Herpes Zoster Following Radiation Therapy: A Single-Institutional Epidemiological Analysis From Patient Database

Herpes zoster (HZ) is common comorbidity among cancer patients, however the relationship between HZ and radiation therapy (RT) has hardly been studied. Here we evaluated the frequency and clinical features of HZ after RT to find the relationship between them. We screened 3192 patients who received any type of external beam radiation therapy in our institute from May 2013 through November 2015. Of those, patients with leukemia were exempted from the study, since bone marrow transplantation is known strong risk factor of HZ. Based on diagnosis database for insurance claim, we identified patients who were diagnosed with HZ by May 2016. All HZ cases were confirmed to have either diagnosis by dermatologists or description of typical skin rash by reviewing patient’s records. Two datasets were integrated to evaluate the incidence rate of HZ after RT. The relevance between the distribution of HZ and the radiation field extent was classified to one of the following: “likely” (rash within or overlapping the radiation field), “possible” (rash related to irradiated nerve root) or “unlikely” (rash without relationship to the radiation field). Information of patient baseline characteristics were gathered, including age, gender, primary cancer site, treatment intent, radiation dose, and concomitant treatments. High risk subgroups for “likely” HZ events were explored using the Fisher’s exact test. Two thousand nine hundred fifty three patients were eligible in this study. Seventy three patients (2.5%) experienced HZ after RT. Relevance between RT and HZ was judged as “likely” in 41 (1.4%), “possible” in 7 (0.2%) and “unlikely” in 25 (0.8%) patients, respectively. Median interval between the initiation of RT and the onset of HZ was 4.4 months, and 73% of the cases were diagnosed within 6 months. HZ within radiation field were more frequently observed among the patients with head and neck cancer (4.7%, p=0.002) and gastrointestinal cancer (2.9%, p=0.05) than other population. Usage of concurrent chemotherapy (2.4%, p=0.01) and definitive intent treatment (2.2% p=0.0004) were statistically significant risk factors for “likely” HZ event. In this population, considerable proportion of patients experienced HZ after RT. Majority of HZ events were observed within the radiation field and the incidence of HZ and RT were temporally close to each other. Our results indicate that HZ should be recognized as one of the toxicity of RT.