Maintenance treatment after first-line platinum-based chemotherapy remains an area of investigation in advanced esophagogastric adenocarcinoma, with the aim of sustaining disease control while maintaining quality of life and allowing recovery from toxicities associated with induction chemotherapy.
The PLATFORM study evaluated maintenance rucaparib after first-line platinum-based chemotherapy in patients with advanced HER2-negative esophagogastric adenocarcinoma.
The results appear in the September 2026 issue of ESMO Open under the title “Maintenance rucaparib after first-line platinum-based chemotherapy in HER2-negative advanced oesophagogastric adenocarcinoma: results from the PLATFORM study.”
Authors: A. Gordon, C. Fong, L. Satchwell, S. Cromarty, K. Piadel, A. Tran, R. Begum, A. Woodward, B. Leamon, O. Zhitkov, M. Terlizzo, M. Hubank, P. Proszek, G. Pietka, E. Peat, E. Cartwright, M. Davidson, P. Das, S. Sothi, R. Herbertson, S. Darby, T. Waddell, A. Bradshaw, S. Rao, N. Starling, I. Chau, and D. Cunningham.
The PLATFORM Study
PLATFORM is a prospective, open-label, multicenter, adaptive phase II trial investigating maintenance strategies in patients with advanced HER2-negative esophagogastric adenocarcinoma following first-line platinum-based chemotherapy.
Eligible patients had inoperable locally advanced or metastatic adenocarcinoma of the esophagus, gastroesophageal junction, or stomach and had achieved either stable disease or a response after first-line fluoropyrimidine-platinum chemotherapy. Patients were required to have HER2-negative disease, an ECOG performance status of 0-2, and adequate organ function. Prior treatment with a PARP inhibitor was not allowed.
Patients were recruited across 46 centers in the United Kingdom. Those who achieved disease control after 18 weeks of first-line chemotherapy and remained eligible could be randomized to surveillance or one of the PLATFORM maintenance treatment arms. For the rucaparib comparison, patients assigned to surveillance were reviewed every 28 days, while patients assigned to rucaparib received 600 mg orally twice daily continuously in 28-day cycles until progression, death, or toxicity.
The primary endpoint was progression-free survival (PFS), defined as the time from randomization to radiological progression according to RECIST 1.1, clinical progression, or death. Secondary endpoints included overall survival (OS), objective response rate, and safety. Exploratory analyses evaluated outcomes according to homologous recombination deficiency (HRD) status and somatic genomic alterations.
Maintenance Rucaparib Versus Surveillance
Between November 2017 and November 2022, 125 patients were contemporaneously randomized to the two groups: 62 to surveillance and 63 to maintenance rucaparib. The median follow-up was 41 months. At the time of analysis, 56 of 62 patients in the surveillance group and 55 of 63 patients in the rucaparib group had experienced disease progression. Forty-six and 51 patients, respectively, had died.
PFS Numerically Longer, but Primary Endpoint Not Met
Median PFS was 4.2 months with rucaparib compared with 2.8 months with surveillance, corresponding to an unadjusted HR of 0.70 (95% CI 0.49-1.02; one-sided P=0.031). The study had specified a one-sided alpha of 0.025. Therefore, the difference in PFS was not statistically significant.
The 12-month PFS rates were 14.0% with rucaparib and 3.7% with surveillance, while the corresponding 24-month rates were 6.0% and 1.9%. In the modified intention-to-treat population, the 12-week progression-free rate was 55% with rucaparib and 36% with surveillance, with an odds ratio of 2.2 (95% CI 1.1-4.6; one-sided P=0.018). This met the interim criterion for non-futility and would have supported continued recruitment to full accrual.
No radiological responses according to RECIST were observed in either group. However, only 39% of patients in each group within the modified intention-to-treat population had measurable disease at randomization.
No Difference in Overall Survival
No difference in OS was observed between the treatment groups. Median OS was 9.2 months with rucaparib and 10.7 months with surveillance, with an unadjusted HR of 1.15 (95% CI 0.77-1.72; one-sided P=0.247). At 12 months, the OS rate was 40.3% with rucaparib and 48.9% with surveillance. At 24 months, the corresponding rates were 13.3% and 26.5%.
At disease progression, 44% of patients in the surveillance group received subsequent treatment compared with 33% in the rucaparib group. In their discussion, the investigators noted that the OS finding may have been influenced by this imbalance in post-progression treatment.
Exploring HRD and Genomic Biomarkers
Adequate tissue for HRD analysis was available for 51 of 125 patients (41%), including 20 patients in the surveillance group and 31 in the rucaparib group. Among these 51 patients, five were HRD-positive, 41 were HRD-negative, and five had inconclusive results.
Four of the five HRD-positive patients received rucaparib and had PFS durations of 2.6, 3.0, 8.3, and 10.6 months. The single HRD-positive patient in the surveillance group had a PFS of 1.8 months. The investigators noted that comparisons according to HRD status were not possible because of the small subgroup size.
Tumor samples from 53 patients were of sufficient quality for next-generation sequencing. At least one somatic alteration was identified in 28 patients. TP53 was the most frequently mutated gene (82%), followed by PIK3CA (18%), CDKN2A (18%), SMARCA4 (14%), and ARID1A (14%).
No clear association between these somatic alterations and PFS was observed. Alterations in key homologous recombination repair genes were uncommon, with BRCA2 mutations identified in three patients. The longest responder in the rucaparib group remained progression-free at 38 months and was classified as HRD-negative, with somatic alterations in TP53, PIK3CA, and CDKN2A.
Overall, the investigators reported that HRD positivity did not predict clinical benefit from maintenance rucaparib and that individual genomic alterations did not consistently correlate with PFS benefit.
Safety With Maintenance Rucaparib
The safety population included 124 patients: 61 in the surveillance group and 63 in the rucaparib group. Adverse events of any grade occurred in 98% of patients in both groups. Grade ≥3 adverse events were more frequent with rucaparib, occurring in 32% of patients compared with 23% with surveillance.
Treatment-related adverse events were reported in 84% of patients receiving rucaparib, and 22% experienced grade 3-4 treatment-related adverse events. The most common grade 3-4 treatment-related events were anemia, fatigue, infection, and neutropenia.
Six serious adverse events occurred in the surveillance group and 16 in the rucaparib group. No treatment-related deaths were reported.
Early Trial Closure and Key Limitations
The rucaparib comparison did not reach its planned full enrollment. The trial had calculated that 154 patients per group and 248 PFS events would be required for the definitive analysis. Recruitment to the rucaparib arm stopped in November 2022 after Clovis Oncology withdrew industry support.
Although the interim 12-week progression-free rate met the criterion for non-futility, recruitment closed prematurely following the withdrawal of industry support. Fewer than half of the required PFS events occurred, leaving the study underpowered for its primary endpoint. The translational analyses were also limited by substantial tumor sample attrition. More than half of available FFPE specimens were excluded because of inadequate tumor content or poor DNA quality, and archival rather than fresh tissue was used.
Another limitation was the absence of a validated HRD assay for esophagogastric adenocarcinoma. The HRD assay used in the study had been validated in ovarian cancer. The investigators therefore considered the HRD analysis exploratory and hypothesis-generating.
Takeaway
In the PLATFORM study, maintenance rucaparib did not significantly improve PFS compared with surveillance in patients with advanced HER2-negative esophagogastric adenocarcinoma whose disease was controlled after first-line chemotherapy. No difference in OS was observed.
The exploratory translational analyses provided preliminary information on HRD in this population, but HRD status and individual gene alterations did not consistently correlate with PFS benefit. The study was underpowered for its primary endpoint because of premature trial closure, while the translational analyses were limited by small sample size and substantial tumor sample attrition.
The full article is available in ESMO Open.
Find more videos on OncoDaily YouTube TV.