Synox; consulting fees from Deciphera and placebo), pimicotinib dose intensity remained high PharmaMar; honoraria from Abbisko Therapeutics, Adium, Bayer, (median 94·6%) during part 1. Dose interruptions were Boehringer Ingelheim, Cogent Bioscience, Deciphera, GSK, PharmaMar, predominantly due to asymptomatic serum enzyme and Roche; travel support from PharmaMar; and is the CEO of Sarcoma elevations (a class eect of CSF-1R inhibitors) and Research Solutions. AAR received institutional support from 23andMe, Abbisko Therapeutics, AbbVie, Adaptimmune, Amgen, AstraZeneca, managed according to protocol-prespecified criteria.32,33 Bayer, BioNTech, Blueprint Medicines, Boehringer Ingelheim, Bristol- Additionally, median duration of exposure was 169 days at Myers Squibb, Cogent Biosciences, Daiichi Sankyo, Deciphera, Frontier the end of part 1. The optimal duration of systemic Biopharma, Gilead, GSK, Intensity Therapeutics, Iterion Therapeutics, treatment in TGCT is currently unknown. In Karyopharm, Kelun-Biotech, MedImmune, Medison, Merck, Neoleukin Therapeutics, Novartis, Pfizer, Polaris, Roche/Genentech, Rain MANEUVER, the median duration of response was not Therapeutics, and Symphogen; honoraria from Boehringer Ingelheim reached. The broader question of optimal length of time and Medison; and participated on a data safety monitoring or advisory on systemic treatment, benefit-to-risk trade-o with board for CD40Life and Inhibrx. CS received honoraria from Deciphera, treatment holidays, and best start and stop times is an PharmaMar, and Roche; travel support from Bayer, Gilead, and PharmaMar; was a steering committee member for NewBay; and ongoing area of research interest in TGCT. participated in advisory boards for Deciphera and IDRX. SS received Strengths of MANEUVER include a broad patient grants (paid to her institution) from Abbisko Therapeutics, Daiichi population from Asia, Europe, and North America that Sankyo, Deciphera, and SynOx Therapeutics; and participated in advisory was generally representative of patients with TGCT. The boards for Daiichi Sankyo, Deciphera, and Merck Serono. GGB received consulting fees from Eli Lilly and PharmaMar; honoraria from Eli Lilly, trial enrolled Chinese and non-Chinese patients with GSK, Istituto Gentili, MSD, and PharmaMar; travel support from 1082 Articles Eli Lilly, Istituto Gentili, and PharmMar; and participated in advisory 13 Dai D, Pan I, Freivogel K, Ye X, Tecson K, Tap W. Patient-reported boards for Deciphera, GSK, and MSD. PR received research funding continued benefits of pexidartinib for tenosynovial giant cell tumor (paid to his institution) from Bristol-Myers Squibb, Novartis, Pfizer, based on a real-world study in the United States. Oncologist 2025; Roche; consulting fees from Bristol-Myers Squibb, Genesis, Medison 30: oyaf028. Pharma, MSD, Novartis, Pierre Fabre, and Philogen; honoraria from 14 Chen D, Yang XW, Fu J, et al. A disproportionality analysis of Bristol-Myers Squibb, Genesis, Medison Pharma, MSD, Novartis, and adverse events caused by pexidartinib from the FDA adverse event Pierre Fabre; participated in speakers bureaus for Bristol-Myers Squibb, reporting system. Sci Rep 2025; 15: 31689. Genesis, MSD, Novartis, and Pierre Fabre; and travel support from 15 Wagner AJ, Tap WD, Bauer S, et al. Long-term ecacy and safety of Pierre Fabre. GT participated in advisory boards for Deciphera and pexidartinib in patients with tenosynovial giant cell tumor: final results of the ENLIVEN study. Oncologist 2025; 30: oyae345. SynOx. All other authors declare no competing interests. 16 Niu X, Ravi V, Shan B, et al. MANEUVER: a phase III study of Data sharing pimicotinib to assess ecacy and safety in tenosynovial giant cell The datasets generated and/or analysed during the MANEUVER study tumor patients. Future Oncol 2024; published online Sept 17. https:// will not be publicly shared due to privacy and confidentiality concerns. doi.org/10.1080/14796694.2024.2396227. Deidentified or aggregated data might be made available from the 17 Xu H, Zhou Y, Li B, et al. Updated ecacy and safety profile of corresponding author upon reasonable request and subject to applicable pimicotinib (ABSK021) in tenosynovial giant cell tumor ethical and regulatory approvals. (TGCT): 1-year follow-up from Phase 1b. Connective Tissue Oncology Society (CTOS) Annual Meeting. Nov 1–4, 2023 (poster 169). Acknowledgments 18 Yang S, Zhao B, Zhang M, et al. Abstract LB-288: a highly selective We thank the patients who agreed to participate in this trial, as well as small molecule CSF-1R inhibitor demonstrates strong their families and carers, and the investigating teams at all participating immunomodulatory activity in syngeneic models. Cancer Res 2018; sites. Abbisko Therapeutics designed and funded the study. Merck holds 78 (suppl 13): LB-288 (abstr). the rights to commercialise pimicotinib worldwide. Merck (CrossRef 19 Xu H, Zhou Y, Li B, et al. Ecacy and safety profile of pimicotinib Funder ID: 10.13039/100009945) funded the medical writing support for (ABSK021) in tenosynovial giant cell tumor (TGCT): phase 1b this manuscript; the authors critically revised the manuscript and update. American Society of Clinical Oncology (ASCO) Annual provided substantive input and ultimate approval. Medical writing Meeting. June 2–6, 2023 (poster 493). support, under the direction of the authors, was provided by Trang H Au 20 Xu H, Li Y, Zhou Y, et al. Long-term ecacy and safety profile of and Abigail Bailey of the Publications Division of Omnicom Health pimicotinib (ABSK021) in tenosynovial giant cell tumor: phase 1b Medical Communications (formerly Caudex, an IPG Health Company), update. Connective Tissue Oncology Society (CTOS) Annual in accordance with Good Publication Practice (GPP 2022) guidelines. Meeting. Nov 13–16, 2024 (poster P407). Editorial note: The Lancet Group takes a neutral position with respect to 21 Gelhorn HL, Ye X, Speck RM, et al. The measurement of physical territorial claims in published text. functioning among patients with Tenosynovial Giant Cell Tumor (TGCT) using the Patient-Reported Outcomes Measurement References Information System (PROMIS). J Patient Rep Outcomes 2019; 3: 6. 1 Mastboom MJL, Verspoor FGM, Verschoor AJ, et al, and the TGCT 22 Peterfy C, Chen Y, Countryman P, et al. CSF1 receptor inhibition of study group. Higher incidence rates than previously known in tenosynovial giant cell tumor using novel disease-specific MRI tenosynovial giant cell tumors. Acta Orthop 2017; 88: 688–94. measures of tumor burden. Future Oncol 2022; 18: 1449–59. 2 Tap WD, Healey JH. Role of colony-stimulating factor 1 in the 23 Marcus J, Lasch K, Wan Y, Yang M, Hsu C, Merante D. An assessment neoplastic process of tenosynovial giant cell tumor. Tumour Biol of clinically important dierences on the worst pain severity item of 2022; 44: 239–48. the modified brief pain inventory in patients with diabetic peripheral 3 Stacchiotti S, Dürr HR, Schaefer IM, et al. Best clinical management neuropathic pain. Pain Res Manag 2018; 2018: 2140420. of tenosynovial giant cell tumour (TGCT): a consensus paper from 24 Gelhorn HL, Tong S, McQuarrie K, et al. Patient-reported the community of experts. Cancer Treat Rev 2023; 112: 102491. symptoms of tenosynovial giant cell tumors. Clin Ther 2016; 4 Stern S, McKenzie PF, Bernthal N, et al. Localized and diuse 38: 778–93. tenosynovial giant cell tumor: real-world results from a patient 25 Gelderblom H, Wagner AJ, Tap WD, et al. Long-term outcomes of observational registry. Future Oncol 2025; 21: 1501–10. pexidartinib in tenosynovial giant cell tumors. Cancer 2021; 5 Bernthal NM, Spierenburg G, Healey JH, et al, and the TOPP Study 127: 884–93. Group. The diuse-type tenosynovial giant cell tumor (dt-TGCT) 26 Dharmani C, Wang E, Salas M, et al. Turalio risk evaluation and patient journey: a prospective multicenter study. Orphanet J Rare Dis mitigation strategy for treatment of tenosynovial giant cell tumor: 2021; 16: 191. framework and experience. Future Oncol 2022; 18: 1595–607. 6 Cook NS, Landskroner K, Shah B, Walda S, Weiss O, Pallapotu V. 27 Verspoor FGM, Mastboom MJL, Hannink G, et al. Long-term Identification of patient needs and preferences in pigmented ecacy of imatinib mesylate in patients with advanced tenosynovial villonodular synovitis (PVNS) using a qualitative online bulletin giant cell tumor. Sci Rep 2019; 9: 14551. board study. Adv Ther 2020; 37: 2813–28. 28 Gelderblom H, Cropet C, Chevreau C, et al. Nilotinib in locally 7 Riedel RF, Eward W, Verini C, et al. Improving resources and advanced pigmented villonodular synovitis: a multicentre, open- support for patients with tenosynovial giant cell tumor. New York, label, single-arm, phase 2 trial. Lancet Oncol 2018; 19: 639–48. NY: CancerCare, 2021. 29 Mastboom MJL, Lips W, van Langevelde K, et al. The eect of 8 Stamatiou A, Nguyen-Ngoc T, Wetterwald L, et al. Overview of imatinib mesylate in diuse-type tenosynovial giant cell tumours pharmacological therapies for diuse tenosynovial giant cell tumor. on MR imaging and PET-CT. Surg Oncol 2020; 35: 261–67. Future Pharmacol 2023; 3: 926–37. 30 Spierenburg G, Grimison P, Chevreau C, et al. Long-term follow-up 9 US Food and Drug Administration. TURALIO® (pexidartinib). of nilotinib in patients with advanced tenosynovial giant cell US Prescribing Information. 2025. https://www.accessdata.fda.gov/ tumours: long-term follow-up of nilotinib in TGCT. Eur J Cancer drugsatfda_docs/label/2022/211810s009lbl.pdf (accessed Oct 27, 2025). 2022; 173: 219–28. 10 US Food and Drug Administration. ROMVIMZATM (vimseltinib). 31 Palmerini E, Veltkamp S, Gelderblom H, et al. 151TiP TANGENT: US Prescribing Information, 2025. https://www.accessdata.fda.gov/ a phase III, global, randomized, double-blind, placebo-controlled trial drugsatfda_docs/label/2025/219304s000lbl.pdf (accessed to assess the ecacy and safety of emactuzumab with tenosynovial April 9, 2025). giant cell tumor. ESMO Open 2025; 10 (suppl 3): 104464. 11 Tap WD, Gelderblom H, Palmerini E, et al, and the ENLIVEN 32 Radi ZA, Koza-Taylor PH, Bell RR, et al. Increased serum enzyme investigators. Pexidartinib versus placebo for advanced tenosynovial levels associated with Kuper cell reduction with no signs of giant cell tumour (ENLIVEN): a randomised phase 3 trial. Lancet hepatic or skeletal muscle injury. Am J Pathol 2011; 179: 240–47. 2019; 394: 478–87. 33 Wang T, Papoutsi M, Wiesmann M, et al. Investigation of 12 Gelderblom H, Bhadri V, Stacchiotti S, et al, and the MOTION correlation among safety biomarkers in serum, histopathological investigators. Vimseltinib versus placebo for tenosynovial giant cell examination, and toxicogenomics. Int J Toxicol 2011; 30: 300–12. tumour (MOTION): a multicentre, randomised, double-blind, placebo-controlled, phase 3 trial. Lancet 2024; 403: 2709–19. --- [PDF原文](https://sci-net.xyz/storage/7932541/ab4bf55e10412514c5f2f6b4a1719810e2e68e6e60b275154465560d06b4a6d4/Pimicotinib-versus-placebo-for-tenosynovial-giant-cell-tumour-MANEUVER-an-international-randomised.pdf) DOI: 10.1016/S0140-6736(25)02602-9