Selenium and Vitamin E for Non–Muscle-Invasive Bladder Cancer (2024)

Key Points

Question Can selenium and vitamin E supplementation prevent recurrence and progression of non–muscle-invasive bladder cancer (NMIBC)?

Findings In this randomized clinical trial of 270 adults, supplementation with selenium was not associated with a decreased risk of NMIBC recurrence; vitamin E supplementation was associated with a significantly increased risk of recurrence. Neither selenium nor vitamin E was associated with progression or overall survival.

Meaning These findings suggest that vitamin E supplements may be harmful to patients with NMIBC.

Importance Selenium and vitamin E have been identified as promising agents for the chemoprevention of recurrence and progression of non–muscle-invasive bladder cancer.

Objective To determine whether selenium and/or vitamin E may prevent disease recurrence in patients with newly diagnosed NMIBC.

Design, Setting, and Participants This multicenter, prospective, double-blinded, placebo-controlled, 2 × 2 factorial randomized clinical trial included patients with newly diagnosed NMIBC recruited from 10 secondary or tertiary care hospitals in the UK. A total of 755 patients were screened for inclusion; 484 did not meet the inclusion criteria, and 1 declined to participate. A total of 270 patients were randomly assigned to 4 groups (selenium plus placebo, vitamin E plus placebo, selenium plus vitamin E, and placebo plus placebo) in a double-blind fashion between July 17, 2007, and October 10, 2011. Eligibility included initial diagnosis of NMIBC (stages Ta, T1, or Tis); randomization within 12 months of first transurethral resection was required.

Interventions Oral selenium (200 μg/d of high-selenium yeast) and matched vitamin E placebo, vitamin E (200 IU/d of d-alfa-tocopherol) and matched selenium placebo, selenium and vitamin E, or placebo and placebo.

Main Outcome and Measures Recurrence-free interval (RFI) on an intention-to-treat basis (analyses completed on November 28, 2022).

Results The study randomized 270 patients (mean [SD] age, 68.9 [10.4] years; median [IQR] age, 69 [63-77] years; 202 male [75%]), with 65 receiving selenium and vitamin E placebo, 71 receiving vitamin E and selenium placebo, 69 receiving selenium and vitamin E, and 65 receiving both placebos. Median overall follow-up was 5.5 years (IQR, 5.1-6.1 years); 228 patients (84%) were followed up for more than 5 years. Median treatment duration was 1.5 years (IQR, 0.9-2.5 years). The study was halted because of slow accrual. For selenium (n = 134) vs no selenium (n = 136), there was no difference in RFI (hazard ratio, 0.92; 95% CI, 0.65-1.31; P = .65). For vitamin E (n = 140) vs no vitamin E (n = 130), there was a statistically significant detriment to RFI (hazard ratio, 1.46; 95% CI, 1.02-2.09; P = .04). No significant differences were observed for progression-free interval or overall survival time with either supplement. Results were unchanged after Cox proportional hazards regression modeling to adjust for known prognostic factors. In total, 1957 adverse events were reported; 85 were serious adverse events, and all were considered unrelated to trial treatment.

Conclusions and Relevance In this randomized clinical trial of selenium and vitamin E, selenium supplementation did not reduce the risk of recurrence in patients with NMIBC, but vitamin E supplementation was associated with an increased risk of recurrence. Neither selenium nor vitamin E influenced progression or overall survival. Vitamin E supplementation may be harmful to patients with NMIBC, and elucidation of the underlying biology is required.

Trial Registration isrctn.org Identifier: ISRCTN13889738

Bladder cancer is the 12th most common cancer worldwide¹; in high-income countries, more than 90% are transitional cell carcinomas of urothelial origin, and most patients (75%-85%) present with non–muscle-invasive bladder cancer (NMIBC) (Union for International Cancer Control stages Ta, T1, and Tis).² Patients with NMIBC are initially treated by transurethral resection of bladder tumor and adjuvant intravesical therapy.³ Recurrence occurs in up to 80% of patients⁴; progression to muscle-invasive bladder cancer (stages T2 or higher) occurs in up to 45% of patients diagnosed with initial stage T1 disease.⁵

Given the frequency of recurrence and progression and the chronic nature of the disease, NMIBC may be amenable to chemoprevention.⁶ Selenium and vitamin E have previously been identified as promising agents^7-9; notwithstanding, more recent clinical trials of selenium and/or vitamin E in the primary and secondary cancer prevention settings have shown no effect^10,11 or a detrimental effect.¹² Between 2005 and 2007, we established a trial protocol with the aim of providing important insights into the use of selenium and vitamin E as adjuvant therapies for NMIBC (SELENIB trial).

From July 17, 2007, to October 10, 2011, SELENIB recruited patients with newly diagnosed NMIBC to a double-blinded, placebo-controlled, 2 × 2 factorial randomized clinical trial from 10 UK hospitals. Eligible patients were 18 years or older with newly diagnosed, pathologically confirmed urothelial NMIBC who were able to give informed consent. Ethnicity data were not collected because they were not considered to be relevant for a study of this nature. Patients were required to be randomized within 12 months of initial transurethral resection of bladder tumor. All patients provided written informed consent. SELENIB was coordinated by the Cancer Research UK Clinical Trials Unit at the University of Birmingham. The trial was conducted in accordance with the principles of the Good Clinical Practice guidelines and the Declaration of Helsinki.¹³ It was approved by the UK Medicines and Healthcare Products Regulatory Agency. Research ethics approval was gained from East Midlands–Derby Research Ethics Committee, and the trial was overseen by an independent data monitoring committee. This report follows the Consolidated Standards of Reporting Trials (CONSORT) reporting guideline.¹⁴ The trial protocol can be found in Supplement 1.

Patients were randomly assigned equally to 1 of 4 groups: oral selenium (200 μg/d of high selenium yeast, 364% recommended daily allowance¹⁵) and matched vitamin E placebo, vitamin E (200 IU/d d-alfa-tocopherol, 600% recommended daily allowance¹⁶) and matched selenium placebo, selenium and vitamin E, or placebo and placebo. Treatment allocations were blinded. Randomization was stratified by recurrence risk group (high vs low or intermediate¹⁷) and treatment center. Patients took 1 tablet (selenium or placebo) and 1 capsule (vitamin E or placebo) once daily with food for up to 5 years. Patients were otherwise treated according to contemporaneous European Association of Urology guidelines.¹⁷ Patients attended a SELENIB follow-up clinic every 6 months for up to 5 years after randomization during which treatment adherence, toxic effects, and disease status were recorded.

The primary outcome was recurrence-free interval (RFI), measured as the time from study entry to recurrence. For patients without recurrence at the time of analysis, the interval was censored at the date the patient was last known to be recurrence free. Recurrences at the first 3-month cystoscopy checkup were excluded. Secondary outcome measures included progression-free interval and overall survival (eMethods in Supplement 2). Quality of life was assessed at each follow-up visit.

The primary hypothesis was addressed on an intention-to-treat basis. As a 2 × 2 factorial design, analysis consisted of 2 comparisons: (1) those patients randomized to selenium vs those randomized to the associated placebo, stratifying by vitamin E allocation; and (2) those patients randomized to vitamin E vs those randomized to the associated placebo, stratifying by selenium allocation. Interaction was not expected. Treatment groups were compared by Kaplan-Meier estimates of recurrence-free and progression-free interval; log-rank tests were used to test the hypothesis of no difference between treatments. Hazard ratios (HRs) from Cox proportional hazards regression models compared treatments, both unadjusted and adjusted for known prognostic factors. Similar methods were used for secondary outcomes. These analyses were completed on November 28, 2022.

All tests used a 2-sided P < .05 to indicate statistical significance, and all analysis was performed in R software, version 4.2.0 (R Foundation for Statistical Computing) using lmtest, survival, survminer and cowplot, ggplot2, and RColorBrewer for graphs.

The study randomized 270 patients (mean [SD] age, 68.9 [10.4] years; median [IQR] age, 69 [63-77] years; 202 male [75%] and 68 female [25%]), with 65 receiving selenium and vitamin E placebo, 71 receiving vitamin E and selenium placebo, 69 receiving selenium and vitamin E, and 65 receiving both placebos (Figure 1). Baseline characteristics are shown in the Table. A total of 256 (95%) of all tumors were pure transitional cell carcinomas, and 226 (84%) had papillary morphology (eTable 1 in Supplement 2). The trial failed to recruit to its prespecified target of 515 patients and was halted because of slow accrual. Two hundred twenty-eight patients were followed up for more than 5 years. To September 26, 2018, median overall follow-up was 5.5 years (IQR, 5.1-6.1 years); with the intervention of the COVID-19 pandemic, ongoing follow-up beyond this time point became unfeasible. All patients are included in the analysis.

Across 3 approaches to assess adherence (returned tablets or capsules, diary, and patient recollection), the median percentage of days taking trial treatment was more than 95% (IQR, 90%-99%) (eTables 2 and 3 in Supplement 2). The median treatment duration was 1.5 years (IQR, 0.9-2.5 years). Of 270 participants, 264 (98%) had more than 3 years of follow-up or died within 3 years of randomization. Standard-of-care treatments received by participants are detailed in eTables 4 to 6 in Supplement 2.

In total, 1957 adverse events were reported; 85 were serious adverse events, and all were considered unrelated to trial treatment. The most common adverse events were fatigue (279 [14%]), cough and/or cold (217 [11%]), and dermatitis (190 [10%]); there was no difference in adverse events between treatment arms (eResults and eTables 7-9 in Supplement 2).

Of 122 recurrences, 60 (49%) occurred in the selenium arm and 62 (51%) in the placebo arm. For selenium, there was no statistically significant difference in RFI (HR, 0.92; 95% CI, 0.65-1.31; P = .65); median RFI was not reached in either arm (Figure 2A). Of the recurrences, 72 (59%) occurred in the vitamin E arm and 50 (41%) in the placebo arm. Vitamin E was associated with a statistically significant decrease in RFI (HR, 1.46; 95% CI, 1.02-2.09; P = .04); median RFI was 3.3 years (IQR, 0.89 years to not reached) for vitamin E and was not reached for the placebo arm (Figure 2B). Recurrence-free survival estimates at yearly intervals for each treatment comparison are shown in eTable 10 in Supplement 2. We observed a 13% difference in RFI at 5 years for placebo vs vitamin E (59.6% vs 46.5%), with the study originally designed for an absolute difference of 12% at 5 years. Adjusted analyses of the primary outcome were also undertaken (eTables 11 and 12 in Supplement 2).

Overall, 37 patients had disease progression; no significant differences in progression-free interval were observed with selenium or vitamin E (Figure 3). Fifty-three patients died, and no significant differences in overall survival were observed with either selenium or vitamin E (Figure 4). No significant differences in quality of life were observed between the arms (eFigure in Supplement 2).

We observed no benefit of selenium for recurrence, progression, or overall survival in patients with NMIBC. We observed an association between vitamin E supplementation and an increased risk of recurrence but no association with progression or overall survival. The assumptions of the factorial trial design were demonstrated to be correct. The randomization used contemporary risk categorization as a stratification factor; over time, risk categorization has changed.^3,17 For enabling up-to-date risk categorization and applicability to current practice, data were collected on each of the contributory factors.

SELENIB is, to our knowledge, the first trial to investigate the use of selenium and vitamin E for preventing recurrence and progression in patients newly diagnosed with NMIBC. In 2011, the Selenium and Vitamin E Cancer Prevention Trial (SELECT) concluded that dietary supplementation with vitamin E (400 IU/d all-rac-α-tocopheryl acetate) significantly increased the risk of prostate cancer among healthy men¹²; a secondary analysis in 2012 investigated bladder cancer incidence, demonstrating no preventive effect of selenium or vitamin E alone or combined on bladder cancer incidence within the SELECT population.¹⁰

Trials of dietary supplements in cancer prevention are important, as demonstrated here. Our data show no evidence of benefit from selenium and evidence of harm from vitamin E—a compound readily available over the counter. Future micronutrient and/or vitamin chemoprevention studies in NMIBC should carefully consider choice of compound, underlying biology, preclinical evidence, and study design.^18,19

This study has some limitations, including the fact that patient accrual fell below the intended 515 due to an unexpectedly high proportion of ineligible patients, the elderly age of the patient population with accompanying comorbidities and medication, and the proposed trial duration. Furthermore, trial funding was withdrawn in 2010, resulting in the decision to end recruitment and treatment in 2011. Consequently, study participants received considerably less selenium and vitamin E than intended, and 245 fewer patients than required were recruited; thus, these analyses are underpowered.

In this randomized clinical trial of selenium and vitamin E in patients with newly diagnosed NMIBC, selenium supplementation did not reduce the risk of disease recurrence, whereas vitamin E supplementation was associated with an increased risk of recurrence. Neither selenium nor vitamin E influenced progression or overall survival. These findings suggest that vitamin E supplementation may be harmful to patients with NMIBC, and elucidation of the underlying biology is required.

Accepted for Publication: August 21, 2023.

Published: October 17, 2023. doi:10.1001/jamanetworkopen.2023.37494

Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2023 Bryan RT et al. JAMA Network Open.

Corresponding Author: Richard T. Bryan, PhD, Bladder Cancer Research Centre, Institute of Cancer and Genomic Sciences, University of Birmingham, Edgbaston, Birmingham B15 2TT, UK (r.t.bryan@bham.ac.uk).

Author Contributions: Dr Bryan had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Drs Zeegers and Cheng contributed equally.

Concept and design: Bryan, Lewis, Wallace, James, Billingham, Zeegers, Cheng.

Acquisition, analysis, or interpretation of data: Bryan, Pirrie, Abbotts, Mayco*ck, During, Lewis, Grant, Bird, Devall, James, Billingham, Zeegers, Cheng.

Drafting of the manuscript: Bryan, Pirrie, Abbotts, Mayco*ck, James, Cheng.

Critical review of the manuscript for important intellectual content: Bryan, Pirrie, During, Lewis, Grant, Bird, Devall, Wallace, James, Billingham, Zeegers, Cheng.

Statistical analysis: Pirrie, Mayco*ck.

Obtained funding: Bryan, James, Billingham, Zeegers, Cheng.

Administrative, technical, or material support: Bryan, Abbotts, Lewis, Grant, Bird, Devall, Wallace, Zeegers.

Supervision: Bryan, Pirrie, Wallace, James, Billingham, Zeegers, Cheng.

Conflict of Interest Disclosures: Dr Bryan reported receiving grants from Cancer Research UK and nonfinancial support from Pharma Nord during the conduct of the study, receiving personal fees from Cystotech ApS outside the submitted work, and serving as an unpaid charity trustee for Action Bladder Cancer UK. Dr Pirrie reported receiving grants from Cancer Research UK during the conduct of the study. Dr Billingham reported receiving grants from Cancer Research UK during the conduct of the study. No other disclosures were reported.

Funding/Support: The SELENIB study and the Bladder Cancer Prognosis Programme were funded by grant C1343/A5738 from the Cancer Research UK, the University of Birmingham, and the Birmingham & The Black Country and West Midlands North & South Comprehensive Local Research Networks. Staff at the Cancer Research UK Clinical Trials Unit are supported by core funding grant C22436/A25354 from Cancer Research UK. Pharma Nord provided the study drugs free of charge.

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Group Information: The SELENIB Investigators appear in Supplement 3.

Data Sharing Statement: See Supplement 4.

Additional Contributions: We thank the patients who took part in the trial; the 15 investigators from 10 recruiting centers and their research staff; and the staff from the Cancer Research UK Clinical Trials Unit, University of Birmingham, including Gareth Bicknell, PhD, for data curation during the trial and Siân Lax, PhD, for contributions to the study. We also acknowledge the contribution of the Trial Steering Committee (Noel Clarke, FRCS[Urol], Bart Kiemeney, PhD, and Per-Uno Malmstrom, MD), the independent data monitoring committee (Chris Parker, FRCR, Gareth Griffiths, PhD, and Howard Kynaston, FRCS[Urol]). Gareth Bicknell, PhD, University of Birmingham, helped with biospecimen handling, curation, and storage. Siân Lax, PhD, University of Birmimgham, helped with manuscript preparation and submission. Noel Clarke, FRCS(Urol), The Christie Hospital Manchester, University of Manchester; Bart Kiemeney, PhD, Radboud University Medical Centre; Per-Uno Malmstrom, MD, Uppsala University; Chris Parker, FRCR, The Royal Marsden National Health Service Foundation Trust; Gareth Griffiths, PhD, University of Southampton; and Howard Kynaston, FRCS(Urol), Cardiff University, served on the data monitoring committee. These individuals have given their permission to be acknowledged and received no compensation for their contributions.