Themes — ASCO 2026¶

Cross-day synthesis of the recurring threads across the 2026 ASCO Annual Meeting (Chicago, May 29 – Jun 2). Each thread cites the specific built trial pages; LBA1–5 anchored the Sunday May 31 plenary. Cross-links reach back to JPM 2026 themes and the AACR 2026 vault where the science rhymes.

Source caveat: Several presentations (TRIPLEX, persevERA, JSKN016, revumenib maintenance, KN026-004) were embargoed or topline-only at writing — figures noted as "not disclosed" reflect that, not missing data.

1. RAS-class drugs become real¶

The headline of the meeting. RASolute 302 is the first positive phase 3 of a pan-RAS(ON) inhibitor: 2L daraxonrasib nearly doubled median OS vs chemo in metastatic PDAC (13.2 vs 6.7 mo, HR 0.40; RAS G12 HR 0.40, P=5.9×10⁻¹⁰), the hardest tumor in oncology, with simultaneous NEJM publication. This validates the direct-RAS franchise that dominated JPM 2026 deal chatter and the KRAS-G12D/pan-RAS programs profiled at AACR 2026 — moving the class from preclinical promise and G12C niche into broad, plenary-grade clinical reality.

2. The PD-1×VEGF bispecific class arrives¶

HARMONi-6 (LBA4) delivered the first OS-relevant win for a PD-1×VEGF bispecific: ivonescimab + chemo beat tislelizumab + chemo in 1L squamous NSCLC (median OS 27.89 vs 23.69 mo) — an active-comparator, head-to-head-vs-PD-1 design, not a placebo bar. This is the clinical readout behind the AbbVie + RemeGen $5.6B PD-1×VEGF license that was the biggest single JPM 2026 headline, and it cements the bispecific backbone as the next 1L IO platform.

3. Perioperative and adjuvant targeting expands — but selection matters¶

Moving active drugs earlier paid off when the target was right and failed when it wasn't. PROTEUS (LBA1) made perioperative apalutamide + ADT a new SOC in high-risk localized prostate (MFS HR 0.80; pCR/MRD 8.9% vs 1.0%). LIBRETTO-432 (LBA3) cut recurrence/death 83% with adjuvant selpercatinib in RET+ NSCLC — the first adjuvant RET TKI. KN026-004 lifted neoadjuvant tpCR in HER2+ early breast cancer, and lidERA cut iDFS risk 30% with adjuvant giredestrant. The counter-example: ANVIL/ALCHEMIST EA5142 — adjuvant nivolumab gave no DFS benefit in EGFR/ALK-WT resected NSCLC (HR 0.97), a reminder that unselected adjuvant IO has hit its ceiling.

4. ADCs and the durability question¶

The Trop-2 ADC story matured from response rate to durability. ASCENT-04 (LBA1000) showed sacituzumab govitecan + pembro beat chemo + pembro in 1L PD-L1+ metastatic TNBC (PFS 11.2 vs 7.8 mo, HR 0.65), with the ASCO 2026 update adding PFS2/subsequent-therapy data and benefit consistent across every biomarker subgroup including low Trop-2 (Trop-2 Q4 HR 0.57). DESTINY-Breast09 reinforced the durability theme (median CR treatment duration 28.0 mo), while novel ADC platforms surfaced behind it: the Nectin-4 topo-I ADC LY4052031, HER2 ADC trastuzumab rezetecan in CRC, and the first-in-class TROP2/HER3 bispecific ADC JSKN016.

5. Endocrine-backbone resistance: the next-gen SERD reckoning¶

The oral SERD class split. SERENA-6 validated ctDNA-guided early switching to camizestrant at emergent ESR1m (PFS2 HR 0.52–0.63), and evERA showed giredestrant + everolimus post-CDK4/6i (PFS2 19.0 vs 13.2 mo). But persevERA missed its primary endpoint — giredestrant + palbociclib gave only a numerical, non-significant 1L PFS gain vs letrozole + palbociclib — even as the same molecule won in the adjuvant (lidERA) and post-progression settings. The lesson: SERD benefit is real but context- and biomarker-dependent (ESR1m), not a universal 1L backbone swap.

6. China-origin assets dominate the LBA slate¶

A structural shift, not a footnote. China-originated molecules ran through nearly every track: ivonescimab (HARMONi-6), zanidatamab (HERIZON-GEA-01), sunvozertinib (WU-KONG28, 1L monotherapy beating platinum doublet in EGFR exon20ins), the TROP2 ADC sac-TMT (OptiTROP-Lung05), anbenitamab (KN026-004), satri-cel Claudin18.2 CAR-T (CT041), olverembatinib (CML), the anti-CD5 nanobody CAR-T (CONQUER), and JSKN016. This is the clinical-data manifestation of the "China is the story" thread from JPM 2026 — assets now reading out as practice-changing plenary data, not just licensing line-items.

7. Biomarker-matched wins crack open rare and hard tumors¶

Precision oncology kept clearing previously untreatable settings. SARC041 (LBA2) was the first positive phase 3 in dedifferentiated liposarcoma — abemaciclib (CDK4 axis) extended PFS 9.7 vs 1.5 mo. OptimUM-02 gave 1L HLA-A*02:01-negative metastatic uveal melanoma a real option with darovasertib + crizotinib (PFS 6.9 vs 3.1 mo, ORR 37% vs 6%). BREAKWATER (EC + FOLFIRI in BRAF V600E mCRC, PFS 15.2 vs 8.3 mo) and HERIZON-GEA-01 (HER2+ GEA) extended the matched-target logic into GI, and in heme revumenib (KMT2Ar/NPM1m/NUP98r maintenance) and PRMT5 inhibitor PRIMAVERA pushed mutation- and epigenetic-target matching into late-line disease.

8. Bispecifics and CAR-T move to the front of heme¶

Off-the-shelf bispecifics started displacing standard regimens and even autologous CAR-T positioning. MajesTEC-9 showed teclistamab monotherapy beat PVd/Kd SOC in RRMM (PFS HR 0.29), and EPCORE DLBCL-1 was the first phase 3 of a CD3×CD20 bispecific monotherapy to beat chemoimmunotherapy in R/R DLBCL. frontMIND brought the antibody tafasitamab into 1L high-risk DLBCL (PFS HR 0.75), SUCCESSOR-2 advanced the oral CELMoD mezigdomide, and the nanobody anti-CD5 CAR-T CONQUER opened T-cell malignancies — collectively pulling redirected-T-cell therapy earlier in the heme treatment line.