Postmarketing Cases of Enfortumab Vedotin–Associated Skin Reactions Reported as Stevens-Johnson Syndrome or Toxic Epidermal Necrolysis

Abstract

This case series study uses data from the FDA Adverse Event Reporting System to investigate a potential association between enfortumab vedotin treatment and Stevens-Johnson Syndrome and toxic epidermal necrolysis.

During routine surveillance, the US Food and Drug Administration (FDA) Division of Pharmacovigilance identified postmarketing cases of Stevens-Johnson Syndrome (SJS) and toxic epidermal necrolysis (TEN) associated with enfortumab vedotin (EV) from the FDA Adverse Event Reporting System (FAERS). To investigate a potential association between EV and SJS/TEN, we evaluated all sources for postmarketing cases of SJS/TEN and report our findings.

Methods

We reviewed FAERS, PubMed, and Embase from December 18, 2019, the FDA approval date for EV, through October 7, 2020, to identify cases of SJS/TEN with EV. The analysis was completed manually over the course of several months in 2020. Cases of SJS/TEN diagnosed by a dermatologist or reporting confirmatory biopsy results were considered confirmed cases. Cases were excluded if there was insufficient information to confirm the diagnosis of SJS/TEN or temporal association with EV, or if an alternative etiology was considered the likely cause of SJS/TEN. Under statutory authority pursuant to public health, no approval from institutional review boards or informed consent was required for this analysis.

Results

We identified 8 cases of serious skin reactions characterized by the reporters as SJS/TEN from FAERS. No additional cases were identified from PubMed or Embase at the time of this analysis. The diagnosis of SJS/TEN was confirmed by a dermatologist and/or biopsy findings consistent with SJS/TEN in 5 cases. The median time to onset of SJS/TEN was 11 (range, 9-21) days from the start of EV. The descriptive characteristics of the cases are presented in the Table. All cases reported serious outcomes, including death in 4 cases for which the cause of death was attributed to SJS/TEN. Other serious outcomes included admission to the burn unit in 4 cases.

Table. Descriptive Characteristics of Patients With SJS and TEN Associated With Enfortumab Vedotin.

Characteristic	No. (n = 8)
Age, y
Mean	75
Median (range)	75 (72-81)
Not reported	1
Sex
Female	1
Male	7
Country
United States	8
Year of event
2020	8
Reported indication for use
Urothelial cancer	7
Not reported	1
Time to onset, d
Mean	13
Median (range)	11 (9-21)
Not reported	1
No. of enfortumab vedotin doses received prior to SJS/TEN onset
2	6
3	1
Not reported	1
Relevant diagnostic informationa
Positive skin biopsy result	5
Diagnosed by a dermatologist	4
Admission to burn unit/ICU	5
Treatment of serious skin reactionb
Corticosteroids	7
Etanercept	1
IVIG	1
Unspecified topical agents	1
Concomitant antineoplastic therapies
Pembrolizumab/carboplatin	1
None	6
Not reported	1

Abbreviations: ICU, intensive care unit; IVIG, intravenous immunoglobulin; SJS, Stevens-Johnson Syndrome; TEN, toxic epidermal necrolysis.

^aMore than 1 method of diagnostic confirmation may have been reported per case.

^bMore than 1 treatment may have been reported per case.

Because SJS/TEN occurences are rare but serious events, the existence of well-documented postmarketing reports with EV is of clinical significance. Moreover, we find the rapid accumulation of cases over an approximate 12-month marketing period a concerning observation. To further investigate, we considered the incidence of locally advanced urothelial cancer in the US per year, which is the population likely to receive EV (12 494-40 000 patients), and calculated a reporting rate of 200 cases per 1 000 000 patients, which well exceeds the expected incidence rate of SJS/TEN (ie, 1-7 cases per 1 000 000 patients).^1,2 We believe this reporting rate likely underestimates the rate of SJS/TEN because we would expect other therapies to be used before EV based on its approved indication (ie, smaller denominator). In addition, underreporting is a known limitation of spontaneous reporting systems like FAERS (ie, underestimated numerator).

Discussion

Although the mechanism for skin toxic effects with EV is unknown, it may be related to the inhibitory effects of EV on nectin-4 expression, which is expressed by epithelial tissues, including skin.³ Skin reactions that occurred in 55% of clinical trial patients were included in the US prescribing information on approval.⁴ This further suggests that EV’s inhibitory effects on nectin-4 may contribute to the development of skin toxic effects, including SJS/TEN.⁴

The US prescribing information for EV was recently updated to include SJS/TEN with recommendations to permanently discontinue EV for suspected SJS/TEN.⁴ This revision is intended to increase clinicians’ awareness of the risk for SJS/TEN and mitigate serious outcomes by improving the likelihood of early identification and intervention. Since our completed analysis, we identified 2 cases of SJS/TEN published in the literature, one of which was initially submitted to FAERS and included in our analysis of this safety issue.^5,6

Although our analysis is limited by the quality of spontaneous adverse event reporting and other limitations inherent to spontaneous reporting systems (ie, underreporting, reporting biases), drug safety surveillance programs remain important because clinical trials are limited in the ability to detect rare safety events like SJS/TEN. The FDA relies on observant clinicians to report suspected drug-related adverse events to MedWatch. This is critical to characterize the overall safety profile of FDA approved products. We encourage continued reporting of adverse events with EV to FDA via the MedWatch portal at https://www.fda.gov/Safety/MedWatch/default.htm.