IN-DEPTH REVIEW A Review of the Risk of Cutaneous Squamous Cell Carcinoma after Vismodegib Therapy

Vismodegib is a hedgehog inhibitor (HHI)


INTRODUCTION
A literature search of PubMed and Ovid MEDLINE was conducted with terms "vismodegib" AND "squamous cell carcinoma." Between May 2013 and January 2022, 245 articles were identified, and 2 additional articles were found through handsearch of the literature (Figure 1). Removing duplicates resulted in 168 articles, of which 130 articles were excluded based on lack of relevance to the research question. 38 total articles were assessed for eligibility, and 14 were ultimately excluded based on lack of applicability. 25 articles were ultimately included in the review.
Of the 25 articles included in the review, 14 reported patient outcome data and statistics on cSCC development after vismodegib exposure, whereas the remaining 11 were review articles or commentaries on the topic. In this review, a total of 2576 patients were treated with vismodegib and 197 cases of cSCC were reported.
The duration of vismodegib treatment until cSCC development ranged from 2 weeks to 2.7 years across all studies included. The median duration of treatment was 3 months ( Table 1).
The initial phase II trial of vismodegib enrolled 96 patients but included safety data of patients from other trials for a total of 400 subjects receiving therapy for a median of 10 months. While the most frequently reported adverse events were muscle spasms, dysgeusia, diarrhea, and alopecia, no secondary malignancies, like cSCC, were recorded. 13 After FDA approval, an openlabel study included 119 patients with BCC receiving vismodegib for a median duration of 5.5 months. Adverse events with a median safety follow-up of 6.5 months were tabulated. One patient developed recurrence of a prior cSCC, however, this adverse event was not deemed to be related to vismodegib. 3 In the years following vismodegib's approval, several studies reported 8,9,10,11,14,15,19,20,24 cSCC development after vismodegib initiation. In one patient who developed cSCC 13 months after vismodegib initiation, genetic testing from the primary BCC and the subsequent cSCC revealed shared tumor drivers, suggesting vismodegib may induce a phenotypic change from BCC to cSCC. 11 A retrospective study of 19 patients with BCC assessing vismodegib's ability to induce metatypical change of BCC demonstrated metatypical change in 5 cases and squamous change in 2 cases. 25 A retrospective cohort study evaluated vismodegib and cSCC development in 180 patients, separated into two groups based on exposure (55 patients) or lack of exposure (125 patients) to vismodegib. 2 The overall rate of subsequent non-BCC cancers was higher in the non-exposed group than in the exposed group. However, a Cox proportional hazards regression analysis revealed an increased risk of non-BCC cancer, most frequently cSCC, in the exposed group compared to the control group (Hazard ratio: 8.12; 95% confidence interval, 3.39-11.96, P<0.001). 2 Invasive cSCC was more common than in-situ cSCC, suggesting vismodegib may potentiate more aggressive cSCC. 2 The median time between initial therapy and diagnosis of secondary malignancy was 0.9 years (ranging from 2 weeks to 2.7 years). However, whether cSCCs emerging just two weeks after initiation could be attributed to vismodegib remains controversial. 6 Additionally, the results of this study have not been reproducible. Figure 1. Study identification and selection using the preferred reporting items for systematic reviews and meta-analyses (PRISMA) flow diagram. *Records excluded due to topic or lack of applicability to the research question. Rates of cSCCs did not significantly differ between those treated with vismodegib and not treated. The exposure-adjusted-incidence rate of cSCC after vismodegib exposure was 0.06 cases per patient-year compared to 0.07 cases per patient-year in the non-exposed group.

RESULTS
Two large, multicenter studies 1,4 , one a retrospective cohort and the other a prospective observational registry, found that the risk of cSCC after exposure to vismodegib use was comparable to the control groups with no exposure. In 2017, a retrospective cohort study of 1675 patients compared patients treated with vismodegib from phase I and II clinical trials and patients who received standard surgical treatment for BCC. Multivariate proportional hazards analysis did not reveal an increased risk of cSCC with vismodegib use (Hazard ratio: 0.57; 95% confidence interval, 0.28-1.16). 1 In 2022, a prospective, multi-center observational registry analyzed over 500 patients with locally advanced BCC not treated with vismodegib. Overall, the study found comparable rates of cSCC between the vismodegib-treatment group (12% of patients developed cSCC, exposureincidence rate of 0.06 cases per year) and the non-vismodegib treatment group (12% of patients developed cSCC, exposureincidence rate of 0.07 cases per year). 4 Since vismodegib's approval by the FDA for the treatment of invasive or metastatic BCC, there have been anecdotal concerns that vismodegib can induce cSCC. While a retrospective cohort study 2 seemed to verify that vismodegib is associated with an increased risk of cSCC, these results have not been reproduced. Based on the available findings, vismodegib does not appear to increase the risk for subsequent cSCC. It is plausible that patients who have already developed BCC are at risk of other skin cancers, like cSCC, due to shared risk factors.
As an HHI, vismodegib's role in cSCC development may be due to select tumor cells circumventing the hedgehog pathway, and instead utilizing other growth pathways, including Ras/MAPK. 2,18 Further investigation has demonstrated the importance of cilium in tumor pathway switching, with a decrease in cilium gene expression in BCCs resistant to HHIs. This results in low hedgehog pathway activation and an increase in Ras/MAPK pathway activation. 16,17 Limitations of this review include only using two databases to search for articles and the limited number of published articles that address the research question. The quality of certain study types included in the review, such as case reports, is another limitation of this article.
The current data available on the relationship between vismodegib use and the subsequent development of cSCC is not definitive and does not demonstrate causality. With a high efficacy rate in the treatment of BCCs not amenable to surgical resection or radiation, vismodegib should not be avoided due to concerns of causing secondary cutaneous neoplasms. However, the data available does suggest that patients treated for BCC with vismodegib should be closely monitored by dermatologists to evaluate lesions for response to treatment, changes, or signs of progression. When in doubt, the lesion should always be re-biopsied. Additional large-scale, prospective, multi-center studies should be conducted to definitively evaluate the risk of vismodegib use in the development of cSCC.