Month in Review
October's major thoracic oncology developments were concentrated at ESMO 2026, which we covered separately in our dedicated conference edition.
Outside the congress, several publications addressed practical questions that receive less attention: whether radiotherapy intensification improves outcomes in stage III NSCLC, whether serial ctDNA monitoring provides clinical benefit, and how treatment might be individualized for older or vulnerable patients.
Two findings deserve particular attention. PET-BOOST did not demonstrate improved locoregional control with PET-guided radiation dose escalation, while R-ALPS found no significant benefit from adding anlotinib to consolidation immunotherapy, despite increased severe toxicity.
Smaller studies explored the clinical value of longitudinal molecular profiling and alternative ROS1 inhibition after resistance to newer targeted therapies.
Core Studies
Early-Stage NSCLC
KEYNOTE-867 — Stereotactic Body Radiotherapy With Pembrolizumab for Unresected Stage I–II NSCLC
Journal of Clinical Oncology
Study: Randomized, double-blind phase III trial comparing SBRT plus pembrolizumab with SBRT plus placebo in medically inoperable or unresected stage I–II NSCLC.
Key data: Among 448 randomized patients, median EFS was 31.2 vs 28.3 months; HR 0.92 (95% CI 0.69–1.24). Grade 3–5 treatment-related adverse events occurred in 20.4% vs 3.7%, with five treatment-related deaths (2.2%), all in the pembrolizumab arm.
Why it matters: Adding pembrolizumab to SBRT did not improve disease outcomes and substantially increased toxicity. Together with the negative SWOG/NRG S1914 trial, these findings argue against routine immunotherapy intensification alongside SBRT in early-stage NSCLC.
Clinical context & caveats: The study was stopped early following the prespecified interim analysis. OS was not formally tested because the primary EFS endpoint was negative. Whether selected high-risk populations might benefit remains unresolved.
Locally Advanced NSCLC
R-ALPS — Benmelstobart With or Without Anlotinib After Chemoradiotherapy in Unresectable NSCLC
JAMA Oncology
Study: Randomized, double-blind phase III trial evaluating benmelstobart, a PD-L1 inhibitor, with or without the multikinase inhibitor anlotinib following chemoradiotherapy for unresectable stage III NSCLC.
Key data: Among 553 randomized patients across two study stages, benmelstobart with or without anlotinib improved PFS compared with placebo in the initial three-arm comparison. However, in the pooled direct comparison, median PFS was 17.4 vs 11.2 months with vs without anlotinib; HR 0.82 (95% CI 0.63–1.08). Grade 5 treatment-related adverse events occurred in 7.2% vs 1.4% with combination treatment versus benmelstobart alone.
Why it matters: The trial addresses whether adding antiangiogenic targeted therapy can improve outcomes beyond consolidation immunotherapy. The additional toxicity without a demonstrated PFS benefit argues against routine treatment intensification with anlotinib.
Clinical context & caveats: Conducted exclusively in China. Benmelstobart was compared with placebo rather than durvalumab, the established PACIFIC consolidation standard. The positive immunotherapy-versus-placebo findings therefore do not establish a new preferred PD-L1 inhibitor.
PET-BOOST — PET-Guided Radiation Dose Escalation in Stage III NSCLC
International Journal of Radiation Oncology, Biology, Physics
Study: Multicenter randomized phase II trial comparing standard chemoradiotherapy (60 Gy in 30 fractions) with selective FDG-PET-guided dose escalation to metabolically active tumor subvolumes, up to 85 Gy in 30 fractions, in unresectable stage III NSCLC.
Key data: Among 78 randomized patients, two-year locoregional failure was 18.4% with dose escalation vs 27.5% with standard radiotherapy (P=0.13). Two-year OS was 73.7% versus 72.5% (P=0.91), and PFS was 57.9% versus 52.5% (P=0.74). Grade ≥3 treatment-related toxicity occurred in 37% vs 25%, including one fatal pulmonary hemorrhage in the dose-escalation arm. Overall, 76% received consolidation durvalumab.
Why it matters: Following the negative results of conventional radiation dose escalation, PET-guided selective boosting was proposed as a way to improve local control while limiting toxicity. This trial did not establish a clinical advantage.
Clinical context & caveats: The study was small, with fewer locoregional failures than anticipated, limiting statistical power. A numerical reduction in failures should not be interpreted as proven benefit. The findings do not support routine PET-guided dose escalation outside clinical trials.
Other Notable Publications
Advanced NSCLC / Liquid Biopsy
COPE — Serial Circulating Tumor DNA Monitoring in First-Line Advanced NSCLC
Clinical Cancer Research
Study: Multicenter randomized, noncomparative phase II study evaluating serial plasma ctDNA profiling with molecular tumor board review vs standard management during first-line treatment for advanced NSCLC.
Key data: Among 176 patients, 117 underwent serial ctDNA monitoring and 59 received standard management. The prespecified primary analysis compared 18-month OS in the monitoring arm against a historical 40% benchmark; observed OS was 53.8% (95% CI 44.3–62.5). Baseline plasma testing increased successful molecular profiling from 69% to 97% among patients with available tissue/plasma assessment. In patients with paired samples, early ctDNA clearance was associated with higher response rates (83% vs 52%) and longer PFS (22.9 vs 5.4 months).
Why it matters: Plasma profiling substantially increased the proportion of patients with successful molecular characterization, while early ctDNA changes showed promising associations with treatment response.
Clinical context & caveats: Although randomized, the study was explicitly noncomparative, with no planned between-arm efficacy test. Serial monitoring did not demonstrate improved clinical outcomes, and the associations between ctDNA clearance and survival cannot establish the value of changing treatment based on molecular response alone.
DURATION — Modified CARG Score–Guided Treatment in Older or Vulnerable Patients With Advanced NSCLC
Clinical Lung Cancer
Study: Randomized phase II trial in older or vulnerable patients with advanced NSCLC. A modified Cancer and Aging Research Group (mCARG) score determined eligibility for platinum-doublet versus single-agent chemotherapy. Within each risk stratum, patients were randomized to four chemotherapy cycles or two cycles followed by durvalumab.
Key data: Among 201 enrolled patients, median age was 76 years and 30% had ECOG PS ≥2. There was no significant difference in grade 3/4 treatment-related adverse events between the chemotherapy and sequential durvalumab strategies (P=0.81). Patients with mCARG ≤3 had longer median OS than those with mCARG >3 (10.2 vs 5.3 months; HR 0.48) and longer PFS (6.1 vs 3.2 months; HR 0.54).
Why it matters: Older patients and those with comorbidities remain underrepresented in treatment trials. DURATION prospectively explored whether a structured vulnerability assessment could guide treatment intensity.
Clinical context & caveats: The survival differences between mCARG strata are primarily prognostic, not evidence that score-guided treatment improves outcomes: patients were not randomized between platinum-doublet and single-agent chemotherapy. The chemotherapy-first strategy also differs from contemporary chemoimmunotherapy practice.
Targeted Therapy
Type II ROS1 Inhibition After Resistance to Type I TKIs — A Phase II Study of Cabozantinib
Clinical Cancer Research
Study: Single-center phase II proof-of-concept study evaluating cabozantinib, a type II ROS1/MET inhibitor, in heavily pretreated ROS1 fusion-positive NSCLC following progression on type I ROS1 inhibitors.
Key data: Among seven prospectively treated patients, ORR was 43% (3/7; 95% CI 10–81.5%). Responses included tumors harboring ROS1 D2033N and G2032R/L2086F resistance mutations. Median PFS was 4.5 months, median duration of response was 8 months, and median OS was 11 months. Grade 3 treatment-related adverse events occurred in 57%, and every patient required dose reduction.
Why it matters: Cabozantinib's different kinase-binding mechanism provides a potential strategy for overcoming resistance mutations that limit currently approved ROS1 inhibitors.
Clinical context & caveats: These results are highly preliminary. The cohort closed after the first stage of its planned two-stage design, and the sample size is too small for reliable efficacy estimates. Substantial toxicity and uncertain comparative benefit prevent routine treatment recommendations, but the findings support further investigation of type II ROS1 inhibitors.
Expert Takeaways
Treatment intensification requires evidence of benefit. R-ALPS and PET-BOOST illustrate why plausible combinations or more aggressive local therapy should not be adopted on the basis of biological rationale or numerical trends alone.
ctDNA is useful for molecular characterization, but monitoring is a different question. COPE supports plasma genotyping when tissue information is inadequate; it does not establish that serial molecular monitoring improves survival.
Vulnerability assessment remains important. DURATION supports risk stratification in older patients but does not establish an optimal treatment regimen.
Resistance mechanisms may open new therapeutic approaches. Type II ROS1 inhibition is promising in selected resistant tumors, although clinical evidence remains extremely limited.
Notes on Generalizability & Bias Signals
R-ALPS was conducted exclusively in China and used benmelstobart rather than durvalumab, limiting direct applicability to current European practice.
PET-BOOST, DURATION, and cabozantinib were relatively small phase II studies, limiting statistical power and the reliability of treatment-effect estimates.
COPE and DURATION illustrate important methodological limitations: ctDNA clearance was associated with outcomes but not validated for treatment guidance, while survival differences between mCARG groups were primarily prognostic rather than treatment-related.
Disclaimer: Publication dates refer to online-first publication.
If you found this useful, subscribe to receive next month’s synthesis.
Know a colleague who would benefit? Forward this issue.

Refer a colleague

