Risk of cognitive impairment in men with advanced prostate cancer treated with NHAs: A systematic review and network meta‐analysis

Abstract

Novel hormonal agents (NHAs) have significantly improved outcomes in men with advanced prostate cancer. However, it remains unclear whether NHAs are associated with subsequent cognitive impairment. Thus, we sought to perform a network meta-analysis to compare the risk of cognitive impairment across NHA types. Databases (PubMed, Embase, Scopus, and Web of Science), trial registries (Clini caltr ial.gov), the European Medicines Agency, and the US Food and Drug Administration drug safety reports were searched from inception through July 30, 2021. Eligible studies were clinical trials evaluating the risk of cognitive impairment between NHAs and placebo/standard care. Two independent investigators extracted the data and performed quality assessments using the Cochrane Risk of Bias Tool and ROBINS-I. We estimated the risk ratios by the frequentist approach and calculated the ranking probabilities of all treatments with the surface under the cumulative ranking probabilities. The primary outcome and secondary outcome were odds ratio (OR) and incidence rate ratio of cognitive impairment, respectively. We identified 15 trials with 14,723 participants comparing HNAs with placebo/standard care. Treatments associated with cognitive impairment, from the most to the least, were enzalutamide (OR, 3.66; 95% confidence interval [CI], 2.84-4.73), apalutamide (OR, 1.76; 95% CI, 1.08-2.87), abiraterone acetate (OR, 1.64; 95% CI, 1.01-2.45), and darolutamide (OR, 1.11 95% CI, 0.51-2.39). After adjustment of treatment time duration, enzalutamide still had the highest risk of cognitive impairment with an incidence rate ratio of 2.17 (95% CI, 1.65-2.78). These findings suggest that NHAs, especially enzalutamide, may increase the risk of cognitive impairment compared with placebo/standard care.

Does hormone therapy impact cognition in patients with prostate cancer? A systematic review and meta‐analysis

“…The difference was detectable from the first month after treatment initiation, 51 and the proportion of patients with perceived cognitive impairment was from two to eight times greater in patients receiving enzalutamide at 2 months 51,53 . These results are in line with those from studies using the Common Terminology Criteria for Adverse Events 12 . In addition, different CNS adverse event profiles were reported for these two treatments, and enzalutamide had a high blood–brain barrier penetrance in murine studies, which could lead to a greater impact on cognition 2 .…”

“…51,53 These results are in line with those from studies using the Common Terminology Criteria for Adverse Events. 12 In addition, different CNS adverse event profiles were reported for these two treatments, and enzalutamide had a high blood-brain barrier penetrance in murine studies, which could lead to a greater impact on cognition. 2 Our meta-analysis points to an increase in perceived cognitive impairment during ADT treatment alone 41,57 or combined with enzalutamide, 51 with a medium-to-large effect size.…”

“…8 Subsequent reviews reported heterogeneous or even contradictory results, and the prevalence rates of cognitive decline during treatment varied between 10% and 69%. 9,10 Two recent meta-analyses focused on adverse events or the incidence of global cognitive impairment 11,12 without identifying the cognitive domains affected. In addition, to our knowledge, no meta-analysis has been performed on both the objective and subjective impact of ADT and NGHT.…”

Neurocognitive effects of androgen deprivation therapy and new hormonal agents in a sample of patients with metastatic prostate cancer

“…In addition, PC patients (between 10% and 69%) experience treatment-related cognitive deficits [ 4 ]. Several studies have implicated treatment-specific relationships between cognitive deficits and systemic PC treatment [ 4 – 6 ]. These cognitive problems may persist for years after treatment [ 7 ].…”

“…However, studies on treatment-specific cognitive effects have yielded inconsistent results [ 5 ]. Thus, a larger empirical base is needed, especially to determine which neurocognitive domains are affected, as emphasized in several reviews [ 4 – 6 ].…”