Absence of Basal Cell Carcinoma in Irradiated Childhood Cancer Survivors of Black Race: A Report from the St. Jude Lifetime Cohort Study Free

Childhood cancer survivors are at a risk of subsequent neoplasms (SN), which comprise basal cell carcinoma (BCC) in approximately 50% to 60% of cases (1, 2). BCC risk is predominantly associated with curative doses of ionizing radiation incurred during cancer therapies (1, 3). Similarly, ionizing radiation exposure has been shown to increase the risk for BCC in atomic bomb survivors, radiologic health care workers, and individuals irradiated for benign medical conditions (4–13).

The incidence of de novo BCC in the general population is dependent upon demographics and geographic ultraviolet radiation (UVR) exposures; therefore, estimated rates vary appreciably. It is evident, however, that in the general population, whites develop BCC at higher rates than blacks (14, 15). A recent report from the United States suggested that whites had a 70-fold higher incidence of BCC than age- and sex-matched blacks (14). In addition, individuals of fair complexion with higher UVR exposure and exposed at younger ages are at highest risk (15). Reduced occurrences in blacks and non-sun–exposed anatomic locations in whites have led some to hypothesize that discrete pathogenic mechanisms may contribute to these variable presentations (16).

Despite the notable influence of race on rates of BCC in populations not exposed to ionizing radiation, the same is not clearly defined in cancer survivors treated with radiotherapy. Compared with irradiated non-white cancer survivors, irradiated whites have been shown to be at significantly increased risk for BCC by some (2), but not other investigators (3, 17); however, outcomes for blacks specifically were not reported. Most studies of populations other than cancer survivors either did not report outcomes by race or were limited by the non-black racial homogeneity of their cohorts (4–8, 10–12). Apart from those occurring in individuals with nevoid basal cell carcinoma syndrome (NBCCS; refs. 18–20), the few reports of BCC in previously irradiated blacks are limited to individuals treated with scalp radiation for tinea capitis (21, 22). These observations led us to investigate the influence of race on the occurrence of BCC in previously irradiated adult survivors of childhood cancer.

The St. Jude Lifetime Cohort Study (SJLIFE) is an Institutional Review Board–approved, ongoing clinical study designed to facilitate longitudinal evaluation of health outcomes among adult survivors of pediatric malignancies. Active patient accrual has been occurring since 2007. Informed consent for this study was obtained for all participants. The study methods have been described previously (23). Briefly, subjects eligible for SJLIFE were previously treated for childhood cancer at St. Jude Children's Research Hospital (Memphis, TN), had survived ≥10 years from diagnosis, and were ≥18 years old at study enrollment. For the purposes of the current study, only those participants who were previously treated with ionizing radiation and had completed an on-campus assessment prior to June 30, 2015, were included. Treatment exposures, cumulative chemotherapy doses, radiation doses to individual anatomic sites, and medical events during and after therapy were abstracted from medical records. All retrospective and prospective data in the current report were collected under the SJLIFE protocol. At the SJLIFE baseline clinical assessment, participants completed comprehensive questionnaires that provided self-reported health outcomes, demographic factors, and medical history. Individuals underwent risk-based screening for chronic health conditions, including, SNs, per the Children's Oncology Group Long-Term Follow-Up Guidelines for Survivors of Childhood, Adolescent, and Young Adult Cancer (24). SNs were confirmed by central review of pathology reports following ascertainment of previously occurring, through questionnaire/patient report, and newly identified SNs, through prospective clinical assessment that included skin examination by SJLIFE clinicians. Suspicious lesions were referred for additional evaluation by a dermatologist. In the current analysis, radiation-associated SNs were restricted to skin (BCC and melanoma), brain, thyroid, and breast. Given prior studies reporting a lack of association with ionizing radiation, squamous cell carcinomas were not considered in the analysis of radiation-associated neoplasms (6, 13).

We calculated an expected number of each type of radiation-associated SN in irradiated blacks, assuming the same rates of development as were seen in irradiated white survivors, accounting for the effects of primary cancer diagnosis, calendar year of diagnosis, attained age, and sex. Specifically, we fitted a piecewise exponential model with these categorical covariates to irradiated whites' person-time data of each type of the radiation-associated SNs, using the logarithm of person-years (PY) as the offset. The fitted model was applied to irradiated blacks' person-time data of the same SN type to obtain their expected count. Multiple occurrences within individual survivors were accounted for using generalized estimating equations modifying the piecewise exponential models above. Observed-to-expected ratios and 95% confidence intervals (CI) were calculated by SN type in irradiated blacks based on Poisson probability models for the observed counts given the expected counts.

Among 1,746 eligible survivors with 33,147 PY of follow-up, 320 SNs occurred in 160 individuals. The median (range) follow-up time was 28.8 (11.1–52.5) years. The ratio of observed-to-expected non-BCC radiation-associated SNs in irradiated blacks was 0.88 (30 observed/34.2 expected; 95% CI, 0.59–1.25). Although not statistically significant, irradiated blacks had lower than expected subsequent thyroid and breast cancers. We expected 56.1 BCCs in irradiated blacks (mean maximum radiation dose, 33.9 Gy; range, 10.0–99.6 Gy), yet observed zero (95% CI, 0.00–0.07; Table 1). Similarly, when considering only the first occurrence of any SN, 19.5 BCCs were expected but none observed (95% CI, 0.00–0.19; Table 2).

We identified an unexpected absence of BCC in irradiated blacks in SJLIFE. These findings underscore a similar pattern among three distinct cohorts (Table 3). Shore and colleagues were the first to report an absence of BCC in blacks treated with irradiation for tinea capitis (median follow-up 25.7 years; delivered doses of 300–600 rad). Despite a clear propensity for irradiated individuals to develop BCC (RR, 3.8; 95% CI, 2.8–5.2), none developed in blacks (3,010 PY) compared with 41 in whites (21,763 PY; P = 0.003; ref. 9). Extended follow-up of this cohort (median 39.3 years) identified only three BCCs in irradiated blacks, whereas irradiated whites remained at 10-fold higher risk for BCC (RR, 10.0; 95% CI, 3.2–31.4) compared with irradiated blacks (21). Similarly, Schwartz and colleagues investigated the occurrence of BCC after total-body irradiation (9.2–15.75 Gy) in transplant survivors with a median follow-up duration of 6.6 years (range 0.3–36.2). Despite the occurrence of 197 BCCs in 3,512 whites (17), none occurred among 60 blacks (W.M. Leisenring; personal communication). Finally, investigators from the Childhood Cancer Survivor Study (CCSS) identified 1,116 BCCs in 7,527 irradiated whites compared with zero in 313 blacks after a median duration of follow-up of 25.1 years (6.5–38.9; ref. 25). Of note, only 39 (12%) of the black survivors in CCSS were also included in our study cohort.

Together, these findings highlight a provocative series of observations suggesting that irradiated blacks are at a substantially lower risk for BCC than expected based upon the increased rate seen in irradiated compared with nonirradiated whites. Even taking into consideration the occurrence of three BCCs in irradiated blacks reported by Shore and colleagues, our finding of no BCC in irradiated blacks contrasts the earlier and more frequent occurrence of BCC observed in previously irradiated whites. This observation remains unexplained within the existing literature, leaving only speculation as to the underlying mechanism.

One might anticipate an additive effect of high intensity, short duration, and ionizing radiation on UVR carcinogenesis, in which case, we would expect a substantial increase in the occurrence of BCC in irradiated, compared with nonirradiated, blacks (Fig. 1). The observed lack of increase among irradiated blacks emphasizes the need for a better mechanistic understanding of BCC pathogenesis in irradiated cancer survivors.

A previous report from the CCSS suggested that darker skin pigmentation is associated with a reduced risk for BCC after ionizing radiation (3). Similarly, the substantially lower risk of BCC in blacks compared with whites irradiated for tinea capitis led investigators to speculate regarding the potentially modifying effects of UVR exposure on carcinogenesis from ionizing radiation (21). However, if this difference can be attributed to the photoprotective effects of melanocytes against UVR alone (26), one might anticipate blacks to experience a reduced, yet still tangible rate of BCC, in contrast to the complete absence we have observed in cancer survivors. Likewise, one might also expect reduced radiation treatment efficacy in black cancer cohorts. Few malignancies are treated with radiation alone, leaving this question difficult to address. Zaki and colleagues reported decreased 5-year overall survival for blacks compared with whites treated with radiation monotherapy for Hodgkin lymphoma (68% vs. 78%; ref. 27). The same survival disparity has not been observed in children treated for Hodgkin lymphoma with a combination of radiation and systemic chemotherapy (28, 29).

Alternatively, genotypic rather than phenotypic mechanisms may be responsible for this seemingly protective effect in blacks. It is well recognized that genetic variants can cause greater DNA damage, inhibit DNA repair, and impair chemotherapy transport mechanisms and metabolism (30). Germline mutations in the tumor suppressor gene PTCH1 have been linked to NBCCS, an autosomal dominant condition in which, among other anomalies, individuals are at substantially higher risk for multiple BCC (31). Despite a reduced rate of BCC in black individuals with NBCCS (32), black individuals exposed to ionizing radiation may develop numerous BCC within previous radiation fields (18). Conversely, protective germline variants may contribute to the reduced risk of BCC we have observed in irradiated black survivors. Given the ongoing genomics initiatives within childhood cancer survivor cohorts, such as SJLIFE and CCSS, our findings come at a crucial juncture in time when focused investigations to better understand potentially underlying genetic mechanisms can be performed. If identified, the implications of these findings may extend beyond surveillance strategies, impacting upfront treatment approaches for survivors of all races and ethnicities.

No potential conflicts of interest were disclosed.

Conception and design: M.J. Ehrhardt, M.M. Hudson, L.L. Robison

Development of methodology: M.J. Ehrhardt, M.M. Hudson, L.L. Robison

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): M.J. Ehrhardt, N. Bhakta, M.M. Hudson, L.L. Robison

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): M.J. Ehrhardt, Q. Liu, Y. Yasui, D.A. Mulrooney, M.M. Hudson, L.L. Robison

Writing, review, and/or revision of the manuscript: M.J. Ehrhardt, N. Bhakta, Q. Liu, Y. Yasui, M.J. Krasin, D.A. Mulrooney, M.M. Hudson, L.L. Robison

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): M.J. Ehrhardt, N. Bhakta

Study supervision: M.J. Ehrhardt, M.M. Hudson, L.L. Robison

We are grateful for the considerable efforts of the research, clinical, and administrative staff who support SJLIFE and the survivors and their families from whom it is our privilege to learn.

The St. Jude Lifetime Cohort Study is supported by the Cancer Center Support (CORE) grant (CA21765) to St. Jude Children's Research Hospital, U01 CA195547 1 (principal investigator: M.M. Hudson), and the American Lebanese Syrian Associated Charities (ALSAC), Memphis, TN.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.