A randomized phase 2 study of sapanisertib in combination with paclitaxel versus paclitaxel alone in women with advanced, recurrent, or persistent endometrial cancer

Objective. This phase 2 study investigated sapanisertib (selective dual inhibitor of mTORC1/2) alone, or in combination with paclitaxel or TAK-117 (a selective small molecule inhibitor of PI3K), versus paclitaxel alone in advanced, recurrent, or persistent endometrial cancer. Methods. Patients with histologic diagnosis of endometrial cancer (1–2 prior regimens) were randomized to 28-day cycles on four treatment arms: 1) weekly paclitaxel 80 mg/m2 (days 1, 8, and 15); 2) weekly paclitaxel 80 mg/m2 + oral sapanisertib 4 mg on days 2–4, 9–11, 16–18, and 23–25; 3) weekly sapanisertib 30 mg, or 4) sapanisertib 4 mg + TAK-117 200 mg on days 1–3, 8–10, 15–17, and 22–24. Results. Of 241 patients randomized, 234 received treatment (paclitaxel, n = 87 [3 ongoing]; paclitaxel +sapanisertib, n = 86 [3 ongoing]; sapanisertib, n = 41; sapanisertib+TAK-117, n = 20). The sapanisertib and sapanisertib+TAK-117 arms were closed to enrollment after futility analyses. After a median follow-up of 14.4 (paclitaxel) versus 17.2 (paclitaxel+sapanisertib) months, median progression-free survival (PFS; primary endpoint) was 3.7 versus 5.6 months (hazard ratio [HR] 0.82; 95% confidence interval [CI] 0.58–1.15; p = 0.139); in patients with endometrioid histology (n = 116), median PFS was 3.3 versus 5.7 months (HR 0.66; 95% CI 0.43–1.03). Grade ≥ 3 treatment-emergent adverse event rates were 54.0% with paclitaxel versus 89.5% paclitaxel+sapanisertib. Conclusions. Our findings support inclusion of chemotherapy combinations with investigational agents for advanced or metastatic disease. The primary endpoint was not met and toxicity was manageable. Trial registration: ClinicalTrials.gov number, NCT02725268


Introduction
Advanced or recurrent endometrial cancer is largely chemoresistant [1], and the prognosis for patients is poor [2], with a 5-year survival rate of 17% in patients with distant metastases [3].In patients failing first-line chemotherapy, single-agent paclitaxel has demonstrated overall response rates (ORRs) of ~27% [4,5].Novel agents and combinations have been investigated in phase 2/3 trials, and advancements in immunotherapy treatments have been made; however, chemotherapeutic treatment options remain limited [6][7][8].
First-generation mTOR inhibitors (rapalogs) that target mTOR complex 1 (mTORC1, e.g., temsirolimus and ridaforolimus), have demonstrated preliminary antitumor activity in small patient populations with recurrent endometrial cancer.However, isolated inhibition of mTORC1 without mTORC2 abrogates normal mTORC1-mediated feedback inhibition of insulin receptor substrate 1, enhancing downstream activity of AKT, which is suspected to play a role in the acquisition of treatment resistance [19].Sapanisertib is an oral, potent, and highly selective inhibitor of mTOR kinase that exhibits dual specificity against mTORC1 and 2; this dual inhibition mitigates the feedback activation of AKT that facilitates resistance to rapalogs [20].Sapanisertib has shown promising antitumor activity in preclinical studies, both alone and in combination with TAK-117 (serabelisib), a selective small molecule inhibitor of PI3K, or paclitaxel, in bladder cancer models [21].We speculated that by targeting the PI3K/AKT/mTOR pathway through inhibition of mTORC1 and 2, sapanisertib could potentiate the cytotoxic effects of paclitaxel in patients with recurrent endometrial cancer.Furthermore, addition of PI3K inhibition may provide more robust inhibition of the PI3K/AKT/mTOR pathway.This randomized, phase 2, 4-arm study assessed the efficacy and safety of sapanisertib alone, or in combination with paclitaxel or TAK-117, versus paclitaxel alone in women with advanced, recurrent, or persistent endometrial cancer.

Patients
Female adults with a histologic diagnosis of advanced, recurrent, or persistent endometrial cancer (including endometrioid, serous, mixed adenocarcinoma, clear-cell carcinoma, or carcinosarcoma) that had progressed after prior platinum-based treatment were eligible.Patients were required to have at least 1 but no more than two prior platinum-based chemotherapeutic regimens for management of endometrial cancer.Patients were excluded if they had received previous treatment with any weekly taxane regimen, PI3K/mTOR, TORC1/2 inhibitors, were taking proton pump inhibitors (PPIs) within 7 days of the first dose of study drug, required treatment with PPIs throughout the trial, or were taking H 2 receptor antagonists within 24 h of the first dose of study drug (see Supplementary Methods for eligibility criteria).The study was conducted according to the protocol, the ethical principles that have their origin in the Declaration of Helsinki, in accordance with International Conference for Harmonisation, and all applicable local regulations.All patients provided written informed consent.
The primary endpoint was progression-free survival (PFS), assessed by the investigator according to the Response Evaluation Criteria in Solid Tumors, version 1.1; secondary efficacy endpoints included time to progression, overall survival (OS), ORR, clinical benefit rate (CBR; ORR + stable disease), CBR-16 (CBR at 16 weeks), and duration of response.PFS, time to progression, and OS were assessed from the time of randomization.The incidence of treatment-emergent adverse events (TEAEs) was a secondary safety endpoint.Quality of life (QoL) endpoints included changes from baseline to end-of-study in European Organisation for Research and Treatment of Cancer Quality-of-Life Questionnaire Core 30 (EORTC QLQ-C30) scores.Response according to phosphatase tensin homolog (PTEN) and MSS status were exploratory endpoints.

Assessments
Baseline computerized tomography (CT) or magnetic resonance imaging (MRI) scans of the chest, abdomen, and pelvis were obtained within 4 weeks of the first dose, then every 2 cycles from cycles 2-8 and every 3 cycles thereafter.Partial responses were confirmed by rescanning patients ~4 weeks after the previous scan.Patients attended an end-of-treatment (EOT) visit 30 to 40 days after receiving their last dose of study treatment, or before the start of subsequent anticancer therapy, if that was required sooner than 30 days after EOT for reasons other than disease progression, at which time patients entered post-treatment follow-up for PFS and OS.For patients who discontinued for reasons other than disease progression, CT or MRI scans were completed every 2 months for the first 6 months after the EOT visit, then every 3 months until disease progression.After disease progression, patients were followed for OS every 3 months.Toxicity was evaluated according to the National Cancer Institute Common Technology Criteria for Adverse Events, version 4.03.Patient reported outcomes were evaluated using EORTC QLQ-C30, at baseline, on day 1 of each cycle, and at EOT.

Biomarkers
Tumor specimens provided at study entry were studied for baseline biomarkers.PTEN mutation status was assessed by immunohistochemistry (IHC) with the monoclonal antibody (mAb) 6H2.1 Clone (Agilent/Dako) with a positive cut-off of >5% of tumor cells.MSI/ deficient mismatch repair (dMMR) was assessed by IHC with a cut-off of <5% of tumor cells carrying any of the following MMR proteins: MLH1, MSH2, MSH6, and PMS2 (stained with Biocare mouse mAbs G168-15 [MLH-1] and FE11 [MSH2] and Epitomics rabbit mAbs EP49 [MSH6] and EP51 [PMS2]) or by nucleotide gain on mononucleotide markers (BAT-25, BAT-26, NR-21, NR-24 and MONO-27) using the Promega MSI Analysis System.Molecular profiling of 1100 genes was determined using a custom-designed Illumina Hybrid Selection Takeda/Broad Cancer Panel v2 (5.7 Mb) with tumor and matched-normal blood, at 500× and 250× coverage respectively.Associations between gene mutations and PFS length were analyzed at the gene level using a regression model, and at the pathway level using Fisher's exact test.

Statistical analysis
Two interim analyses using the Bayesian predictive probability design with early stopping rules for futility were planned for the sapanisertib and sapanisertib + TAK-117 arms [22].Arms were closed to enrollment if ≤6 patients from the first 20 patients treated (≥4 cycles of treatment), or ≤10 patients from the first 30 patients treated (≥4 cycles of treatment), achieved clinical benefit at 16 weeks in each arm.
The primary comparison of the primary endpoint of PFS was conducted between the paclitaxel and paclitaxel + sapanisertib treatment arms.Assuming an increase in median PFS from 4 months with paclitaxel alone to 6.5 months with sapanisertib in combination with paclitaxel (hazard ratio [HR] 0.615; approximately 38% reduction in the hazard rate), a total of 134 PFS events and 90 patients per treatment arm were required.This calculation is based on 80% power using a two-sided alpha of 5% and assuming a dropout rate of 15%.The intent-to-treat (ITT) population was defined as all randomized patients and was used for all time-to-event analyses, including the primary endpoint; distributions were estimated using Kaplan-Meier methodology.HRs along with two-sided 95% confidence intervals (CIs) were estimated using a stratified Cox regression model adjusted for histological subtype, lines of prior chemotherapy and prior taxane therapy.Response-evaluable and safety populations are described in the Supplementary Methods.No statistical adjustments were made for multiple comparisons.All statistical analyses were conducted using SAS Version 9.4.
Response was assessed in all patients who received one dose of study drug (safety population; paclitaxel [n = 87], paclitaxel + sapanisertib [n = 86]) (Table 2).Confirmed ORR was 18.4% with paclitaxel (complete response [CR], n = 2; partial response [PR], n = 14) versus 24.4% (CR, n = 2; PR, n = 19) with paclitaxel + sapanisertib; CBR was 57.5% versus 80.2%; CBR-16 was 36.8% versus 51.2%.Response rates by patient histological subtype, PTEN mutation status, and MSI status are shown in Table 3.In patients with endometrioid histology, CBR was 55% with paclitaxel alone versus 84% with paclitaxel + sapanisertib; and in patients with non-endometrioid histology CBR was 63% versus 72%.No patients with high MSI status responded to paclitaxel + sapanisertib, compared with 15 with MSS status; similar results were observed with paclitaxel where only one patient with high MSI status had a response compared with 15 with MSS status.In patients receiving single-agent sapanisertib and sapanisertib + TAK-117, response was assessed at the planned interim analyses after 4 cycles of treatment.CBR-16 was 17.1% with sapanisertib and 5.0% with sapanisertib + TAK-117, and both arms were closed to enrollment per protocol.

Genomic analysis
Molecular profiling in a subgroup of 67 patients in this study confirmed that mutations in PIK3CA and PTEN were more common in endometrioid than non-endometrioid tumors, but were not associated with improved PFS with paclitaxel + sapanisertib compared with paclitaxel in patients with endometrioid histology, consistent with findings from other studies.However, mutations in the beta-catenin signaling pathway were associated with improved PFS with paclitaxel + sapanisertib compared with paclitaxel in patients with endometrioid histology (p < 0.05; Supplementary Tables 1-3; Supplementary Fig. 3).
Forty patients discontinued treatment due to TEAEs (paclitaxel, n = 12; paclitaxel + sapanisertib, n = 13; sapanisertib, n = 9; sapanisertib + TAK-117, n = 6).There were 15 on-study deaths that occurred between the first dose of study drug and 30 days after the last dose; four with paclitaxel, eight with paclitaxel + sapanisertib, one with sapanisertib and two with sapanisertib + TAK-117.Thirteen were considered related to the disease under study.The remaining two were due to hyperbilirubinemia (paclitaxel + sapanisertib) and general physical health deterioration (sapanisertib + TAK-117), both considered unrelated to study drug.

Patient-reported outcomes
In the ITT population, EORTC QLQ-C30 compliance was 96.1% with paclitaxel, and 93.6% with paclitaxel + sapanisertib.Mean global health status/QoL scores tended to show deterioration from baseline with both paclitaxel and paclitaxel + sapanisertib.The difference between mean change from baseline global health status/QoL scores for paclitaxel and paclitaxel + sapanisertib was <10 points (Supplementary Fig. 4) and not significant (p > 0.05).Median time to deterioration of global QoL was 2.1 months for both treatment arms.

Discussion
In this randomized phase 2 study, there was no statistically significant difference in PFS with paclitaxel + sapanisertib versus paclitaxel alone, and the primary endpoint was not met.Median PFS reported with paclitaxel (3.7 months) was as expected for this pre-treated population [23].The addition of sapanisertib increased median PFS to 5.6 months, consistent with previously reported outcomes with other mTOR inhibitors and immunotherapeutic combinations in similar patient populations [24][25][26].Median PFS values reported with the chemotherapy-free treatment arms of sapanisertib and sapanisertib + TAK-117 were inferior to the treatment arms containing paclitaxel.
Although chemotherapeutic treatment options for advanced endometrial cancer remain limited, there have been significant advances in immunotherapy in this setting in the last five years.Pembrolizumab, a programmed death receptor-1 (PD-1)-blocking antibody, has been approved since 2020 as monotherapy for the treatment of patients with advanced endometrial carcinoma that is MSI-high or dMMR, who have progressed after prior systemic therapy, and are not candidates for curative surgery or radiation [27][28][29].Pembrolizumab in combination with lenvatinib has also been approved since 2020 for patients with endometrial cancer that is not MSI-high or dMMR, who have progressed after prior systemic therapy, and are not candidates for curative surgery or radiation [29,30].Furthermore, pembrolizumab in combination with carboplatin-paclitaxel (NRG-GY018) resulted in longer progression-free survival than with carboplatin-paclitaxel alone in patients with advanced or recurrent endometrial cancer [31].Another anti-PD-1 antibody, dostarlimab, has been approved since 2021 for the treatment of dMMR recurrent or advanced endometrial cancer that has progressed on or following a prior platinumcontaining regimen [32].Dostarlimab plus carboplatin-paclitaxel has recently been shown to improve PFS, compared to placebo, among patients with primary advanced or recurrent endometrial cancer, particularly in the dMMR and MSI-high population [33].As treatment for advanced and recurrent endometrial cancer shifts towards combined chemotherapy and immunotherapy, additional combination strategies, such as with mTOR inhibitors may be worth investigating.
Single-agent temsirolimus mTOR inhibition previously demonstrated modest activity in advanced/recurrent endometrial carcinoma, although durable disease stabilization was observed in some patients.In our study, the ORR was lower with paclitaxel alone (18.4%) versus paclitaxel + sapanisertib (24.4%) and less improvement was observed in CBR (57.5% vs 80.2%, respectively) and CBR-16 (36.8% vs 51.2%, respectively).The ORR of 24.4% for patients treated with paclitaxel + sapanisertib was consistent with the ORR of 18% reported in a phase 1 study of sapanisertib in combination with paclitaxel in patients with advanced solid tumors, and the CBR was considerably higher (80% vs 50%) [34].The ORR was also consistent with that reported with other mTOR inhibitors and immunotherapeutic combinations in similar pre-treated populations.
Subgroup analyses based on endometrioid histology were performed to assess differential response to paclitaxel versus paclitaxel + sapanisertib based on histologic subtype.The trend for improvement in PFS with the addition of sapanisertib to paclitaxel appeared to be driven by longer PFS in patients with endometrioid histology (HR 0.66; 95% CI 0.43-1.03)while there was no apparent benefit in patients with non-endometrioid histology (HR 1.09; 95% CI 0.62-1.90).Improvements in CBR and CBR-16 with paclitaxel + sapanisertib versus paclitaxel were also numerically greater in the endometrioid than non-endometrioid subgroup.Endometrioid histology is typically associated with a better prognosis and more mutations affecting the PI3K/AKT/mTOR pathway [9][10][11], which may explain why this subgroup was particularly responsive to inhibition of this pathway by sapanisertib.Molecular profiling in a subgroup of 67 patients in this study confirmed that mutations in PIK3CA and PTEN were more common in endometrioid than non-endometrioid tumors, but were not associated with improved PFS with paclitaxel + sapanisertib compared with paclitaxel in patients with endometrioid histology; however, mutations in the beta-catenin signaling pathway were associated with improved PFS with paclitaxel + sapanisertib compared with paclitaxel in patients with endometrioid histology.Further translational studies are needed to confirm these observations.
Patients with MSS status appeared to have a higher ORR versus patients with high MSI status overall, but there were no apparent differences according to whether patients received paclitaxel or paclitaxel in combination with sapanisertib.Although there were no treatment effects observed in this study according to MSS status, it is important to analyze clinical trial data according to different patient subgroups given the heterogeneity of this disease and the approval of therapies for specific subtypes.Immunotherapy has shown promise in patients with MSI-high endometrial cancer; single-agent pembrolizumab has recently received regulatory approval for patients with advanced endometrial cancer that is MSI-high or mismatch repair-deficient [35].However, there is still an unmet need given the modest activity observed in patients with MSS status (response rates ranging from 3%-23%) [36].The frequency of alterations in the PI3K/AKT/mTOR pathway across all histologies supports rational approaches to further develop this drug pathway.
The incidence of grade ≥ 3 TEAEs was higher with paclitaxel + sapanisertib versus paclitaxel, but toxicity with this combination was limited, with no new safety signals or notable differences in QoL.The most frequently reported grade ≥ 3 TEAEs included anemia, neutropenia, fatigue, and hypophosphatemia, which is consistent with the known safety profiles, and as expected for inhibition of the PI3K/AKT/mTOR pathway [23,34,37,38].
The primary endpoint of this study was not met and treatment with paclitaxel + sapanisertib did not improve outcomes.The trend for improvement in PFS of combination therapy was driven by outcomes in patients with endometrioid subtype tumors, who demonstrated greater numerical improvements in PFS, CBR, and CBR-16 with paclitaxel + sapanisertib versus paclitaxel alone than patients with non-endometrioid histology.The incidence of grade ≥ 3 TEAEs was higher in patients receiving paclitaxel + sapanisertib, but treatment with the combination had limited toxicity with no new safety signals.Given these encouraging preliminary data, future studies should investigate sapanisertib in combination with other agents in patients with endometrioid histology.
• There were no significant differences in PFS between the paclitaxel and paclitaxel + sapanisertib arms.

•
Treatment with paclitaxel + sapanisertib had limited toxicity with no new safety signals.Progression-free survival in the intent-to-treat population.
Table 1 Patient baseline demographics and disease characteristics.Gynecol Oncol.Author manuscript; available in PMC 2024 May 17.
Abbreviations: ECOG PS, Eastern Cooperative Oncology Group performance status; G, grade; ITT, intent-to-treat; NOS, not otherwise specified.aAt initial diagnosis.bAccording to patient's record prior to study entry.

Table 2
Confirmed overall response rate in the safety population.

Table 3
Overall response rate by histological subtype, PTEN mutation status and MSI status in the safety population.
b MSI/dMMR defined as IHC <5% tumor cells of any MMR proteins or nucleotide gain by PCR.Gynecol Oncol.Author manuscript; available in PMC 2024 May 17.

Table 4
Most frequent (≥20% any treatment arm) any grade treatment-emergent adverse events and most frequent (≥2% of patients overall) grade ≥ 3 treatment-emergent adverse events by preferred term in the safety population.