Incidence of HHV‐6 Reactivation in Drug Reaction With Eosinophilia and Systemic Symptoms and Correlation With RegiSCAR Score: A Retrospective Cohort Study

Drug reaction with eosinophilia and systemic symptoms (DRESS) is a severe cutaneous drug reaction that remains challenging to diagnose [1]. It has been associated with human herpesvirus-6 (HHV-6), the reactivation of which has been suggested to correlate with symptom severity [2]. The RegiSCAR score uses clinical and laboratory criteria to estimate the likelihood of DRESS [3], but does not consider HHV-6 positivity in its assessment. We investigated the incidence of HHV-6 reactivation in patients treated for DRESS at a large academic medical center that utilizes RegiSCAR for diagnostic purposes. We also explored the association between HHV-6 reactivation and RegiSCAR score within our cohort. This retrospective cohort study included patients treated for DRESS at Massachusetts General Hospital from January 2016 to October 2023 (IRB approval no. 2023P001945). Billing codes were utilized to identify patients, and variables including demographics, clinical presentation, and HHV-6 testing were abstracted from the medical record by manual chart review. A RegiSCAR score was calculated for each patient. Only those patients scoring more than or equal to 2, indicating ‘possible’ DRESS or higher, were included; those determined not to meet criteria for DRESS were removed before analysis. Statistical analysis was completed in STATA version 18.0.124 and frequencies were compared across groups, using Fisher's exact test for categorical variables. A p-value of 0.05 was considered significant. Fifty-six patients with a RegiSCAR score of 2 or higher were included. The median age was 61 (range 5–86) years old and 27 (48.2%) were female. Twenty-three patients (41.1%) scored 2–3 on RegiSCAR, 27 (48.2%) scored 4–5, and 6 (10.7%) scored greater than 5 (Table 1). A total of 18 patients were tested for HHV-6, and 4 (22.2% of those tested) were found to be positive. No statistically significant association was found between RegiSCAR score and the completion of HHV-6 testing (p = 0.322), nor between RegiSCAR score and positive HHV-6 result (p = 0.103) (Table 1). In this cohort of DRESS patients, approximately 22% of patients tested were positive for HHV-6. This differs from several previously published studies which have reported HHV-6 reactivation to occur in 50%–62% of patients with DRESS [2, 4], but is fairly consistent with a recent study that observed a 28% positivity rate among 93 patients with probable or definite DRESS [5]. Only 18 of 56 (32.1%) of our patients were tested, resulting in a relatively small cohort that may limit the interpretation of these results. There are no formal guidelines in place for HHV-6 testing at our institution. Given the retrospective design of this study, the specific factors driving HHV-6 testing in the patients for whom this was ordered are uncertain; however, it does not seem that RegiSCAR score factored heavily into this decision (Table 1). Additionally, prior studies have indicated that HHV-6 reactivation occurs 2–4 weeks following symptom onset [2] and have found that testing for reactivation early in disease course is of low clinical utility, with some patients becoming viremic weeks to months after initial diagnosis [6]; a recent study observed the mean latency time to HHV-6 reactivation to be 16 days [5]. The retrospective nature of this study and reliance on medical record review precluded the ability to assess the timing of viral testing; it is possible that patients were tested outside of this optimal window, potentially affecting the positivity rate within our sample. We found no statistically significant association between RegiSCAR score and HHV-6 reactivation, suggesting that HHV-6 may not be a marker of disease severity, though only patients with ‘probable’ or ‘definite’ cases tested positive. Future studies should look to clarify the relationship between HHV-6 status and RegiSCAR score in a larger patient cohort to aid in diagnostic reasoning at institutions utilizing RegiSCAR. Allison Holt: conceptualization, data curation, formal analysis, writing – original draft, writing – review and editing. Baraa Hijaz: writing – review and editing. Joseph Ebriani: writing – review and editing. Bethany Brumbaugh: writing – review and editing. Mike Westmeijer: writing – review and editing. Madeline E. DeWane: writing – review and editing. Michael Stephens: writing – review and editing. John Trinidad: writing – review and editing. Kevin Sheng-Kai Ma: writing – review and editing. Steven T. Chen: conceptualization, writing – review and editing, supervision. The authors have nothing to report. Reviewed and approved by Mass General Brigham IRB; protocol #2023P001945. Steven T. Chen has received honoraria from ArgenX, Pfizer, and Scholar Rock. The remaining authors declare no conflicts of interest. Data are available upon reasonable request from the corresponding author.