Annals of Oncology最新文献

Background: Intrahepatic cholangiocarcinoma (ICCA) is a rare but highly lethal adenocarcinoma arising within the hepatic parenchyma. Diagnosis presents a significant clinical challenge as the histological features of ICCA substantially overlap with those of metastatic liver cancers. This diagnostic ambiguity often necessitates extensive and costly exclusionary investigations, such as upper and lower gastrointestinal endoscopy, to rule out an occult primary site. Consequently, this process results in treatment delays and an increased financial burden on healthcare systems.

Patients and methods: We retrospectively analyzed 544 patients across five European centers, comprising cases of either ICCA or metastases from extrahepatic cancers. Three deep-learning architectures utilizing foundation models were investigated: Ctranspath/HistoBistro, UNI/CLAM, and CONCH/TITAN. Performance was assessed using the Area Under the Receiver Operating Characteristic curve (AUROC) and the False Positive Rate (FPR). Furthermore, we implemented a confidence estimation system using the Generalized-ODIN (G-ODIN) approach, utilizing predictive entropy as a metric. The final model, designated AI2CCA, was prospectively validated in 161 patients across four international centers in France, India, and Korea.

Results: In the retrospective test set, the CONCH/TITAN architecture yielded the best performance (AUROC: 0.840). Predictive entropy derived via G-ODIN was significantly higher in misclassified cases, validating its utility as a confidence metric. Implementation of confidence thresholding improved the AUROC to 0.958 with an FPR of 0, while retaining 46% of samples for high-confidence prediction. In prospective validation, AI2CCA achieved AUROCs of 1.00 and 0.965 in the French and Asian cohorts, respectively (with only one misclassified case in the Asian series).

Conclusion: Collectively, our study demonstrates the real-world clinical utility of a confidence-based AI biomarker for assisting in the diagnosis of liver cancer. By accurately discriminating ICCA from metastasis, this tool offers the potential to reduce unnecessary investigations and accelerate therapeutic decision-making.

Background: Nivolumab plus ipilimumab (NIVO+IPI) demonstrated significant long-term survival and response benefits in patients with previously untreated advanced renal cell carcinoma (aRCC) in the phase III CheckMate 214 trial (NCT02231749). We report final efficacy and safety results with 9.3 years median follow-up.

Patients and methods: Patients (N=1096) were randomized to NIVO 3 mg/kg plus IPI 1 mg/kg Q3W × four doses, followed by NIVO (3 mg/kg or 240 mg Q2W or 480 mg Q4W); or sunitinib (SUN) (50 mg) QD (4 weeks on, 2 weeks off).

Endpoints: overall survival (OS), and independent radiology review committee-assessed progression-free survival (PFS) and objective response rate (ORR) in IMDC intermediate/poor-risk (primary), intention-to-treat (secondary), and IMDC favorable-risk (exploratory) patients.

Results: With a median (range) follow-up of 9.3 (8.6-9.9) years, the hazard ratio (HR; 95% confidence interval [CI]) for OS with NIVO+IPI versus SUN was 0.71 (0.62-0.82) in intention-to-treat patients, 0.69 (0.59-0.81) in intermediate/poor-risk patients, and 0.80 (0.59-1.09) in favorable-risk patients; 108-month OS probabilities were 31.4% versus 19.5%, 30.2% versus 18.7%, and 35.3% versus 21.8%, respectively. PFS probabilities at 96 months were 22.7% versus 9.0% (intention-to-treat), 25.4% versus 8.5% (intermediate/poor-risk), and 12.5% versus 11.3% (favorable-risk). Probabilities of remaining in response at 96 months with NIVO+IPI versus SUN were 48.0% versus 19.0% (intention-to-treat), 50.0% versus 23.0% (intermediate/poor-risk), and 36.0% versus not estimable (favorable-risk). Incidence of any-grade (grade 3-4) treatment-related adverse events (AEs) was 94.1% (48.6%) with NIVO+IPI versus 97.6% (64.1%) with SUN. Exploratory post hoc analyses reported include descriptive analyses of OS by immune-mediated AE discontinuation status.

Conclusions: In the longest phase III follow-up of a first-line checkpoint inhibitor combination in aRCC (>9 years), NIVO+IPI maintained a substantial survival benefit with durable responses versus SUN. Grade 3-4 treatment-related AEs were lower with NIVO+IPI versus SUN at 9 years. NIVO+IPI remains a first-line standard of care in aRCC.

Background: ORCHARD (NCT03944772) was a phase II platform study conducted to characterize resistance mechanisms and evaluate novel treatment combinations following progressive disease (PD) on first-line osimertinib. Datopotamab deruxtecan (Dato-DXd) is an anti-TROP2 antibody-drug conjugate approved as monotherapy in EGFR-mutated advanced non-small-cell lung cancer (NSCLC). We report final data from the osimertinib plus Dato-DXd module.

Methods: Eligible patients had EGFR-mutated advanced NSCLC and PD on first-line osimertinib. Patients received oral osimertinib (80 mg once daily) plus intravenous Dato-DXd (4 or 6 mg/kg every 3 weeks). The primary endpoint was investigator-assessed confirmed objective response rate (ORR) per RECIST v1.1. Secondary endpoints were progression-free survival (PFS), duration of response (DoR), overall survival (OS), and safety.

Results: Sixty-nine patients received study treatment. Among patients in the 4 mg/kg (N = 35) and 6 mg/kg (N = 34) cohorts, respectively: confirmed ORR was 43% (80% confidence interval [CI] 32-55) and 36% (80% CI 25-49); median PFS was 9.5 (95% CI 7.2-9.8) and 11.7 (95% CI 8.3-21.7) months; median DoR was 6.3 (95% CI 3.8-8.1) and 20.5 (95% CI 6.2-not calculable [NC]; estimated median of at least 16) months; median OS was 19.8 (95% CI 13.5-23.3) and 26.2 (95% CI 14.8-NC; estimated median greater than or equal to the 4 mg/kg cohort) months. In the 4 mg/kg and 6 mg/kg cohorts, respectively, grade ≥3 adverse events (AEs) were reported in 49% and 76% of patients; AEs leading to Dato-DXd dose reduction in 23% and 59%; and adjudicated interstitial lung disease/pneumonitis in 3% and 15%.

Conclusions: Osimertinib plus Dato-DXd demonstrated clinical benefit in patients with EGFR-mutated advanced NSCLC who progressed on first-line osimertinib. AEs in the 6 mg/kg cohort were of higher frequency and severity but could be managed with prophylaxis, careful monitoring and dose reduction. The safety profile was consistent with the known profiles of the individual drugs. Considering the overall benefit-risk profile, 6 mg/kg is suggested as the preferred Dato-DXd starting dose for combining with osimertinib 80 mg.

Background: In DESTINY-Breast06, trastuzumab deruxtecan (T-DXd) improved progression-free survival (PFS) versus physician's choice of chemotherapy (TPC) for patients with hormone receptor-positive (HR+), HER2-low/-ultralow metastatic breast cancer (mBC), without new safety signals. We report additional analyses exploring outcomes for patients with characteristics that affect prognosis.

Patients and methods: Patients were randomized 1:1 to receive T-DXd (5.4 mg/kg) once every 3 weeks or TPC. Subgroups were specified post hoc from the intent-to-treat population (N = 866). Outcomes were PFS, objective response rate (ORR), and duration of response (DOR) by blinded independent central review via Response Evaluation Criteria in Solid Tumours 1.1. Time from randomization until second progression or death (PFS2) by investigator and safety were also assessed.

Results: Within subgroups, baseline disease characteristics and prior therapies were balanced across treatments. Median PFS (95% CI) favored T-DXd versus TPC regardless of time to progression (TTP) on prior first-line endocrine therapy (ET) + CDK4/6 inhibitor (CDK4/6i): <6 months: 14.0 (11.2-15.9) versus 6.5 (4.2-8.4) months for T-DXd versus TPC; 6-12 months: 13.2 (8.6-16.4) versus 6.9 (4.3-10.4) months; >12 months: 12.9 (9.8-17.1) versus 8.2 (6.9-10.9) months. A consistent PFS benefit with T-DXd over TPC was observed for patients with primary or secondary endocrine resistance, and across indicators of high or low disease burden (defined as visceral/non-visceral disease, presence/absence of liver metastases, ≥3/<3 disease sites, and >median/≤median baseline tumor size). In all subgroups, confirmed ORR favored T-DXd (36.7%-67.7%) versus TPC (16.7%-37.5%), and median DOR (confirmed response) was longer with T-DXd. T-DXd also prolonged PFS2 versus TPC across subgroups. Safety profiles of T-DXd and TPC in subgroups were consistent with the overall safety population.

Conclusions: These data support T-DXd as a broadly efficacious treatment for patients with HR+, HER2-low/-ultralow mBC after ≥1 ET, regardless of TTP on prior first-line ET + CDK4/6i, endocrine resistance, or disease burden.