For decades, small cell lung cancer was considered a monolithic, uniformly lethal disease — one that simply didn't yield to the targeted therapy revolution that transformed other cancers. A comprehensive review in Cancers challenges that assumption at two levels simultaneously, with implications for how oncologists approach treatment selection and how researchers design the next generation of trials.
Transcriptomic profiling has fractured SCLC into at least four biologically distinct subtypes: SCLC-A (driven by ASCL1), SCLC-N (NEUROD1-driven), SCLC-P (POU2F3-driven), and SCLC-I (characterized by immune inflammation). These subtypes differ not only in transcription factor dominance but in surface antigen expression, immune microenvironment composition, and — critically — their vulnerability to specific therapeutic classes. The review further uses delta-like ligand 3 (DLL3) as a molecular case study in what the authors term the "target-modality problem": rovalpituzumab tesirine, an antibody-drug conjugate targeting DLL3, failed across three randomized trials, while tarlatamab, a bispecific T-cell engager against the same antigen, achieved FDA accelerated approval in 2024 and full traditional approval in 2025 following positive Phase 3 data. Same target, opposite outcomes — explained by how the therapeutic payload is delivered rather than whether the target is valid.
This framework carries significant weight for the broader oncology field. The DLL3 story is arguably the clearest real-world demonstration that target validation and modality selection are separable problems — a lesson that may inform ADC versus bispecific development strategies across other tumor types. The review also contextualizes emerging maintenance strategies, including the lurbinectedin-atezolizumab combination from the IMforte trial, within the subtype framework. Key limitations remain: intratumoral heterogeneity and phenotypic plasticity mean clean subtype assignment is rarely achievable in clinical practice, and most subtype-specific therapeutic data remain early-stage. This is a synthesis that is more paradigm-organizing than paradigm-shifting, but it arrives at a moment when the field badly needs conceptual scaffolding to match its experimental ambition.