For patients with locally advanced or metastatic basal cell carcinoma, Sonic hedgehog inhibitors (SSHi) are effective but are associated with primarily partial responses, and the two Food and Drug Administration–approved agents have significant toxicities, results of a systematic review and meta-analysis indicated.
Data on patients with metastatic or locally advanced basal cell carcinoma (BCC) treated with either vismodegib (Erivedge) or sonidegib (Odomzo) showed that the two agents had roughly similar overall response rates (ORR). Vismodegib, however, had a significantly higher rate of complete responses (CR) in patients with locally advanced disease, reported, and , from McGill University, Montreal.
“Common side effects of SHHi therapy tend to incapacitate patients, leading to high discontinuation rates. Over 25% of patients stopped treatment due to side effects. Most side effects are reversible after therapy cessation, except cases of persistent alopecia have been reported,” they wrote in the.
The authors conducted the review to evaluate SHHi as a class and to get a better idea of the efficacy and safety of each agent in the class. They searched the literature to identify all studies using SHHi to treat BCC with vismodegib, sonidegib, itraconazole, or the investigational compound TAK-441, and identified 14 studies focused on vismodegib, 2 on sonidegib, and 1 each on itraconazole and TAK-441.
Of the 18 studies, data from 16 were pooled in fixed-effects linear models to analyze efficacy. The pooled ORR for all patients was 59.6%, “indicating that most patients receiving SHHi achieve at least a partial response.”
Combined ORR results showed a rate of 61.9% for vismodegib, 55.2% for sonidegib, 50% for itraconazole, and 20% for TAK-441, although data for the latter two agents were limited.
In studies looking at locally advanced and metastatic BCC separately, vismodegib was numerically better but statistically similar to sonidegib for locally advanced disease (ORR, 68.8% vs. 56.6%). However, vismodegib was significantly superior to sonidegib for patients with metastatic BCC (ORR, 39.7% vs. 14.7%; P = .007).
The pooled CR for all patients was 23.5%, and there were no CRs with either itraconazole or TAK-441. The combined CR rate for vismodegib was 28% (P = .012). The combined CR rate for sonidegib was just 8.9% and was not statistically significant, the investigators found.
In subgroup analyses for locally advanced BCC, the CR rate for vismodegib was 30.9% for vismodegib (P = .012), “meaning that many patients can expect cure.” In contrast, only 3% of patients treated for locally advanced disease had a CR with sonidegib. The difference between the drugs in this subpopulation was significant (P less than .0001).
Neither drug produced significant CRs in patients with metastatic melanoma, and the pooled clinical benefit rate (all patients with stable disease or better) was 94.9%, with rates similar among all four drugs.
Pooled prevalences of adverse events showed a 67.1% prevalence of muscle spasms, 54.1% prevalence of dysgeusia, and a 57.7% prevalence of alopecia. The proportions of side effects were similar between vismodegib and sonidegib, but sonidegib was associated with a higher prevalence of upper gastrointestinal tract distress than vismodegib.
The study was partially funded by the Canadian Dermatology Foundation. The authors reported having no conflicts of interest.