Authored by Aiko Nagayama

The evolution of cancer therapeutics began back in the 1800s, when surgery was the only option. From the 1900s to the 1940s, radiation therapy and chemotherapy emerged as other modalities, and targeted therapy in the 2000s and immunotherapy (IO) followed in the 2010s. Although the targeted therapies and IO demonstrated survival benefit over conventional therapy, there is a struggle in the early phase drug development, where only 7% of patients who were tested for the genetic panel benefit in phase I populations. Moreover, less than 20% of response is seen across all cancers. The precision medicine to apply the right drug to the right patient in oncology and immuno-oncology has become increasingly important.

Dr. Hammerman mentioned that comprehensive approaches such as the Cancer Cell Line Encyclopedia, Project DRIVE, and high-throughput screening using patient-derived tumor xenografts are emerging as effective ways to identify and validate targets. One of the successful examples of a predictive biomarker is the presence of PIK3CA mutation associated with BYL719 sensitivity, which leads to the drug approval based on the pre-selected patients with PIK3CA-mutated tumors. This biomarker enriched design of clinical trials proved the concept, showing that the significantly prolonged median PFS of PIK3CA-mutant cohort treated with alpelisib plus fulvestrant compared to the same cohort treated with placebo plus fulvestrant, whereas the median PFS was not significantly different in the PIK3CA-non-mutant cohort. 

The mechanism of activation of receptor tyrosine kinase MET differs depending on the type of cancer, and large scale pharmacological screening proved to successfully identify hits with activated MET particularly due to amplification, not due to autocrine activation. A MET inhibitor, capmatinib, demonstrated antitumor efficacy in animal models of lung cancer, validating that MET exon 14 mutations, MET amplification, and MET overexpression can be predictive biomarkers. Although the MET exon 14 skipping is found in 3% of NSCLC, almost all patients experienced tumor shrinkage regardless of prior treatment in a clinical trial. Dr. Hammerman also referred to the shRNA and CRISPR dropout screen to identify the synthetic lethal target, including alterations of ARID1A, BRG1, CDKN2A, SMAD4, APC, PTEN and Rb (McDonald et al. Cell 2017). These comprehensive screens can be complemented with in silico target validation.

The question that Dr. Hammerman raised was whether the predictive biomarker of response to IO and gene therapy would be identified. He drew an example of Wnt signaling pathway preventing anti-tumor immunity due to suppressed DC transcripts/chemolines. He pointed out that the new era of translation research requires ways to understand the dynamic state of the immune microenvironment and to elucidate the acquired resistance mechanism to develop novel combinations of therapy. As a great example of targeted therapy development, an attempt of pan-RAF inhibition was described. To overcome the resistance of immune “cold” tumors treated with a combination of BRAF/MEK inhibitors, LXH254, a highly selective RAF inhibitor, was introduced because it was found to overcome the issue of paradoxical activation in RAF inhibitors. cEPIC screen with LHX254 identified ARAF as the top sensitizer hit, and it was later validated in the preclinical study that LXH254 plus trametinib in NRAS mutant melanoma model in vitro showed an anti-tumor effect with less rash in animal model.