Nat cancer: prediction of the response of lung cancer patients to immunotherapy by multimodal genome analysis

February 17, 2020 / biool / -- researchers from Kimmel Cancer Center, Johns Hopkins, Bloomberg Kimmel Cancer Immunotherapy Institute and Johns Hopkins University School of medicine have developed a comprehensive genomic approach, which may help doctors predict which non-small cell lung cancer patients respond to immunosuppressive treatment Immunocheckpoint inhibitors are changing the field of cancer treatment However, these successes are limited because of the lack of biomarkers to determine the most likely response of patients Tumor mutation burden (TMB) is a measure of the number of mutations carried by tumor cells, which is considered as a new reactive biomarker, but TMB value is affected by the purity of tumor samples (the number of tumor and normal cells) A team led by Dr valsamo Anagnost, assistant professor of oncology, has developed a new method for calculating TMB more accurately The researchers also developed a comprehensive response model that combined detailed genomic features with the antigen expression ability of each patient to correct TMB A description of the patent application was published in the inaugural issue of nature cancer This method can also be used to accurately estimate TMB and optimize the prediction of immunotherapy response in patients with lung cancer, colon cancer, melanoma and other solid tumors "Immunotherapy is an exciting treatment for many tumors, but what we really don't know is who will respond to immunotherapy, why, and whether there are specific molecular characteristics that can help predict the response," Anagnostou said Biomarkers currently used to assess patients' response to immunotherapy include the detection of PD-L1 protein in TMB and cancer cells "More and more studies show that TMB is not as predictive as we think," said Victor Velculescu, senior author, Professor of oncology, MD, of the study "Some tumors with high TMB do not respond to immunotherapy, while some tumors with low TMB benefit from immunotherapy We urgently need to develop comprehensive biomarkers to explain the subtle differences of tumor immune system, so as to better understand the clinical course of patients "Anagnostou and his colleagues initially evaluated 3788 tumor samples (from bladder, breast, colon, head and neck, kidney and non-small cell lung cancer and melanoma) from the National Cancer Institute Cancer Genome Atlas Database, as well as 1661 tumor samples from a previously published immunotherapy patient cohort They studied the complexity of TMB estimates observed from the entire exome sequencing and target down generation sequencing They found a significant correlation between TMB and tumor purity - the higher the purity of the tumor, the closer it is to the true TMB of the tumor, and the lower the purity of the tumor, the more likely it is that TMB estimates are inaccurate "The observed TMB is strongly influenced by low tumor purity, a simple concept that is completely underestimated in clinical settings," said Dr noushin niknafs, one of the first authors of the study and a postdoctoral fellow To overcome this limitation, the team developed a calculation method based on tumor purity to estimate the corrected TMB value of each tumor They used information from the cancer genome map to simulate 20000 tumors with different levels of TMB and sequencing coverage, and generated a correction factor for each simulated tumor based on its purity "The correction factor can be summed up as a lookup table," Anagnostou said "For example, if the purity of a tumor sample is 20% to 30%, clinicians can look at the table and see a coefficient multiplied by the sample value to better achieve true TMB "The team also developed a method to modify the TMB from the target sequence data, and in reanalysis of 1661 tumor samples treated with immunosuppressive checkpoint inhibitors, the researchers found that using the modified TMB estimates improved overall survival prediction Next, the team looked at other molecular features that play a role in a patient's response to immunotherapy They sequenced all exons of 104 lung tumors treated with immunosuppressant Through a comprehensive analysis of sequence and structural changes, they found that in some patients who did not respond to immunotherapy, receptor tyrosine kinase (RTK) gene (receptor is a key regulator of cell process including cell proliferation, survival and metabolism) has more active mutations In addition, they found that smoking related mutations predominated in patients who responded to treatment Correction of TMB, RTK mutations, mutant smoking signals and human leukocyte antigen (HLA) germline mutations provided the team with a more accurate correction method The team is working on more immunotherapy patients and generating tumor sequencing data for them "We expect this approach to be incorporated into clinical practice and it can change the way doctors make decisions about patients," said Anagnostou "For example, if clinicians can be sure that a tumor has a high mutation burden, they may choose immunotherapy as an independent treatment, and if a tumor has a low mutation burden, they may choose chemotherapy plus immunotherapy "Reference: valsamo Anagnostou et al, multimodal generic features predict outcome of immune checkpoint blockade in non small cell lung cancer, nature cancer (2020) Doi: 10.1038/s43018-019-0008-8