The mechanism of primary resistance to inhibitors of PI3K such as idelalisib has not been well understood. Thus, we conducted a study investigating the mechanisms underlying PI3K inhibitor resistance in CLL using whole exome sequencing (WES), RNA sequencing, and biochemical confirmation. We performed WES on matched tumor and germline samples from a cohort of 28 relapsed patients with CLL who had been treated with idelalisib, pilaralisib, or voxtalisib in our early PI3K inhibitor trials. Of these, 10 patients were considered nonresponders, and we looked for genetic correlations with response or resistance. We found that mutations, particularly in MAPK pathway genes (ie, KRAS, BRAF, MAP2K1), were recurrent among patients who did not respond to PI3K inhibition and were absent entirely from responders. Given these findings, we then examined the role of the extracellular signal–regulated kinase (ERK)/MAPK pathway in idelalisib resistance and found that the nonresponder cells had persistent ERK activation that was not affected by idelalisib, whereas AKT, the direct downstream target, was still inhibited.

Based on these data, we are potentially interested in conducting clinical trials combining ERK inhibitors with idelalisib. For practicing clinicians, it would be reasonable to choose a drug other than idelalisib for patients who have a mutation in one of these pathways. However, the difficulty is that the very abbreviated somatic mutation panels that are available from many laboratories do not necessarily include these mutations. Another area for future research is the identification of mechanisms involved in the development of acquired resistance to PI3K inhibition. Data in patients with Waldenström’s macroglobulinemia show that individuals who acquire resistance to ibrutinib also have ERK activation. It is possible that activation of the ERK pathway might confer resistance to novel agents.