Month in Review
November delivered a concentrated set of advances across targeted therapy, immunotherapy, perioperative biomarkers, CNS-directed strategies, and SCLC. ctDNA-driven personalization matured with interventional and surrogate-endpoint data; ROS1 and MET-directed therapies continued sharpening post-TKI resistance pathways; and early-stage tools, from methylation-based liquid biopsy to radiotherapy QoL analyses, pushed the field toward more refined patient selection. SCLC saw an unusual breadth of prospective work, including induction IO + surgery signals and novel maintenance concepts. Collectively, November reinforces the direction toward biology-driven escalation and functional de-escalation, supported by increasingly rich molecular and imaging biomarkers.
Core Studies
Targeted Therapy & ADCs
Savolitinib + osimertinib in MET-amplified post-osimertinib NSCLC
Clinical Lung Cancer
Study: Double-blind phase II; EGFR-mutated, MET-amplified disease after osimertinib; savolitinib + osimertinib vs savolitinib alone.
Key data: ORR 57% vs 13%; mPFS 7.4 vs 1.6 mo; OS NR vs 13.4 mo; grade ≥3 AEs 21% vs 31%.
Why it matters: Strengthens combined EGFR + MET inhibition as a strategy for MET-driven osimertinib resistance.
Clinical context & caveats: Small trial (n=30) with early termination; results directionally consistent with SAVANNAH/SACHI but require phase III confirmation.
Lorlatinib in TKI-naive ROS1-positive NSCLC
JAMA Oncology
Study: Multicenter, single-arm phase II in untreated or minimally pretreated ROS1+ NSCLC.
Key data: ORR 73% (90% in untreated); mPFS 53.6 mo; intracranial ORR 71%; grade ≥3 lipid disorders 25%.
Why it matters: One of the longest PFS outcomes reported in ROS1 disease; strong CNS efficacy.
Clinical context & caveats: Small, Korea-only cohort; single-arm; comparisons to repotrectinib should not be made directly.
Immunotherapy – Metastatic & Oligometastatic
Tumor-informed ctDNA guiding first-line immunochemotherapy
Advanced Science
Study: ctDNA-informed adjustment of 1L chemo-IO in advanced NSCLC using tumor-informed assays.
Key data: ctDNA dynamics strongly predicted PFS/OS; ctDNA-clearers had substantially prolonged benefit vs non-clearers.
Why it matters: Early interventional evidence supporting ctDNA as a real-time treatment-guidance tool.
Clinical context & caveats: Not a phase III study; heterogeneity of IO backbones; assay-specific.
ctDNA clearance as an intermediate endpoint after 1L chemo-IO (IMpower150 biomarker analysis)
JTO Clinical and Research Reports
Study: Retrospective analysis of ctDNA clearance and radiographic response as predictors of PFS/OS.
Key data: ctDNA clearance → PFS HR 2.06, OS HR 0.35; combined ctDNA_CL + cResp gave strongest stratification.
Why it matters: Supports ctDNA clearance as a robust patient-level prognostic marker and potential co-endpoint.
Clinical context & caveats: Surrogacy at trial level remains modest; assay/platform dependent.
BOOSTER trial: ablation + IO vs IO alone in oligo-residual NSCLC
Signal Transduction and Targeted Therapy
Study: 2:1 randomization to ablation + IO vs IO alone after initial response to PD-1/L1 therapy.
Key data: mPFS 26.7 vs 11.7 mo (HR 0.21); exploratory OS trend favoring ablation; limited grade ≥3 toxicity.
Why it matters: Prospective evidence that local ablation can meaningfully extend PFS after IO response in select patients.
Clinical context & caveats: Single center, China-only; small sample; highly selected responders.
Early-Stage & Perioperative
cfDNA methylation to predict neoadjuvant pathologic response
BMC Medicine
Study: Prospective phase II; ΔMeth-score after 2 cycles neoadjuvant chemo-IO.
Key data: AUC for predicting MPR 0.83; mutation-based ctDNA sensitivity only 29% vs 82% for methylation; MPR 45%.
Why it matters: Methylation-based cfDNA outperforms mutation ctDNA in early-stage disease; strong candidate for perioperative response assessment.
Clinical context & caveats: Single center; internally derived thresholds; survival data immature.
QoL after SBRT vs hypofractionated RT for inoperable stage I NSCLC
International Journal of Radiation Oncology Biology Physics
Study: Multicenter randomized trial; SBRT (2:1) vs CHRT (60 Gy/15).
Key data: Slight short-term advantage for SBRT; no clinically meaningful long-term QoL differences; toxicity similar except one grade 5 hemoptysis (SBRT).
Why it matters: Supports equivalence in long-term QoL and reinforces individualized selection.
Clinical context & caveats: Closed early; missing QoL at later timepoints; central/ultracentral mix.
Surgery vs SABR in screen-detected stage I NSCLC (YLST)
Lung Cancer
Study: Prospective screening cohort; 132 surgery vs 58 SABR.
Key data: Cancer-specific survival similar; OS worse in SABR due to non-cancer deaths (29% vs 7%).
Why it matters: In screening-era stage I disease, patient fitness, not tumor biology, drives OS differences.
Clinical context & caveats: Strong selection bias; many SABR patients lacked histologic confirmation.
SCLC Advances
LungMate-005: induction anti–PD-L1 + chemotherapy followed by surgery
Cell Discovery
Study: Anti–PD-L1 + platinum/etoposide ×4 → surgery or RT → PD-L1 maintenance; 95% stage III.
Key data: ORR 92.5%; MPR 62%; pCR 43%; 24-mo OS 85.7% in surgery group.
Why it matters: Rare prospective evidence supporting surgery after IO + chemo in select stage III SCLC.
Clinical context & caveats: Non-randomized; surgery assigned by MDT; small, single-center.
Other Notable Publications
Translational & Biomarker Work
AI-powered spatial TME profiling after EGFR-TKI resistance
Journal for ImmunoTherapy of Cancer
Study: Retrospective multi-cohort analysis (n=143) using AI-based whole-slide spatial profiling (Lunit SCOPE IO) to characterize TME remodeling after EGFR-TKI resistance and its predictive value for ICI monotherapy.
Key data: Post-TKI biopsies showed decreased intratumoral TILs and increased endothelial cells; high post-TKI TILs predicted better ICI outcomes (ORR 41.7% vs 9.7%; PFS 4.9 vs 1.8 mo). Spatial transcriptomics validated AI-derived signatures.
Why it matters: Provides a practical, pathology-integrated biomarker framework to stratify post-TKI ICI benefit in EGFR-mutant NSCLC; a key unmet clinical domain.
Clinical context & caveats: Single-center Korean cohort; heterogeneous biopsy timing; cutoffs internally derived and need external validation; PD-L1 TPS not predictive despite post-TKI increase.
Brain Metastases
Brain Metastasis Velocity predicts survival and WBRT need after repeated SRS (CYBER-SPACE)
Radiotherapy & Oncology
Study: Prespecified secondary analysis of the CYBER-SPACE randomized phase II trial assessing prognostic models for OS and WBRT indication in patients undergoing repeated SRS for up to 10 brain metastases.
Key data: BMV was the only score consistently predictive of OS (mOS 24.6 vs 9.9 mo using 8.5 cutoff) and WBRT need (WBRTi 4.8 mo vs NR using 16.2 cutoff). Neurologic death rate 10%.
Why it matters: Establishes BMV as a practical metric to guide focal-therapy sequencing and WBRT deferral in modern repeated-SRS paradigms.
Clinical context & caveats: Single-center; optimized cutoffs may overfit; requires systems with high MRI surveillance capacity.
Immunotherapy and Radiation
Determinants of durvalumab non-completion after CRT in stage III NSCLC
Translational Lung Cancer Research
Study: Retrospective Japanese cohort (n=65) evaluating predictors of failing to complete 12 months of durvalumab after definitive CCRT.
Key data: Completion rate 46%; low-dose carboplatin CCRT strongly predicted non-completion (20% vs 54%; p=0.04); discontinuations mainly due to pneumonitis (n=12). Non-completion associated with worse PFS (7.8 vs 44.8 mo) and OS (22.3 mo vs NR).
Why it matters: Highlights how regimen selection and pulmonary toxicity shape real-world PACIFIC-era outcomes, especially in elderly/frail patients.
Clinical context & caveats: Small, single-center dataset; frailty-driven treatment selection explains many associations; findings are hypothesis-generating.
Expert Takeaways
ctDNA is emerging both as a real-time interventional guide and as a strong prognostic co-endpoint across 1L and perioperative settings.
MET-amplification and ROS1 data reinforce precision-resistant strategies and highlight CNS activity as a defining feature of next-gen TKIs.
Early-stage personalization accelerates: methylation cfDNA and real-world SABR vs surgery reinforce functional rather than purely anatomical decision-making.
SCLC shows diversification: surgical exploration in stage III, maintenance angiogenesis inhibition, and critical negative data refining PD-1 strategies.
CNS management continues shifting toward high-frequency MRI + repeat SRS, guided increasingly by BMV rather than lesion count.
Notes on generalizability & bias signals (cross-cutting)
Region-locked trials: Several key datasets (BOOSTER, methylation cfDNA, SCLC maintenance) are China- or Asia-only.
Selection bias: SABR patients in YLST significantly older/comorbid; LungMate-005 surgical cohort MDT-selected.
Assay specificity: Methylation scores and ctDNA-informed pathways are platform-specific.
Operational constraints: Repeat SRS strategies require tight MRI access; real-world replication may vary.
Disclaimer: Publication dates reflect journal issue month; some articles may have appeared online earlier ahead of print.
Refer a colleague