A phase II randomized controlled trial of nicotinamide for skin cancer chemoprevention in renal transplant recipients

Dear Editor, Immunosuppressed organ transplant recipients have an 80‐fold increased risk of squamous cell carcinoma (SCC) and a 16‐fold increased risk of basal cell carcinoma (BCC);1 SCCs are more aggressive and more likely to metastasize in these patients.2 While mammalian target of rapamycin (mTOR) inhibitors may help reduce the incidence of SCC,3 the current mainstay of nonmelanoma skin cancer (NMSC) chemoprevention post‐transplant is oral retinoids, which have side‐effects, including liver and lipid abnormalities, mucocutaneous dryness and teratogenicity.4,5 Nicotinamide (vitamin B3) enhances repair of photodamaged DNA and prevents the inhibitory effects of ultraviolet radiation on the immune system without altering baseline immune reponses.6,7,8 In 386 immunocompetent participants at high risk of having skin cancer, nicotinamide 500 mg twice daily reduced new NMSCs by 23% (P = 0·02), with 20% fewer BCCs and 30% fewer SCCs compared with placebo.9 Actinic keratoses (AKs) were also significantly reduced by nicotinamide in the same phase III randomized controlled trial.9 However, it is unknown whether nicotinamide prevents NMSCs and is safe in immunosuppressed renal transplant recipients. Eligible adult immunosuppressed participants included those who had received a renal transplant ≥ 12 months ago, had stable renal function and a history of ≥ 2 NMSCs in the previous 12 months. Participants were ineligible if they commenced retinoids or mTOR inhibitors within 6 months, or used nicotinamide supplements or AK field treatments within the 4 weeks prior to enrolment. This double‐blind study was conducted at the Royal Prince Alfred Hospital (Sydney, Australia; Australian New Zealand Clinical Trials Registry ACTRN12612000628842) after ethical approval was received. All participants provided written informed consent. Participants were enrolled (by A.C.C. and R.A.D.) and randomized 1:1 to receive either nicotinamide 500 mg twice daily or matched placebo (Insolar™; Blackmores, Warriewood, Australia) by centralized randomization (permuted blocks; National Health and Medical Research Council Clinical Trials Centre) stratified by lifetime NMSC history (< 6 or ≥ 6 NMSCs), oral retinoid use and mTOR inhibitor use. Compliance was monitored by tablet counts at each visit (A.C.C). Skin cancer checks were performed at baseline and twice monthly up to 6 months by dermatologists blinded to treatment allocation (D.L.D., P.M.L.). AKs on the face, scalp, forearms and hands were counted at baseline and twice monthly up to 6 months by a dermatology Fellow blinded to allocation (A.C.C). All squamous lesions were reviewed by a single (blinded) histopathologist (C.A.M.) to ensure consistent reporting of SCC differentiation.10 Blood and urine samples were taken at baseline, 2 and 4 weeks, and 2, 4 and 6 months for full blood count, electrolytes, renal and liver function tests, drug levels and urine microalbumin/creatinine ratio. Blood pressure and weight were measured at baseline and twice monthly for 6 months. The primary end point was the number of new NMSCs (BCC + SCC, including invasive and in situ SCC) up to 6 months. Secondary end points included new BCCs, new SCCs, AK counts up to 6 months and safety. The planned sample size was 80 participants, in order to provide 80% power to detect a 50% reduction in the 6‐month NMSC rate at the 5% level of significance, assuming that NMSC counts followed a Poisson distribution with a mean of 1·5 for the placebo group, allowing for ≤ 10% noncompliance. However, owing to slow recruitment because of an overestimation of the eligible participant population at our site, the study was stopped early; 22 participants were recruited over 1·5 years. Analyses were by intention to treat. As per the provision specified in the statistical analysis plan, a negative binomial model was used to analyse NMSC, SCC and BCC data, owing to overdispersion that rendered the Poisson model inappropriate. Models included an offset term to account for variation in follow‐up duration. The 2‐monthly postbaseline AK data were analysed using a mixed model for repeated measures, including treatment group, baseline value, time point and a time‐by‐treatment group interaction as covariates. From August 2012 to March 2014, 25 participants were assessed for eligibility and 22 were randomized [11 in each group; Figure S1 (see Supporting Information)]. The groups had similar baseline characteristics (Table 1). Follow‐up was from August 2012 to August 2014. Median compliance was 93% and 98% for the placebo and nicotinamide groups, respectively. Baseline characteristics Data are mean ± SD (range) or n (%). NMSC, nonmelanoma skin cancer; BCC, basal cell carcinoma; SCC, squamous cell carcinoma; AKs, actinic keratosis; mTOR, mammalian target of rapamycin. Baseline characteristics Data are mean ± SD (range) or n (%). NMSC, nonmelanoma skin cancer; BCC, basal cell carcinoma; SCC, squamous cell carcinoma; AKs, actinic keratosis; mTOR, mammalian target of rapamycin. The 6‐month NMSC rate was not significantly lower for the nicotinamide group [mean 2·7, 95% confidence interval (CI) 1·4–5·3; total 30 cancers] compared with placebo (mean 4·2, 95% CI 2·2–7·8; total 45 cancers); however, the numeric trend was dominated by one patient in the placebo group with 20 NMSCs (eight BCCs and 12 SCC). The estimated relative rate difference (RRD) was 0·35 (95% CI −0·62 to 0·74; P = 0·36). The 6‐month BCC rate was 0·6 (95% CI 0·2–2·0) for the nicotinamide group (total of seven BCCs) and 1·9 (95% CI 0·7–5·0) for placebo (21 BCCs) (estimated RRD 0·67, 95% CI −0·4 to 0·9; P = 0·14). The 6‐month SCC rate was 2·1 (95% CI 1·0–4·2) for the nicotinamide group (total of 23 SCCs) and 2·2 (95% CI 1·1–4·5) for placebo (24 SCCs), although one patient in the placebo arm with six clinically apparent SCCs died without histological confirmation of these lesions. The estimated RRD in histologically‐confirmed SCCs was 0·07 (95% CI −1·51 to 0·65; P = 0·89). Table 2 summarizes the BCC subtypes and SCC differentiation in each group. Histologically‐confirmed nonmelanoma skin cancers Data are n (%). SCC, squamous cell carcinoma; BCC, basal cell carcinoma. Histologically‐confirmed nonmelanoma skin cancers Data are n (%). SCC, squamous cell carcinoma; BCC, basal cell carcinoma. Actinic keratoses counts were not significantly lower in participants receiving nicotinamide compared with placebo by 1% at 2 months (95% CI −26·0 to 22·0; P = 0·93), 6% at 4 months (95% CI −20·0 to 26·0; P = 0·62) and 16% at 6 months (95% CI −7·0 to 34·0; P = 0·15). There were no statistically significant differences between groups in the number or types of adverse events nor in changes in weight, blood pressure, full blood count, creatinine, liver function or microalbumin/creatinine ratio. Oral nicotinamide 500 mg twice daily was associated with a statistically nonsignificant 35% relative difference in the rate of NMSCs (P = 0·36) and statistically nonsignificant 16% reduction in AKs (P = 0·15); these magnitudes are similar to the statistically significant reductions in NMSCs and AKs observed in the phase III ONTRAC study (Oral Nicotinamide To Reduce Actinic Cancer)9, with greater chemopreventive efficacy of nicotinamide seen in participants with higher numbers of previous NMSCs. No safety concerns were detected (Tables S1 and S2; see Supporting Information). Nicotinamide has an excellent safety profile at doses ≤ 3 g daily,9,11 although it has not been evaluated in transplant recipients. This pilot study suggests that nicotinamide may be an effective and safe chemopreventive agent in renal transplant patients. Phase III clinical trials are now warranted. We are most grateful to our study volunteers for their generous participation in this project. We thank Blackmores Ltd (Warriewood, Australia) for supply of the placebo and nicotinamide tablets, and clinical trials pharmacists Katherine Snowden, Tuong‐Vi Phan and Romana Cecchele for coordinating tablet distribution. Thanks also to Mr David Espinoza for assistance with participant randomization, and Mr Tamati Paki for database setup and maintenance. Funding sources: This project was funded by the Australasian College of Dermatologists Scientific Research Fund. A.C. Chen was supported by an Australian Postgraduate Award and a University of Sydney Postgraduate Scholarship in Dermatology. The funders had no role in study design, data collection, data analysis, manuscript preparation and/or publication decisions. Conflicts of interest: none declared. Fig S1. Study flow diagram. Table S1. Safety data analysis. Table S2. Adverse and serious adverse event assessment.