A little-known immune cell could explain why some small cell lung cancer patients respond to immunotherapy

Patients whose small cell lung cancer (SCLC) tumors carried high levels of an immune cell called the gamma delta (γδ) T cell lived longer on anti-PD-L1 immunotherapy, around 15.4 months, compared with 10.8 months for patients with low levels, according to a study published Oct. 12 in Cancer Cell.

SCLC is one of the most aggressive forms of lung cancer, and most patients are diagnosed after it has already spread. Chemotherapy paired with anti-PD-L1 drug, Tecentriq (atezolizumab) or Imfinzi (durvalumab), is standard first-line treatment for extensive-stage disease, but few patients see responses that last. This is likely due because SCLC tumors switch off MHC class I (MHC-I), the surface machinery that normally flags a cancer cell for killer T cells. Without it, the immune cells those drugs are meant to unleash have trouble finding the tumor.

This gap matters for payers as much as clinicians. Imdelltra (tarlatamab), the bispecific T-cell engager approved for previously treated extensive-stage SCLC in 2024, carried a list price of $31,500 for the first cycle, and there’s still no reliable way to predict who will benefit.

The work was led by Jin Ng, Ph.D., a researcher at the Walter and Eliza Hall Institute of Medical Research (WEHI) in Melbourne, Australia, and his team who set out to find immune cells that can recognize tumors without relying on MHC-I. γδ T cells fit: unlike conventional T cells, they spot stressed or transformed cells directly, they cluster naturally in the lungs and they appear less prone to burning out.

Single-cell profiling of lymph node biopsies

The researchers ran single-cell RNA sequencing on biopsy samples from 12 patients with SCLC — 11 from needle biopsies of chest lymph nodes taken during routine diagnosis, plus one surgical sample. In all, 13,002 cells passed quality checks, about 3,600 of them immune cells. The team confirmed those results by sorting and staining cells, then tested its findings against stored tumor gene-expression data from the IMpower133 (271 patients) and CASPIAN trials and in lab experiments using SCLC cell lines, healthy-donor immune cells and tumor organoids grown from patients' own samples.

Higher γδ T cell levels tracked with longer survival

γδ T cells turned up in every patient sample, at both the lymph nodes and the primary tumor, carrying the genetic signature of cells equipped to kill. In the trial data, patients with high levels of a γδ gene marker who received atezolizumab plus chemotherapy trended toward longer overall survival at 15.4 months compared to 10.8 months for low-level patients in the same arm.

A second γδ marker showed the same pattern in IMpower133 (14.4 months versus 10.8 months) and in CASPIAN, where patients on durvalumab plus chemotherapy reached 12.7 months compared to 8.6 months. All of those differences were trends, not statistically significant results. One finding did reach significance: patients with high tumor expression of BTN2A1, a protein the cells use to recognize tumors, survived a median 14.4 months on atezolizumab plus chemotherapy, compared with 9.7 to 10.4 months in the other groups (hazard ratio, 0.67; 95% confidence interval, 0.45-0.99).

A γδ marker did not predict survival in patients who never received immunotherapy, suggesting it tracks treatment response rather than simply a better prognosis.

In the lab, γδ T cells killed SCLC cells about as effectively as conventional CD8+ T cells when tarlatamab was added, across every molecular subtype tested. Zoledronate, a generic bone drug already approved for osteoporosis and cancer-related bone disease, made SCLC cells easier to kill, and expanded γδ T cells from one patient's blood roughly 260-fold over 13 days, from 0.15% of cells to 38.5%. Deleting BTN2A1 from tumor cells blunted the killing, confirming the recognition route.

The gene signature now used to flag the “inflamed” SCLC subtype most likely to respond to checkpoint inhibitors leaves out γδ genes entirely, and the study found these cells across subtypes generally considered immunologically cold. The authors argued γδ T cells are “an underappreciated immune population with therapeutic relevance in SCLC,” and that adding their genes to existing signatures could flag a broader group of patients worth treating. The zoledronate finding also raises the prospect of a cheap, long-available drug in a very expensive category.

Strengths and limits of the research

The study's main strength is that it moves in both directions, from patient samples to trial data and back into lab experiments, rather than resting on one line of evidence, and the γδ signal held up across all 12 patients.

However, there are limitations. Most samples came from lymph nodes rather than the lung tumors themselves, so the researchers infer that these cells reach primary tumors but did not show it. The sequencing covered relatively few immune cells.

The trial subgroups were also small, and the survival differences were not statistically significant. The killing experiments were done in dishes and organoids, not animal models, and the researchers focused on one γδ subset (Vδ2+) while acknowledging other subsets may do similar work.

The authors recommend larger cohorts to establish whether circulating Vδ2+ levels can serve as a prognostic marker, further investigation of γδ T cell adoptive transfer in SCLC and in vivo work to confirm the lab findings. They also call for research into zoledronate's effect on overall survival in SCLC specifically, noting it remains unclear even though up to 25% of patients present with bone metastases at diagnosis.