Result: Ivonescimab + chemotherapy significantly improved OS versus an active PD-1 + chemotherapy comparator.

Key data: Median OS 27.9 vs 23.7 months; HR 0.66, 95% CI 0.50–0.87. OS benefit was seen in both PD-L1 <1% and PD-L1 ≥1% subgroups.

Why it matters: One of the clearest survival-positive head-to-head challenges to a PD-1 + chemotherapy backbone in first-line squamous NSCLC.

Caveat: China-based trial with tislelizumab comparator; global applicability will depend on regulatory context and comparison with local standards.

Result: Sunvozertinib significantly improved PFS versus chemotherapy.

Key data: Median PFS 10.3 vs 7.5 months; HR 0.65, 95% CI 0.50–0.85. ORR 58.9% vs 31.1%; median DoR 11.2 vs 7.1 months. OS immature; 90.2% crossed over from chemotherapy to sunvozertinib after progression.

Why it matters: Strengthens first-line targeted therapy for EGFR exon 20 insertion NSCLC and offers an oral chemotherapy-free option.

Caveat: OS interpretation will be difficult due to high crossover.

Result: Sac-TMT + pembrolizumab significantly improved PFS versus pembrolizumab alone.

Key data: N=413. Median PFS not reached vs 5.7 months; HR 0.35, 95% CI 0.26–0.47. ORR 70.2% vs 42.0%. OS immature but favorable trend: HR 0.55. Grade ≥3 TEAEs 55.3% vs 31.4%.

Why it matters: One of the strongest phase III signals for first-line ADC–IO escalation in PD-L1-positive NSCLC.

Caveat: Pembrolizumab monotherapy is a debated comparator for PD-L1 1–49% disease in many settings; OS maturity is essential.

Result: Selpercatinib significantly improved EFS versus placebo after definitive locoregional treatment.

Key data: In stage II–IIIA disease, EFS HR 0.172, 95% CI 0.058–0.509. Median EFS not reached vs 31.8 months. Two-year EFS 91.5% vs 61.1%.

Why it matters: Extends the adjuvant targeted therapy paradigm beyond EGFR and ALK into RET-positive early-stage NSCLC.

Caveat: OS immature; reinforces the need for broad genomic testing in resectable disease.

Result: Adding thoracic RT to chemo-immunotherapy did not improve outcomes and increased toxicity.

Key data: Median OS 10.0 vs 11.1 months; HR 1.12, 95% CI 0.82–1.54. Median PFS 5.1 vs 5.1 months; HR 1.09. Esophagitis occurred in 47.8%; grade 3–4 esophagitis in 10.4%. Trial stopped early for futility and safety.

Why it matters: Provides randomized evidence against routine early thoracic RT intensification with first-line durvalumab-based chemo-IO in ES-SCLC.

Caveat: Timing, dose, and patient selection remain open questions, but this regimen should not be broadly adopted.

Result: ctDNA-guided escalation to osimertinib + chemotherapy improved PFS in patients with persistent plasma EGFR mutation after initial osimertinib.

Key data: N=80 randomized. Median PFS 23.1 vs 12.7 months; HR 0.53, 95% CI 0.31–0.92. ORR 50% vs 35%. Grade ≥3 TRAEs 65% vs 10%.

Why it matters: A practical example of early molecular response being used to personalize intensification.

Caveat: Small phase II trial; toxicity is substantially higher with chemotherapy.

Result: Upfront cranial RT reduced intracranial progression but did not improve survival outcomes.

Key data: N=208. Two-year cumulative incidence of intracranial progression 21.7% vs 50.0%; sub-HR 0.35, 95% CI 0.21–0.59. Median OS 23.3 vs 28.7 months; p=0.06. Grade >3 radiation necrosis occurred in 5.8% in the upfront RT arm.

Why it matters: Adds level I evidence to a common clinical dilemma: intracranial control improves with upfront RT, but survival does not.

Caveat: QOL and neurocognition data are pending and will be critical.

Result: Pralsetinib improved PFS and response versus platinum-based standard therapy.

Key data: N=223. Median PFS 18.7 vs 9.0 months; HR 0.59, 95% CI 0.42–0.84. ORR 65.5% vs 41.6%; median DoR 20.6 vs 9.7 months. Infection-related deaths occurred in 8 patients in the pralsetinib arm.

Why it matters: Confirms first-line RET inhibition as superior to chemotherapy-based standard therapy.

Caveat: Safety signal around infections needs attention.

Result: Adjuvant nivolumab did not improve DFS versus best supportive care.

Key data: N=935. Median DFS 71.3 vs 68.8 months; HR 0.97, 95% CI 0.81–1.17. In PD-L1 ≥50%, DFS HR 0.86, 95% CI 0.59–1.25. Grade 3–5 treatment-related toxicities occurred in 25%.

Why it matters: Reinforces that adjuvant IO benefit cannot be assumed from metastatic or perioperative data.

Caveat: Conducted in an evolving perioperative landscape; current practice increasingly uses neoadjuvant or perioperative IO strategies.

Result: Lorlatinib continues to show exceptional long-term disease control.

Key data: Median PFS still not reached after 7 years; HR 0.19, 95% CI 0.13–0.26. Seven-year PFS 55% vs 3%. No new intracranial progression events after 30 months on lorlatinib.

Why it matters: Sets a durability benchmark for first-line ALK therapy.

Caveat: Comparator is crizotinib, not a modern next-generation ALK inhibitor.

Result: Neoadjuvant lorlatinib produced high radiographic and pathological response rates in stage III ALK-positive NSCLC.

Key data: N=43. ORR 83.7%; no progressive disease. Among 32 operated patients, R0 resection 96.9%, pCR 46.9%, MPR 81.3%, and nodal downstaging 90.6%. In initially unresectable stage III disease, 75.0% achieved conversion surgery.

Why it matters: Suggests ALK-directed neoadjuvant therapy may become a serious strategy in locally advanced ALK-positive disease.

Caveat: Single-arm phase II with short follow-up.

Result: Divarasib + pembrolizumab showed promising activity in untreated KRAS G12C-positive disease.

Key data: In PD-L1-positive patients treated with divarasib + pembrolizumab, ORR 73%; median PFS 19.3 months. Grade 3–4 TRAEs occurred in 65%, including ALT increase 20% and AST increase 18%.

Why it matters: KRAS G12C inhibition may move into first-line combination strategies.

Caveat: Early-phase, non-randomized; hepatotoxicity remains a practical concern.

Result: Tarlatamab improved CNS PFS and OS versus chemotherapy in patients with baseline brain metastases.

Key data: CNS PFS 6.5 vs 4.2 months; HR 0.40, 95% CI 0.24–0.66. CNS complete response 15% vs 5%. In patients with baseline brain metastases, median OS was 13.9 vs 6.8 months; HR 0.51, 95% CI 0.34–0.74.

Why it matters: Supports tarlatamab as a relevant second-line option even in SCLC patients with brain metastases.

Caveat: Post-hoc CNS analysis; most patients had prior CNS treatment.

Bispecifics are no longer theoretical: HARMONi-6 provides an OS-positive head-to-head signal against an active PD-1 comparator in squamous NSCLC.

ADC–IO is moving earlier: OptiTROP-Lung05 suggests major first-line PFS activity.

Early-stage genomics is expanding: LIBRETTO-432 pushes RET into the adjuvant targeted therapy conversation; LORIN suggests ALK may move further into neoadjuvant strategies.

ctDNA-guided escalation is becoming tangible: FLAME shows how early molecular persistence could identify EGFR-mutant patients needing intensified therapy.

More treatment is not always better: ES-SCLC thoracic RT and adjuvant nivolumab both show that plausible escalation can fail when tested prospectively.

CNS strategy remains unsettled: Upfront brain RT improves intracranial control in oncogene-driven NSCLC without clear survival benefit so far, while tarlatamab shows meaningful CNS activity in SCLC.

ALK remains the benchmark for durable metastatic control: CROWN’s 7-year data continue to set a high bar for targeted therapy outcomes.