Northwestern Medicine scientists have identified a stem-like state in small-cell lung cancer that identifies a group of patients with exceptionally poor outcomes following chemoimmunotherapy, according to a study published in Cell Reports Medicine.
The findings could pave the way for new biomarkers and treatment strategies for one of the deadliest forms of lung cancer, said Anish Thomas, MD, the Roswell Park Professor of Hematology and Oncology and senior author of the study.
Small-cell lung cancer (SCLC) accounts for roughly 15 percent of lung cancer cases, according to estimates from the Centers for Disease Control and Prevention, and is characterized by rapid growth, early spread and frequent recurrence after treatment. Despite recent advances, clinicians have few tools to help determine which patients are most likely to have more aggressive cancers.
In the study, Thomas and his collaborators analyzed the chromatin landscape of SCLC, examining which regions of the genome were accessible to the tumor cells. The approach revealed a previously unrecognized tumor state, dubbed SCLC-SL, that appears particularly adaptable and is associated with poor treatment outcomes.
“We identified a previously unrecognized stem-like regulatory state of small-cell lung cancer by examining how DNA is packaged and which regions of the genome are accessible to the tumor cell,” said Thomas, who is also a member of the Robert H. Lurie Comprehensive Cancer Center of Northwestern University. “This state, which we call SCLC-SL, is less differentiated, metabolically adaptable, and has reduced immune signaling compared with other SCLC states.”
Most previous SCLC classification systems have relied on bulk gene-expression profiles. A limitation of this approach is that RNA measured from a tumor reflects a mixture of cancer cells and surrounding immune and stromal cells, making it difficult to distinguish tumor-intrinsic biology from differences in the tumor microenvironment. This may help explain why existing molecular classifications have not consistently separated patients according to clinical outcome.
To overcome this limitation, the investigators analyzed patient-derived xenograft tumors, which enabled tumor-specific regulatory signals to be distinguished from stromal signals, and used a technology called ATAC-seq to map chromatin accessibility, revealing which genes tumor cells are able to turn on. They then integrated the data with gene-expression profiles from more than 800 tumors across multiple patient cohorts.
The work revealed three major chromatin-defined regulatory states: a neuroendocrine state, an immune-active state and the newly discovered stem-like state. Tumors classified as SCLC-SL showed increased activity in biological processes associated with hypoxia, oxidative metabolism and cellular stress responses, and also exhibited reduced immune activity.
“We looked at which genes tumor cells can use. This revealed a biologically distinct stem-like state that was not fully captured by conventional SCLC classifications and was strongly associated with clinical outcome,” Thomas said.
Most notably, patients whose tumors were enriched for the stem-like state experienced substantially worse outcomes after chemoimmunotherapy. In a clinical trial cohort analyzed in the study, patients with SCLC-SL tumors had a median overall survival of 7.41 months, compared with 15.9 months for the immune-active subtype and 12.6 months for the neuroendocrine subtype. The association remained significant even after accounting for liver metastases, brain metastases and elevated lactate dehydrogenase levels (LDH), according to the study.
The findings represent one of the first molecular classifications of SCLC to distinguish patient outcomes even after accounting for established clinical risk factors, Thomas said.
“To our knowledge, this is one of the first studies to show that a SCLC molecular feature has independent prognostic significance after accounting for established clinical risk factors, including liver and brain metastases and serum levels of LDH,” Thomas said.
The findings also suggest that SCLC tumors are more dynamic than previously appreciated.
“This shifts the way we think about SCLC subtypes,” Thomas said. “Rather than viewing tumors as belonging to fixed categories, our data suggest that SCLC can occupy different regulatory states and potentially transition between them. That plasticity may be central to how these cancers adapt to treatment and develop resistance.”
The investigators also found evidence that the stem-like state becomes more common after treatment. Analysis of matched patient samples collected before and after therapy showed enrichment of stemness following treatment, suggesting a more adaptable, treatment-resistant tumor type.
While more research is needed to fully understand SCLC subtypes, the study establishes a potential framework for identifying patients at highest risk for poor outcomes and tailoring future therapies.
The biology of the newly identified state may also provide new therapeutic opportunities, Thomas said. The SCLC-SL state was associated with increased reliance on metabolic and stress-response pathways.
“If validated prospectively, a chromatin-informed signature could help identify patients whose tumors have this high-risk stem-like biology and who may be less likely to benefit from current approaches alone,” Thomas said. “It could also help direct these patients toward clinical trials testing alternative strategies.”
Looking ahead, the team plans to determine what drives tumors into the stem-like state and whether that process can be prevented or reversed.
“The next step is to validate this signature prospectively in patients and determine whether it can reliably identify patients with high-risk disease before treatment,” said Thomas, who is also a member of the Center for Human Immunobiology (CHI). “We also want to understand what drives tumor cells into this stem-like state and whether we can therapeutically prevent or reverse it.”
The Intramural Research Program of the Center for Cancer Research, National Cancer Institute, National Institutes of Health, supported the study under grant ZIA BC 011793.
