Metabolomics Crucial To Personalizing Lung Cancer Therapy
By Deborah Borfitz
August 17, 2022 | A popular hypothesis in oncology research over the past few years is that immunotherapy works synergistically with glutaminase inhibition, based on promising results from preclinical studies using cell models grown in a dish. Banking on that assumption, clinical trials have launched to evaluate the combination approach in patients with non-small cell lung cancer—a terminated phase 1 trial and a terminated phase 2 trial—as well as a completed melanoma study.
While no results from these trials have yet been released, it may have been a doomed strategy from the start, according to Sarah Best, Ph.D., and Kate Sutherland, Ph.D., researchers from WEHI (Melbourne, Australia).
Best is a laboratory head in The Brain Cancer Centre but, until this year, she worked as a lung cancer researcher under the supervision of Sutherland who is a lung cancer biologist. Together, they led a study published in Cell Metabolism (DOI: 10.1016/j.cmet.2022.04.003) that used metabolomics analysis to identify what looks to be a very big problem.
In an aggressive form of lung cancer, KRAS-mutant lung adenocarcinoma, a glutaminase inhibitor used to suppress tumors was found to counteract the benefits of anti-PD1 immunotherapy intended to boost the immune system, says Sutherland. The discovery was made with genetically engineered mouse models with a fully active immune system and metabolism, whose tumors developed from cells in their lungs rather than being implanted subcutaneously, which is a notable departure from other preclinical models.
Tumors arose in vivo in the mice, making the experiments closely mimic what might happen in human patients. The project was expected to confirm that immunotherapy would synergize with glutaminase inhibition, Sutherland says, given all the high-profile papers suggesting it was so.
Instead, “it was really quite striking that something was going wrong,” Best says. Delving further, they learned that “the T cells were unable to fully activate because they also needed the function of glutaminase, the enzyme that we were inhibiting.”
Best says the “energy fingerprint” of different cell types, as measured by metabolomics, turned out to be critically important. Researchers found that genetic alterations in lung adenocarcinomas caused by STK11/Lkb1 mutations are more susceptible to glutaminase inhibition, which wasn’t previously appreciated. “So, we have hypothesized that glutaminase inhibition by itself would have a really good impact in this patient population, but we shouldn’t be combining it with immunotherapy.”
Most surprising was the fact that in the mice where the glutaminase inhibitor was added on top of the immunotherapy, T cells became incapable of clonal expansion enabling them to mount an effective immune response—unlike in the mice receiving immunotherapy alone, Best says.
Prior studies elsewhere indicated immunotherapy would activate T cells and glutaminase inhibition would dampen the energy sources of the tumor, making them an ideal therapeutic pairing, she says. The unexpected discovery that immunotherapy and metabolic inhibitors could sometimes be potentially detrimental finds a parallel in research conducted a few years ago where T cells were found to be sensitive to cancer drugs targeting the pentose phosphate pathway.
Best and Sutherland say they are hopeful the latest Cell Metabolism paper will “pump the brakes” on some planned combination trials while researchers reconsider the actual scientific evidence backing their advancement and the kinds of preclinical models that they have been tested in. They say it has certainly highlighted the importance of publishing negative results when things don’t turn out as expected.
A basket trial terminated earlier this year due to a lack of clinical benefit was in fact a step toward a more personalized therapeutic approach, says Best. It was investigating the efficacy of glutaminase inhibition (Telaglenastat) relative to mutations in KEAP1, STK11/Lkb1, or NF1.
As suggested by the literature, it was thought that glutaminase inhibitor would be effective in lung adenocarcinoma that had the KEAP1 mutation, she notes. But many patients have the combination of both KEAP1 and STK11/Lkb1 mutations. “Assuming this drug works well in patients with KEAP1 mutation is a fair assumption, but because we have these genetically engineered mouse models, we could take the lung adenocarcinoma that just had a KEAP1 mutation and compare it to one that just had a STK11/Lkb1 mutation or a combination of both mutations, and then use metabolomics assays to see who would benefit from inhibiting the glutaminase enzyme.”
Glutaminase monotherapy may in fact benefit patients with STK11/Lkb1-mutant cancers, which is not currently recognized in the clinic, Best says. “We can use our mouse models to investigate personalized therapies that can have direct translational impact.”
This research approach continues in the Sutherland lab. “There is a real need to identify new therapeutic targets that enhance the response to immunotherapy for these hard-to-treat lung cancers,” Sutherland says.
Meanwhile, since being appointed lab head in The Brain Cancer Centre, Best has shifted her attention away from lung cancer to brain cancer, which shares some of the same energy pathways. Forthcoming investigations will use sophisticated, immune-competent mouse models to look at the energy fingerprints of different types of brain cancers and the potential clinical utility of metabolic inhibition.
An approach to generating personalized therapy for brain cancer is “absolutely lacking,” Best says, but metabolomics may well hold the key.