The FDA’s Oncologic Drugs Advisory Committee voted in favor of pexidartinib for an indication as a treatment of adult patients with symptomatic tenosynovial giant cell tumor.
The FDA’s Oncologic Drugs Advisory Committee (ODAC) voted 12 to 3 in favor of pexidartinib for an indication as a treatment of adult patients with symptomatic tenosynovial giant cell tumor (TGCT).
Under the Prescription Drug User Fee Act, the FDA will now make a decision on the pexidartinib indication by August 3, 2019.
A priority review designation was initially granted to the application for the investigational, small molecule, CSF1R inhibitor in February 2019. The designation was based on results from the international phase III ENLIVEN study, which demonstrated a 39.3% overall response rate (ORR; 95% CI, 28%-52%) with pexidartinib versus 0% (95% CI, 0%-6%; P <.0001) with placebo following 24 weeks of treatment based on central review of MRI scans.1
At the ODAC meeting, the committee assessed the clinical benefit of pexidartinib in this patient population, as the interpretation of the activity was limited due to a proportion of data missing at 25 weeks for several secondary endpoints, including range of motion (ROM), physical function, and worst stiffness.2
Moreover, the panel also aimed to characterize the risk of liver injury in patients with TGCT who received therapy with pexidartinib. In the phase III ENLIVEN trial, which comprised the majority of data that were submitted to the FDA, pexidartinib was associated with increases in alanine aminotransferase (ALT; 67%), aspartate aminotransferase (AST; 90%), and total bilirubin (12%); these events reached grade ≥3 severity in one-third of patients. The FDA briefing document for ODAC also stated the long-term safety profile of pexidartinib was unknown.
“I voted, ‘yes.’ While I am quite concerned about the ramifications of the drug on the liver, this drug really does have the potential to be life-changing,” said Valerae O. Lewis, MD, professor and chair, department of Orthopaedic Oncology, John Murray Professor in Orthopaedic Oncology, The University of Texas MD Anderson Cancer Center.
TGCT, which is also known as pigmented villonodular synovitis or giant cell tumor of the tendon sheath, is a nonmalignant tumor of the joint or tendon sheath. The disease can be locally aggressive and debilitating, and is associated with severe morbidity or function limitations. Currently surgery is the primary treatment for patients with TGCT; there are no FDA-approved systemic therapies.
The multicenter, double-blind, phase III ENLIVEN trial had 2 parts, the first of which was a double-blind phase. Here, 120 patients with symptomatic advanced TGCT—in whom surgical removal of the tumor would lead to potentially worsening functional limitation or severe morbidity—were randomized 1:1 to receive either pexidartinib (n = 61) or placebo (n = 59) at 1000 mg daily for 2 weeks followed by 800 mg daily for 22 weeks.
To be eligible for enrollment, patients had histologically confirmed, advanced, symptomatic TGCT with measurable disease ≥2 cm by RECIST v1.1 criteria. Patients were also stratified by US or non-US sites, as well as upper versus lower extremity. Fifty-nine percent of patients were female, and 88% of patients overall were Caucasian. The median age was 45 years (range, 18-79). Eight percent of patients had upper extremity involvement and the remaining 92% had lower extremity involvement.
The primary endpoint was percentage of patients achieving a complete or partial response (PR), assessed with centrally read MRI scans using RECIST 1.1 criteria, after 24 weeks of treatment. Secondary endpoints included ROM, response by tumor volume score (TVS), Patient-Reported Outcomes Measurement Information System (PROMIS) physical function, stiffness, and measures of pain reduction.
A hierarchical procedure had been specified in the analysis plan in order to adjust for multiplicity in testing the following secondary efficacy outcomes: mean change from baseline in ROM of the affected joint, relative to a reference standard for the same joint at week 25; proportion of responders based on a 50% decrease in TVS at week 25 as measured in centrally evaluated MRI scans; mean change from baseline score in the PROMIS Physical Function Scale at week 25; mean change from baseline score in the Worst Stiffness numeric rating scale (NRS) item at week 25; and proportion of responders from patients who experience ≥30% reduction in the mean Brief Pain Inventory (BPI) Worst Pain NRS item and also did not experience a ≥30% increase in narcotic analgesic use (BPI-30) at 25 weeks.
Results of the study showed that the ORR was 39.3% at the end of part 1. The median duration of response (DOR) was not reached with pexidartinib (range, 4.6+ to 24.9+) and was not available in the placebo arm. Ninety-six percent of patients had a DOR ≥6 months, and 50% of patients had a DOR ≥12 months.
At a median 6 months of follow-up, no responders in the trial progressed. Tumor response was assessed by TVS, which was 56% (95% CI, 43.3%-67.5%) with pexidartinib and 0% (95% CI, 0%-6.1%) with placebo (P <.0001). Responses also correlated with improved patient symptoms and function.
There was no statistically significant difference between the 2 arms for worst pain. Moreover, while these improvements in the secondary endpoints of mean change in ROM, ORR per TVS, mean change in physical function per PROMIS, and mean change in worst stiffness for patients on pexidartinib versus placebo were statistically significant, the FDA stated in the briefing document that interpretation of the data for these secondary endpoints were to be viewed cautiously. This is due to a high proportion of missing data at 25 weeks for ROM (27%), physical function (43%), and worst stiffness (43%). There was a similar proportion of patients with missing data across arms.
Regarding safety, the most common adverse events (AEs; ≥20%) that occurred in the pexidartinib arm were changes in hair color, fatigue, increased AST and ALT, edema in the eye(s) and face, vomiting, dysgeusia, and rash. Serious AEs were reported in 13% (n = 8) of patients. AEs that led to treatment discontinuation (13%) were increased ALT/AST and hepatoxicity, and 33% of patients had dose interruptions.
Also, there were patients on pexidartinib who had laboratory abnormalities that indicated drug-induced liver injury (4.9%; upper bound of the 95% CI, 13.7%). These data were consistent with that reported in a pooled analysis of all patients with TGCT in a clinical development program of pexidartinib (n = 130). In the majority of those in the TGCT population who had transaminase elevations and total bilirubin increase did return to baseline levels with dose reductions, interruptions, and/or treatment discontinuation.
Additional results of a pooled safety population of patients with solid or hematologic cancers treated with pexidartinib (n = 630) showed that there was a total bilirubin of ≥2 x upper limit of normal (ULN) and an AST/ALT ≥3 x ULN.
Two patients in an overall development program of pexidartinib (n = 768) experienced irreversible liver injury which led to death 1 in patient and liver transplant in the second. Eight patients whose liver biopsies were obtained to evaluate these liver abnormalities showed a pattern of hepatocellular injury and ductopenia.
Hepatic toxicities were more frequent with pexidartinib versus placebo (AST or ALT ≥3 x upper limit of normal [ULN], 33% vs 0% and total bilirubin ≥2 x ULN, 5% vs 0%). Additionally, 8 patients discontinued pexidartinib treatment because of hepatic AEs. Additionally, 4 patients experienced serious nonfatal AEs with increased bilirubin, one of which lasted about 7 months.
In non-TGCT development trials of pexidartinib, there were 2 severe liver toxicity cases observed. One case required liver transplant and the other was associated with death.
Forty-eight (75%) and 30 (51%) who previously received pexidartinib and placebo in part 1 of the trial and completed therapy continued onto part 2, an open-label extension portion of the ENLIVEN trial. At the time of data cutoff, 39 of the 48 pexidartinib-treated patients and 26 of the 30 patients previously on placebo were still receiving pexidartinib in part 2. Five patients (6%) had discontinued pexidartinib due to AEs and 6 (8%) patients withdrew consent to stay on trial.
A TGCT cohort (n = 39) of the phase I PLX108-01 trial, which evaluated pexidartinib in patients with advanced incurable solid tumors, was also mentioned in the ODAC briefing document. Key disease characteristics of these patients were similar to that of the ENLIVEN trial. Results showed that the ORR was 62% (95% CI, 45%-77%) and the median DOR was not reached (1.8+ to 53.1+).
Pexidartinib was also mentioned in ASCO’s January 2019 announcement of selecting progress in treating rare cancers as its Advance of the Year.3
One of the 3 dissenting ODAC votes was Doris Strader, MD, professor, Department of Medicine, Department of Gastroenterology & Hepatology, The Robert Larner, MD, College of Medicine, University of Vermont Burlington.
“I was concerned about the missing data and was not convinced that there was a real clinically meaningful benefit. Likewise, while I understand that the hepatic injury is not liver failure, I am concerned that this might be persistent for a lifetime and I worry that there was not enough to suggest that there was going to be rigorous monitoring of patients over their lifetime,” said Strader.