Efficacy of Lorlatinib After Failure of a First-Line ROS1 Tyrosine Kinase Inhibitor (ROS1 TKI) In Patients (Pts) With Advanced Ros1-Positive Non-Small Cell Lung Cancer (ROS1+ NSCLC) (IFCT-2003 ALBATROS)

Background

Current ESMO guidelines recommend ROS1 TKIs crizotinib and entrectinib as preferred first-line treatments for advanced ROS1+ NSCLC. Lorlatinib is an oral, reversible, ATP-competitive macrocyclic TKI targeting ROS1. This potent and highly selective inhibitor was designed to penetrate the blood-brain barrier.

Aim

To evaluate the efficacy of lorlatinib after failure of a 1L ROS1 TKI in 54 pts with advanced ROS1-positive NSCLC

Methods

ALBATROS is an open-label, single-arm phase II study. Advanced ROS1+ NSCLC pts progressing after one prior line of TKI with crizotinib or entrectinib (± one line of chemotherapy ± immunotherapy before TKI treatment), PS 0-2, and with a measurable disease according to RECIST 1.1 received lorlatinib 100 mg/day until progression or intolerable toxicity. Baseline brain MRI was required. Primary endpoint was the investigator-assessed ORR at 8 weeks, confirmed at 16 weeks. Secondary endpoints were the blinded independent central review (BICR)- assessed ORR at 8 weeks (confirmed at 16 weeks), DoR, DCR, PFS, OS, brain metastasis (BM) ORR, and safety.

Results

• Median follow-up was 31 mo. The median duration of treatment was 8.8 mo.
• Efficacy: Investigator-assessed confirmed ORR at 8 wk was 30.0% and DCR at 8 weeks was 84.0%. BICR-assessed confirmed ORR at 8 wk was 34.0%. The median DoR was 20.4 mo. Median PFS was 7.4 mo. Median OS was 42.3 mo. In the 13/31 pts with measurable BM, BM ORR was 92.3% and DCR was 100%. 
• Safety: Grade ≥ 3 treatment-related AE (TRAE) was observed in 45.3% of pts. 30.2 % of pts had at least 1 dose reduction (100% for AE). One pt permanently discontinued lorlatinib due to TRAE.

Conclusions

In this phase II trial, lorlatinib demonstrated robust clinical activity in ROS1+ NSCLC pts who received one line of ROS1+ TKI, mostly crizotinib. 

Reference 

Duruisseaux M et al, Efficacy of lorlatinib after failure of a first-line ROS1 tyrosine kinase inhibitor (ROS1 TKI) in patients (pts) with advanced ROS1-positive non-small cell lung cancer (ROS1+ NSCLC) (IFCT-2003 ALBATROS). Annals of Oncology, 1987MO, (2025).

Durvalumab + Paclitaxel/Carboplatin + Bevacizumab Followed By Durvalumab, Bevacizumab + Olaparib Maintenance in Patients with Newly Diagnosed Non-Tbrca-Mutated Advanced Ovarian Cancer: Final Overall Survival From Duo-O/Engot-Ov46/Gog-3025

Background

DUO-O showed statistically significant and clinically meaningful PFS benefit in both non-tBRCAm HRD+ and non-tBRCAm ITT pts treated with paclitaxel/carboplatin (PC) + bevacizumab (B) + durvalumab (D) followed by B + D + olaparib (O) maintenance vs PC + B at the planned interim analysis (first data cutoff [DCO1] 5 Dec 2022). 

Aim

To report final overall survival (OS), updated PFS, and safety from DUO-O/ENGOT-ov46/GOG-3025

Methods

Following 1 cycle of PC ± B, 1130 non-tBRCAm pts were randomised 1:1:1 at Cycle 2 to PC + B followed by B (Arm 1; control), PC + B + D followed by B + D (Arm 2), or PC + B + D followed by B + D + O (Arm 3). Final OS was formally tested (DCO3 17 Mar 2025) in non-tBRCAm ITT pts (Arm 3 vs Arm 1) per predefined multiple testing procedure. 

Results

• Median follow up:∼56 months.
• OS: In non-tBRCAm ITT pts, final OS (53% maturity) demonstrated no statistically significant differences when comparing Arm 3 vs Arm 1 or Arm 2 vs Arm 1 (Table). PFS HRs (73% maturity) in all treatment comparisons were consistent with DCO1. In non-tBRCAm HRD+ pts, OS HR (36% maturity) was 0.80 (95% CI 0.54–1.18) for Arm 3 vs Arm 1 and 0.84 (95% CI 0.57–1.23) for Arm 2 vs Arm 1. 
• PFS: PFS HRs (62% maturity) for Arm 3 vs Arm 1 and Arm 2 vs Arm 1 were consistent with DCO1. In non-tBRCAm HRD– (unstratified subgroup) pts, PFS and OS HRs were consistent with DCO1 in all treatment comparisons. 
• Safety: Safety was generally consistent with prior DCOs and the known profile of each agent.

Conclusions

Final OS for B + D + O (Arm 3) vs control (Arm 1) was not statistically significant in non-tBRCAm ITT pts, and the OS HR showed numerical improvement in non-tBRCAm HRD+ pts. In non-tBRCAm HRD+ pts, PC + B + D, followed by B + D + O maintenance, resulted in a median PFS of 45.1 months with 47% of pts progression-free at 48 months. 

Reference 

Aghajanian C et al. Durvalumab + paclitaxel/carboplatin + bevacizumab followed by durvalumab, bevacizumab + olaparib maintenance in patients with newly diagnosed non-tBRCA-mutated advanced ovarian cancer: Final overall survival from DUO-O/ENGOT-ov46/GOG-3025. Annals of Oncology, LBA44, (2025).

Single-Cell Atlas of Circulating Immunity Identifies Shared Specific DLBCL Signatures for Predicting Response to R-CHOP and Anti-CD19 CAR-T Therapies

Background

While the immunological heterogeneity of DLBCL involves both tumor microenvironment and peripheral immune systems, blood-based immune profiling to assess systemic status and predict treatment response remain lacking.

Aim

To prospectively collect longitudinal blood samples for cytokine quantification, to profile circulating immunity, develop a machine learning model for refractory distinction, and investigate treatment-related immune variations.

Methods

We prospectively collected longitudinal blood samples from a newly diagnosed DLBCL cohort, a R/R DLBCL cohort treated with anti-CD19 CAR-T (sampled up to 365 days) and healthy controls, using 22-marker single-cell spectral flow cytometry, algorithm-guided analysis and ELISA for cytokine quantification, to profile circulating immunity, develop a machine learning model for refractory distinction, and investigate treatment-related immune variations.

Results

• DLBCL patients exhibited peripheral immune subset dysregulation, characterized by reduced innate cells and T cell imbalance.
• Notably, two CXCR5+ T cell subsets and one CD4+ NKT subset were found to be very low in healthy controls but markedly increased in DLBCL. 
• Machine learning analysis showed a random forest-derived 3-feature signature (HLA-DRhi PD1hi CD127- TFH1, classical monocytes, Tregs) achieved high accuracy/F-measure for differentiating primary refractory DLBCL patients and responders, with subset kinetics correlating to treatment efficacy. 
• Furthermore, longitudinal CD19 CAR-T cohort analysis revealed shared and divergent patterns of effective immune homeostasis recovery related to longer PFS. 
• All patients exhibited sustained expansion of CXCR3+ CD4 Th1 cells from pre-lymphodepletion through day 365. Terminally differentiated CD8+ TEMRA cells reached their peak at day 90, after which a decline was observed. Pre-leukapheresis and D90 blood immunophenotypes (elevated HLA-DRhi PD1hi CCR7-CD127- TFH1, classical monocytes, Tregs, pDCs and reduced CCR6+CD4+ TEM, CD8 naïve T) correlated to poorer PFS, paralleling primary refractory cohort findings.

Conclusions

Our systemic analysis of peripheral immune signatures revealed refractory DLBCL patients may exhibit shared immune profiles. These new systemic immunity-based liquid biomarkers can be utilized to predict and monitor responses in a clinical context.

Reference 

Meng X et al, Single-cell atlas of circulating immunity identifies shared specific DLBCL signatures for predicting response to R-CHOP and anti-CD19 CAR T therapies. Annals of Oncology, 1246MO, (2025).

Trifluridine/Tipiracil in Combination with Capecitabine and Bevacizumab as Upfront Treatment for Metastatic Colorectal Cancer: First Results of the Phase II TriComB Study by GONO

Background

 Capecitabine (CAP) plus bevacizumab (BEV) is the standard 1L treatment for the 15–20% of patients with unresectable mCRC, deemed ineligible for combination chemotherapies. Preclinical studies have shown a synergistic effect from the sequential use of CAP and trifluridine/tipiracil (FTD/TPI). The phase I of the TRICOMB study established the recommended dose of FTD/TPI to be sequentially combined with CAP + BEV. 

Aim

To evaluate Trifluridine/tipiracil in combination with capecitabine and bevacizumab as upfront treatment for metastatic colorectal cancer

Methods

 In the phase II part of the TRICOMB (NCT04564898) study previously untreated mCRC patients ineligible for intensive therapy were included in 11 Italian centers. The 38 patients received CAP (1000 mg/sqm orally twice daily, days 1–14), followed by FTD/TPI (25 mg/sqm orally twice daily, days 15–19 and 22–26), combined with BEV (5 mg/kg IV on days 1 and 15) in 28-day cycles until disease progression. Primary endpoint was ORR per RECIST v1.1. Assuming an ORR of 25% and 45% as null and alternative hypothesis, respectively, with α and β errors of 0.05 and 0.20, 14 responses out of 36 enrolled patients were needed to deem the strategy promising.

Results

• Efficacy: Twenty pts achieved partial response (ORR: 52.6% [90% CI 38-67%]). The disease control rate was 84%. At a median follow-up of 11.6 months, 24 (63%) out of 38 pts experienced disease progression, with a median PFS of 10.3 months (95% CI, 7.7–14.2), and 14 (58%) received at least one subsequent line of therapy. Overall survival data are still immature.
• Safety: The most common grade 3/4 adverse events were neutropenia (38%), anemia (38%), thromboembolic events (16%), asthenia (11%), febrile neutropenia (5%), and diarrhea (5%).

Conclusions

 TRICOMB met its primary endpoint. The sequential combination of CAP, FTD/TPI, and BEV, according to the TRICOMB regimen, deserves further investigation as upfront therapy of mCRC.

Reference

Conca V et al. Trifluridine/tipiracil in combination with capecitabine and bevacizumab as upfront treatment for metastatic colorectal cancer: first results of the phase II TriComB study by GONO. Annals of Oncology, 730MO, (2025).

Enfortumab Vedotin plus Pembrolizumab (EV + P) as First-line (1L) Treatment in Recurrent or Metastatic Head and Neck Squamous Cell Carcinoma (R/M HNSCC): Results from a Cohort of the EV-202 Trial

Background

Pts with R/M HNSCC have a poor prognosis despite treatment advancements. P +/- chemotherapy is the SoC for R/M HNSCC. EV, a nectin-4-directed ADC, demonstrated clinically meaningful activity as monotherapy in pts with heavily pretreated R/M HNSCC, with a confirmed ORR of 23.9% and median DOR not reached (NR) after a median follow-up (f/u) of 9.3 months. EV-202 (NCT04225117) is an international phase 2 basket study evaluating the efficacy and safety of EV +/- P in various solid tumors. Here, we present primary results from cohort 9 (1L EV + P in R/M HNSCC).

Aim

To evaluate the efficacy and safety of EV +/- P in R/M HNSCC as a 1L treatment

Methods

Pts with R/M HNSCC, PD-L1 combined positive score (CPS) ≥1, ECOG PS 0–1 and no prior systemic therapy in R/M setting (systemic therapy completed >6 mo prior for locoregional disease allowed) were eligible. Pts received EV (1.25 mg/kg IV) on days 1, 8 and P (200 mg IV) on day 1 in 21-day cycles. Primary endpoint: investigator-assessed cORR (RECIST v1.1). Secondary endpoints: DOR, DCR, PFS, OS and safety.

Results

• Median f/u was 11.0 mo (95% CI: 9.5, 12.1)
•  Median number of treatment cycles was 6.0 for EV and 6.0 for P
• Efficacy: cORR was 39.0% (95% CI: 24.2, 55.5) overall; 43.8% in PD-L1 CPS 1-19 and 36% in CPS ≥20. Complete response rate was 9.8%; DCR was 75.6% (95% CI: 59.7, 87.6); median DOR was NR; DOR rate at 6 mo was 81.7% (95% CI: 42.0, 95.4); median PFS was 5.1 mo (95% CI: 3.5, NE); and median OS is not yet mature. 
• Safety: Grade ≥3 treatment-emergent adverse events occurred in 70.7% of pts; most commonly (>5%) fatigue (9.8%) and acute respiratory failure, dehydration, dysphagia, maculopapular rash, syncope and tumor bleeding (all 7.3%).

Conclusions

 EV + P demonstrates promising clinical activity in 1L pts with PD-L1 CPS ≥1 R/M HNSCC with an encouraging antitumor response rate that met the protocol-defined efficacy threshold and was often durable. Safety profile was consistent with previous reports, reinforcing the manageable tolerability profile of EV + P.

Reference 

Swiecicki PL et al. Enfortumab vedotin plus pembrolizumab (EV + P) as first-line (1L) treatment in recurrent or metastatic head and neck squamous cell carcinoma (R/M HNSCC): Results from a cohort of the EV-202 trial. Annals of Oncology, 1329MO, (2025).

IMbrave152/SKYSCRAPER-14: A Phase 3 Study Of First-Line Tiragolumab (Tira) + Atezolizumab (Atezo) + Bevacizumab (Bev) Vs Placebo (Pbo) + Atezo + Bev For Patients (Pts) With Untreated Locally Advanced Or Metastatic Hepatocellular Carcinoma (HCC)

Background

 Atezo + bev is a standard first-line therapy for advanced HCC; however, not all pts experience clinical benefit and novel treatment options are needed. 

Aim

 To evaluate the addition of tira (an anti-TIGIT antibody) to atezo + bev in pts with advanced HCC.

Methods

 Eligible pts with treatment-naïve, unresectable locally advanced or metastatic HCC were randomised 1:1 to receive tira 600 mg + atezo 1200 mg + bev 15 mg/kg or pbo + atezo 1200 mg + bev 15 mg/kg via intravenous infusion every 3 weeks until loss of clinical benefit or unacceptable toxicity. Dual primary endpoints were investigator-assessed progression-free survival (INV-PFS) and OS. Secondary endpoints included ORR, DOR, and safety.

Results

• Median follow-up: 12.5 months
• Median INV-PFS was 8.3 months for tira + atezo + bev and 8.2 months for pbo + atezo + bev (HR 0.97)
• Median OS data were immature (HR 0.94)
• ORR was 29.9% for tira + atezo + bev and 26.0% for pbo + atezo + bev. 
• MDoR: 15 vs 13.2 mo
• G3–5 TRAEs were reported in 147 pts (44.3%) in the tira + atezo + bev arm and 122 pts (36.6%) in the pbo + atezo + bev arm. Overall, safety results were similar in both treatment arms and generally consistent with previous data from IMbrave150

Conclusions

 The IMbrave152/SKYSCRAPER-14 study did not meet its primary endpoint of INV-PFS; combining tira with atezo + bev did not show an added benefit in pts with advanced HCC. OS data are not expected to reach statistical significance. The study has been unblinded and long-term survival follow-up is ongoing.

Reference 

Finn RS et al, IMbrave152/SKYSCRAPER-14: A phase III study of first-line tiragolumab (tira) + atezolizumab (atezo) + bevacizumab (bev) vs placebo (pbo) + atezo + bev for patients (pts) with untreated locally advanced or metastatic hepatocellular carcinoma (HCC). Annals of Oncology, LBA50, (2025).

Primary Results From ASCENT-03: A Randomized Phase 3 Study of Sacituzumab Govitecan (SG) vs Chemotherapy (Chemo) in Patients (pts) With Previously Untreated Advanced Triple-Negative Breast Cancer (TNBC) Who Are Unable to Receive PD-(L)1 Inhibitors (PD-[L]1i)

Background

Significant PFS benefit was observed with SG vs chemo in pretreated metastatic (m)TNBC (ASCENT) and with SG + pembrolizumab vs chemo + pembrolizumab in first line (1L) PD-L1+ mTNBC (ASCENT-04). For pts with mTNBC who cannot receive PD-(L)1i, treatment options are limited. 

Aim

To evaluate 1L SG vs chemo in pts with locally advanced unresectable or mTNBC who are unable to receive a PD-(L)1i

Methods

Pts had centrally confirmed PD-L1− mTNBC (defined as combined positive score [CPS] < 10) or PD-L1+ mTNBC (CPS ≥ 10) but were unable to receive PD-(L)1i due to a comorbidity or prior use in the curative setting. Randomization (1:1) of 558 pts to SG (10 mg/kg IV, days 1 & 8 in 21-day cycles) or chemo (paclitaxel, nab-paclitaxel, or gemcitabine + carboplatin) was stratified by disease status and geography. The primary end point was PFS by BICR. Key secondary end points included overall survival (OS), ORR and DOR by BICR, and safety.

Results

• Median follow-up of 13.2 mo
• mPFS: SG showed a significant improvement in median PFS vs chemo (9.7 vs 6.9 mo; HR, 0.62; 95% CI, 0.50-0.78; P < .0001)
• mDOR: median DOR was 12.2 mo vs 7.2 mo 
• OS: OS data were immature
• Safety: The most frequent grade ≥ 3 TEAEs were neutropenia (43%) and diarrhea (9%) with SG and neutropenia (41%) and anemia (16%) with chemo.

Conclusions

SG led to a statistically significant and clinically meaningful improvement in PFS and more durable responses vs chemo in 1L mTNBC. The safety profile of SG was manageable and consistent with its known profile; treatment discontinuation rate due to TEAEs was lower with SG vs chemo. These data support SG as a potential new standard of care for pts with previously untreated mTNBC who are unable to receive a PD-(L)1i. 

Reference 

Cortés JC et al, Primary results from ASCENT-03: A randomized phase III study of sacituzumab govitecan (SG) vs chemotherapy (chemo) in patients (pts) with previously untreated advanced triple-negative breast cancer (TNBC) who are unable to receive PD-(L)1 inhibitors (PD-[L]1i). Annals of Oncology, LBA20, (2025).

Randomised Phase 3 Trial Of Androgen Deprivation Therapy (ADT) with Radiation Therapy With Or Without Enzalutamide for High Risk, Clinically Localised Prostate Cancer: ENZARAD (ANZUP 1303)

Background

 Adjuvant ADT with an LHRH analog (LHRHA) given before, during and after radiotherapy (RT) is standard of care for high risk localised prostate cancer (PC). Enzalutamide improves overall survival (OS) in castration-resistant, metastatic prostate cancer. 

Aim

To determine the effects of enzalutamide versus a conventional non-steroidal anti-androgen (NSAA) as part of neoadjuvant and adjuvant ADT in men undergoing RT for high risk, localized PC.

Methods

The 802 participants (pts) having 24mo ADT and RT for high-risk L/LAPC were randomized to ENZA 160mg daily for 24 mo vs a non-steroidal anti-androgen (NSAA) for 6 mo with stratification across 7 clinico-pathological factors. Metastasis-free survival (MFS) on conventional imaging (ICECaP) was the primary endpoint. Secondary endpoints included OS, PC-specific survival (PCSS), PSA progression-free survival (PSA-PFS), clinical-PFS, and adverse events (AE). 200 MFS events provided 80% power assuming a true MFS hazard ratio (HR) of 0.67 and 2-sided alpha of 0.05. Main effects were tested with unstratified logrank p-values. Five subgroups of interest were prespecified and subgroup effects were tested with interaction p-values (i-p).

Results 

• Median age 71; 89% Gleason 8-10; 47% T3-4; 35% PSA≥20; and 11% N1. 
• 207 MFS events occurred during a median follow-up of 92 mo.
•  ENZA and NSAA had similar effects on MFS, OS, and PCSS; ENZA had larger effects on PSA-PFS and clinical-PFS.
• Outcome at 8 years: ENZAN vs NSAAN 
  1. MFS: 74% vs 72% (HR 0.88)
  2. OS: 83% vs 80% (HR 0.87)
  3. PCSS: 97% vs 96% (HR 0.90)
  4. PSA-PFS: 67% vs 62% (HR 0.78)
  5. Clinical PFS: 72% vs 66% (HR 0.76)

Conclusions

 The addition of ENZA to ADT and RT had limited effects on MFS during 8 years of follow-up in high-risk L/LAPC. Effects on MFS were larger in those with positive pelvic nodes on conventional imaging, and in those with pelvic RT planned. 

Reference 

Nguyen PL et al, Randomised phase III trial of androgen deprivation therapy (ADT) with radiation therapy with or without enzalutamide for high risk, clinically localised prostate cancer: ENZARAD (ANZUP 1303). Annals of oncology, LBA86, (2025).

EMBARK: Overall Survival with Enzalutamide In Biochemically Recurrent Prostate Cancer

Background

The phase 3 EMBARK trial (NCT02319837) showed significant improvements in metastasis-free survival (MFS) and secondary efficacy endpoints with enzalutamide plus leuprolide (enza combo) and enzalutamide monotherapy (enza mono) vs leuprolide alone (LA). 

Aim

To show final overall survival (OS) data and other secondary efficacy outcomes from EMBARK.

Methods

Patients (pts) with high-risk biochemically recurrent prostate cancer (hrBCR), defined as a prostate-specific antigen (PSA) doubling time of ≤9 months, were randomised 1:1:1 to receive enza combo, LA, or enza mono. The primary endpoint was MFS for enza combo vs LA. OS, the time between randomisation and death from any cause, was an alpha-protected (final adjusted 2-sided alpha level of 0.0499) key secondary endpoint. Other secondary endpoints included MFS for enza mono vs LA, time to PSA progression, time to first use of new antineoplastic therapy, and time to first symptomatic skeletal event (SSE). Progression-free survival on first subsequent therapy (PFS2) was an exploratory endpoint. Survival data were collected every 12 weeks until final OS analysis, planned after 271 events. At final data cutoff (27 May, 2025), OS was compared between treatment groups with a 2-sided stratified log-rank test. Hazard ratios (HRs) were estimated by stratified Cox models. Pts without an OS event were censored at last contact.

Results 

• Enza combo reduced the risk of death by 40.3% vs LA (HR 0.597; 95% CI 0.444–0.804; P=0.0006). 
• Enza mono reduced the risk of death by 17.0% vs LA, which did not reach statistical significance (HR 0.830; 95% CI 0.630–1.095; P=0.1867). 
• Enza combo and enza mono both significantly prolonged time to first use of new antineoplastic therapy, time to first SSE, and PFS2.
•  Safety findings were consistent with prior publications.

Conclusions

Enza combo reduced the risk of death vs LA by over 40%. This unprecedented survival advantage reinforces the MFS results and further supports enza combo as the standard of care for pts with hrBCR. Table: LBA87

Reference 

Freedland S et al, EMBARK: Overall survival with enzalutamide in biochemically recurrent prostate cancer. Annals of Oncology, LBA87, (2025).

A Phase III Study of Capivasertib (Capi) + Abiraterone (Abi) Vs Placebo (Pbo) + Abi In Patients (Pts) With Pten Deficient De Novo Metastatic Hormone-Sensitive Prostate Cancer (mHSPC): CAPItello-281

Background

In HSPC tumours, PTEN deficiency is associated with dysregulation of PI3K/AKT and androgen receptor (AR) signalling resulting in reduced benefit from AR pathway inhibition and shorter time to progression. An unmet need exists for a targeted treatment for PTEN deficient mHSPC. CAPItello-281 (NCT04493853) is a randomised, double-blind, phase 3 study evaluating capi + abi in pts with PTEN deficient de novo mHSPC.

Aim

To evaluate capi + abi in pts with PTEN deficient de novo mHSPC.

Methods

The 1012 Pts aged ≥18 years with PTEN deficient tumours (defined as ≥90% of viable malignant cells with no specific cytoplasmic staining by immunohistochemistry) and ECOG 0-1 received capi or matched pbo (1:1) in combination with abi + prednisone/prednisolone and androgen deprivation therapy. Primary endpoint was investigator-assessed radiographic progression-free survival (rPFS); overall survival (OS) was a key secondary endpoint. rPFS in post hoc exploratory subgroups at more stringent PTEN cutoffs was also assessed.

Results 

• Statistically significant improvement in rPFS was observed with capi + abi vs pbo + abi (HR 0.81, P=0.034). 
• rPFS HRs at increasing PTEN deficiency cutoffs (≥95%, n=814 to 100%, n=331) ranged from 0.75 to 0.68. 
• rPFS: 33.2 vs 25.7 mo (HR 0.81; p= 0.034)
• OS: NC vs NC (HR 0.90; p = 0.401) [NC, not calculable]
• Interim OS analysis numerically favoured capi + abi. 
• 18.3% of pts discontinued capi due to an AE vs 4.8% pbo (9.5 vs 5.4% abi). 

Conclusions

CAPItello-281 met its primary objective showing a statistically significant rPFS benefit with capi + abi vs pbo + abi in pts with PTEN deficient de novo mHSPC. Increasing benefit was observed in subgroups selected using progressively more stringent PTEN cutoffs. The safety profile was broadly consistent with the known profiles of capi and abi. Capi in combination with abi represents a potential first-in-class targeted treatment for this poor prognosis population with high unmet need.

Reference 

Fizazi K et al, A phase III study of capivasertib (capi) + abiraterone (abi) vs placebo (pbo) + abi in patients (pts) with PTEN deficient de novo metastatic hormone-sensitive prostate cancer (mHSPC): CAPItello-281. Annals of Oncology, 2383O, (2025).

ICON8B: GCIG Phase III Randomised Trial Comparing First-Line Weekly Dose-Dense Chemotherapy + Bevacizumab to Three-Weekly Chemotherapy + Bevacizumab in High-Risk Stage III-IV Epithelial Ovarian Cancer (EOC): Final Overall Survival (OS) Analysis

Background

In ICON8B the use of dose-dense weekly paclitaxel (ddwT) with 3-weekly (q3w) carboplatin (C) and bevacizumab (BEV) as first-line treatment improved median progression-free survival (PFS) by 5.5 months (m) compared to standard q3w paclitaxel (T) dosing with C+BEV (22.2m vs 16.7m; Hazard Ratio (HR) 0.75, 95% CI 0.62-0.90 p=0.002). We now report the final OS analysis conducted at trial closure.

Aim

To report final OS analysis conducted at ICON8B trial closure.

Methods

The 579 eligible participants (pts) with high-risk stage III (residual disease >1cm diameter after immediate primary surgery (IPS) or requirement for primary chemotherapy) or stage IV EOC were randomised 1:1:1 to Arm B1 (standard- q3w C AUC5/6+q3w T 175mg/m2+ q3w BEV 7.5mg/kg); Arm B2- (q3w C AUC5/6+ddwT 80mg/m2); Arm B3- (q3w C AUC5/6+ddwT 80mg/m2+ q3w BEV 7.5mg/kg). Up to six cycles chemotherapy and 18 BEV cycles were administered. Arm B2 recruitment discontinued after ICON8 saw no evidence of PFS improvement with q3wCddwT vs q3wCT. OS was a key secondary outcome and pts were followed for survival endpoints until trial closure on 18th Dec 2024 at end of academic funding.

Results 

• After a median follow-up of 72.0m, 411 deaths were reported (197 in B3; 214 in B1).
• Median OS was 49.8m (95% CI 43·7-54.5m) in B3 and 39.6m (95% CI 34·7-45·0m) in B1 (HR 0·79, 95% CI 0·65-0·95, p=0·010).
• In pts receiving primary chemotherapy, median OS was 47.3m (95% CI 42.0-52.6m) in B3 and 37.1m (95% CI 32.3-42.1m) in B1.

Conclusions

In pts with high-risk stage III-IV EOC, the use of ddwT in combination with q3w C + BEV as first-line systemic therapy improves median OS by 10.2m compared to q3w T dosing. ddwT with q3wC+BEV should now be considered a standard-of-care first-line treatment option in this group. Further research is required to determine whether efficacy of this regimen is impacted by tumour homologous recombination deficiency and intrinsic chemosensitivity.

Reference 

Clamp AR et al. GCIG phase III randomised trial comparing first-line weekly dose-dense chemotherapy + bevacizumab to three-weekly chemotherapy + bevacizumab in high-risk stage III-IV epithelial ovarian cancer (EOC): Final overall survival (OS) analysis. Annals of Oncology, 1064O, (2025).