Oncology · Medical Literature
Palbociclib in Patients With Ovarian Cancer With CDKN2A Alterations: Results From the Targeted Agent and Profiling Utilization Registry Study.
Abstract
The Targeted Agent and Profiling Utilization Registry Study is a phase II basket trial evaluating the antitumor activity of commercially available targeted agents in patients with advanced cancer and genomic alterations. Results of a cohort of patients with ovarian cancer (OC) with CDKN2A alterations treated with palbociclib are reported. Eligible patients had advanced OC, measurable disease (RECIST), Eastern Cooperative Oncology Group performance status 0-2, adequate organ function, tumors with CDKN2A alterations, and no remaining standard treatment options. The primary end point was disease control (DC), defined as objective response (OR) or stable disease (SD) of at least 16+ weeks duration (SD16+). Simon two-stage design with a null DC rate of 15% versus 35% (power = 0.85; α = .10) was used. Secondary end points included OR, progression-free survival (PFS), overall survival (OS), duration of response, duration of SD, and safety. Twenty-eight patients were enrolled from February 2017 to April 2023, and all had OC with a CDKN2A alteration (loss [n = 21], mutation [n = 5], and loss and mutation [n = 2]). One patient was not evaluable for efficacy because of an excluded genomic finding. Three patients had a partial response and seven had SD16+ for DC and OR rates of 38% (90% CI, 25 to 100) and 11% (95% CI, 2 to 29), respectively. The null hypothesis of 15% DC rate was rejected (P = .004). The median PFS was 8 weeks (95% CI, 8 to 18) and the median OS was 33 weeks (95% CI, 24 to 73). Nineteen patients (68%) had at least one treatment-related grade 3-4 adverse event (AE) or serious AE including anemia, dyspnea, fatigue, leukopenia, neutropenia, and thrombocytopenia. Palbociclib met the prespecified criteria to declare a signal of activity in patients with OC and CDKN2A alterations.
Clinical relevance
In patients with advanced ovarian cancer harboring CDKN2A alterations who have exhausted standard therapy, does single-agent palbociclib achieve a disease control rate sufficient to declare a signal of antitumor activity? __SNAPSHOT__ {"studyDesign":"Phase II basket trial, single-arm cohort (Simon two-stage design) — part of the TAPUR (Targeted Agent and Profiling Utilization Registry) study (NCT02693535)","setting":"Multi-site, community and academic oncology centers across the United States (CLIA-certified, CAP-accredited genomic testing laboratories)","sampleSize":"n=28 enrolled; n=27 evaluable for efficacy (1 patient excluded post-enrollment due to an excluded genomic finding)","population":"Adults with advanced ovarian cancer, measurable disease by RECIST v1.1, ECOG PS 0–2, adequate organ function, CDKN2A alteration (loss, mutation, or both), RB wild-type, and no remaining standard treatment options. CDKN2A alteration subtypes: loss (n=21), mutation (n=5), loss and mutation (n=2). Most common co-alterations: CDKN2B (n=17), TP53 (n=15), PIK3CA (n=5). Median age 57 years (range 33–85); ECOG PS 0 in 32%, PS 1 in 68%; majority (71%) had received ≥3 prior systemic regimens.","intervention":"Palbociclib 125 mg orally once daily for 21 days followed by 7 days off (standard 28-day cycle), continued until disease progression","comparator":"Not applicable (single-arm study)","primaryOutcome":"Disease control (DC) rate, defined as objective response (complete or partial response by RECIST v1.1) or stable disease of ≥16 weeks duration (SD16+). Null DC rate: 15%; alternative DC rate: 35%; if ≥7/28 evaluable patients achieved DC, the null rate was rejected.","followUp":"Enrollment from February 2017 to April 2023; individual follow-up duration not reported in abstract","fundingSource":"Not reported in the abstract; the TAPUR study is an ASCO-sponsored trial; Pfizer (manufacturer of palbociclib) provides drug support — independent non-industry statistical leadership noted in prior TAPUR publications"}
Key findings
The primary endpoint was met: the null DC rate of 15% was rejected (p=0.005). Among 27 evaluable patients, the DC rate was 37% (90% CI: 24%–100%), exceeding the pre-specified threshold of ≥7/28 patients with DC. Three partial responses (PRs) were observed, all in patients with CDKN2A deletion, yielding an objective response rate (ORR) of 11% (95% CI: 2%–28%). Seven patients achieved SD16+: CDKN2A deletion (n=5), mutation (n=1), and mutation plus deletion (n=1). Durations of response for 2 patients with PR were 26 and 60 weeks; DOR was not available for the third patient with PR (discontinued study procedures at time of response). Median duration of SD for SD16+ patients was 30 weeks (range: 16–78 weeks). Median PFS was 9 weeks (95% CI: 8–16 weeks). Median OS was 36 weeks (95% CI: 25–56 weeks). NNT for DC (calculated): 1/0.37 ≈ 2.7. Almost all patients with DC had serous carcinoma histology. Safety: 19 of 28 patients (68%) had at least one treatment-related grade 3–4 adverse event (AE) or serious AE, including neutropenia, leukopenia, anemia, thrombocytopenia, fatigue, and dyspnea — a clinically significant toxicity burden given the heavily pre-treated population.
Practice implications
This study formally established a statistically significant signal of antitumor activity for single-agent palbociclib in patients with advanced, heavily pre-treated ovarian cancer harboring CDKN2A alterations, meeting its pre-specified Simon two-stage criterion (p=0.005). The 37% DC rate exceeded the null hypothesis threshold and was driven predominantly by patients with CDKN2A loss (deletion), with all PRs occurring in that subgroup. However, the objective response rate of 11% is modest, median PFS of only 9 weeks reflects limited durability in the overall cohort, and median OS of 36 weeks (~8.3 months) reflects a heavily pre-treated, poor-prognosis population. Clinically, the signal is meaningful in the context of refractory ovarian cancer with no remaining standard options, but the wide confidence interval on the DC rate (90% CI: 24%–100%), small sample size, single-arm design, absence of a comparator arm, and lack of centralized genomic confirmation limit the strength of inference. Patients were required to have RB wild-type tumors, and most had CDKN2B co-alterations and TP53 co-mutations, which may influence generalizability. Patients with non-serous histology, RB mutations, CCNE1 amplification, or earlier lines of therapy were not adequately represented. The high rate of grade 3–4 toxicity (68%) warrants careful patient selection and monitoring. This study does not establish palbociclib as a new standard of care and should not change practice in isolation; it provides the justification for further confirmatory study in a genomically selected population. Note: this summary is based on abstract-level data; full published manuscript data (JCO Precision Oncology, July 2026) should be reviewed for complete survival analyses, subgroup data, and additional safety detail. __LIMITATIONS__ Small sample size (n=27 evaluable) with wide confidence intervals, limiting precision of all effect estimates,Single-arm design with no comparator; DC rate cannot be attributed to palbociclib independent of natural disease history or selection bias,Heterogeneous genomic testing platforms (CLIA-certified but not centrally mandated), raising concerns about comparability of CDKN2A alteration classification across sites,Basket trial design: while enabling signal detection, it introduces cohort-level confounding from variation in prior treatments, co-mutations (TP53, CDKN2B, PIK3CA), and histologic subtypes,Median PFS of 9 weeks and OS of 36 weeks with wide confidence intervals suggest limited durability of benefit across the full evaluable cohort,Absence of blinding and investigator-assessed response (not independently reviewed) introduces potential for subjective bias in SD classification,No patient-reported outcomes or quality-of-life data reported,High grade 3–4 toxicity rate (68%) may limit real-world applicability in a PS 0–2 population,Limited generalizability: all but one patient with DC had serous carcinoma histology; findings may not apply to non-serous ovarian cancer subtypes,Results based on abstract only; full-manuscript data for PFS/OS curves, subgroup analyses, and dose modification details are not yet available for this summary __BOTTOM_LINE_CATEGORY__ promising_but_preliminary __BOTTOM_LINE__ Single-agent palbociclib met the pre-specified signal criterion for disease control (DC rate 37%, p=0.005) in heavily pre-treated ovarian cancer patients with CDKN2A alterations, particularly those with CDKN2A loss, establishing proof-of-concept for genomically directed CDK4/6 inhibition in this tumor type. However, the objective response rate was only 11%, median PFS was 9 weeks, and 68% of patients experienced grade 3–4 toxicity, underscoring modest clinical magnitude of benefit. Confirmatory randomized or larger prospective studies are needed before palbociclib can be considered a practice-changing option in this biomarker-selected population.