Abstract
SMO variants drive resistance to the Smoothened inhibitor vismodegib in basal cell carcinoma.
Major finding: SMO variants drive resistance to the Smoothened inhibitor vismodegib in basal cell carcinoma.
Mechanism: SMO mutants inhibit drug binding or relieve receptor auto-inhibition to confer constitutive activity.
Impact: Patients with resistance to SMO inhibitors may benefit from inhibitors of targets downstream of SMO.
Basal cell carcinoma (BCC) is characterized by hyperactivation of the Hedgehog signaling pathway and is driven by inactivating mutations in the suppressive Hedgehog receptor Patched 1 or activating mutations in Smoothened (SMO), a transmembrane G protein–coupled receptor that activates GLI transcription factors. Despite the clinical benefit of SMO inhibitors such as vismodegib in patients with advanced BCC, resistance remains a challenge. Using genomic analysis of patient biopsies, Sharpe and colleagues identified recurrent SMO mutations in vismodegib-resistant BCCs, implicating reactivation of Hedgehog signaling in drug resistance. SMO mutations in the transmembrane region were enriched post-treatment, suggestive of drug-induced selection. Computational modeling of the SMO drug-binding pocket and functional assays revealed that expression of SMO mutants conferred resistance by inhibiting vismodegib binding and promoted cross-resistance to structurally related SMO inhibitors. Sequencing of distinct tumor regions highlighted intratumor heterogeneity and copy-number alterations in the Hedgehog pathway genes suppressor of fused homolog (SUFU) and GLI2 as potential alternative resistance mechanisms. Consistent with these findings, Atwood, Sarin, and colleagues identified SMO mutations in 50% of resistant BCCs analyzed and showed that these mutations maintained Hedgehog signaling in the presence of SMO inhibitors. Mutations in the SMO ligand-binding pocket were selected for as a mechanism of acquired resistance, whereas constitutively activated SMO mutations in structural pivot regions were present in both untreated and vismodegib-resistant BCCs, suggesting that relief of SMO auto-inhibition may contribute to intrinsic resistance. Expression of either type of SMO variant conferred a growth advantage over cells expressing wild-type SMO in the presence of vismodegib. Importantly, both groups showed that cells expressing SMO resistance variants were sensitive to treatment with inhibitors of downstream signaling components, including atypical protein kinase C ι/λ or GLI2 inhibitors and the bromodomain inhibitor JQ1. Together, these data indicate that SMO mutations drive clinical SMO inhibitor resistance through multiple mechanisms and suggest therapeutic strategies to overcome resistance.