Specialist Referral in Prostate Cancer

Evidence

This pragmatic, multicenter, international randomized clinical trial evaluated routine cardiovascular-specialist referral among patients with prostate cancer recruited at participating sites in eight countries. Eligible participants had been diagnosed during the prior year, had first received androgen-deprivation therapy within the preceding six months, or were planning to begin that therapy within the next month. The analysis included 2487 patients with prostate cancer. The study compared usual care with usual care plus routine referral to an internist or cardiologist; patients already taking a statin and with systolic blood pressure at or below the protocol target were ineligible. This population and eligibility structure matter because the results describe people entering the trial with scope for additional cardiovascular risk-factor management, not every person with prostate cancer. (pmid:42669035)

Study personnel used a web-based, computer-generated schedule to assign eligible participants to the referral or usual-care groups. Randomization was stratified by center and androgen-deprivation-therapy use, with permuted blocks used for allocation concealment. In the referral group, internists or cardiologists implemented a systematic strategy that included blood-pressure management, statin medication irrespective of cholesterol level, smoking-cessation encouragement, and diet and exercise guidance. Usual-care participants could still receive cardiovascular care through their routine clinicians. The intervention therefore tested a package centered on systematic specialist referral and protocol-directed risk-factor management rather than isolating the effect of a single medication, visit, or counseling component. (pmid:42669035)

The primary outcome was a hierarchical composite assessed by a win ratio. It incorporated cardiovascular death, myocardial infarction, stroke, heart failure, suboptimal cholesterol, and suboptimal blood pressure. During follow-up, the win ratio favored the specialist-referral intervention at 1.60 (95% CI, 1.42-1.81). The reported result was mostly attributable to lower cholesterol in the intervention group, in the context of greater protocol-mandated statin use. Thus, the composite result reflects both clinical-event components and risk-factor-control components; it should not be read as a result based solely on cardiovascular death, myocardial infarction, stroke, or heart failure. (pmid:42669035)

For the prespecified clinical-event comparison, there was no difference between groups in time to cardiovascular death, myocardial infarction, stroke, or heart failure, with a subdistribution hazard ratio of 1.08 (95% CI, 0.79-1.49). Exploratory analyses of individual clinical components reported estimates in both directions, and none was statistically significant. The investigators also reported lower cholesterol throughout follow-up among participants assigned to an internist or cardiologist. Reported close-out systolic blood-pressure values were similar between groups, while the discussion described greater use of statin and antihypertensive medications in the referral group. These results separate a measurable difference in risk-factor treatment and cholesterol control from the trial’s clinical cardiovascular-event comparison. (pmid:42669035)

Implementation was not confined perfectly to assigned groups. Some usual-care participants visited an internist or cardiologist for cardiovascular risk management at their routine physician’s request, while some intervention-group participants did not adhere to the referral strategy. These crossovers are part of the study context: the comparison was between assignment to routine systematic referral plus usual care and usual care in a pragmatic setting, rather than between wholly exclusive forms of cardiovascular management. The trial also collected safety outcomes and reported no cases of myositis; it reported no between-group differences for the listed major bleeding, kidney-injury, and liver-injury outcomes. (pmid:42669035)

Analysis — Trial interpretation

The reported pattern is internally consistent with the intervention’s design. A referral pathway that specified statin use and structured attention to cardiovascular risk factors was associated, in this randomized comparison, with greater protocol-directed treatment and lower cholesterol. Because the hierarchical primary outcome included suboptimal cholesterol and suboptimal blood pressure alongside major cardiovascular events, improvement in a risk-factor component could materially influence the win ratio even when the clinical-event comparison did not differ. The investigators explicitly attributed most of the win-ratio result to cholesterol control. Statistical significance for the win ratio therefore does not establish clinical significance or demonstrate a reduction in cardiovascular death, myocardial infarction, stroke, or heart failure. The clinical-event estimate is compatible with effects in more than one direction within its confidence interval. This is a randomized trial, so the principal interpretation concerns the tested referral-and-management strategy in the enrolled prostate-cancer population, rather than an observational association. It did not isolate which element of the multicomponent specialist strategy accounted for the cholesterol difference. (pmid:42669035)

Limitations

The principal reported limitation was a lower-than-expected incidence of cardiovascular events, which limited power to detect differences in clinical outcomes. Crossovers also reduced study power: usual-care participants could adopt components of the intervention, and intervention-group participants could be nonadherent. Consequently, the investigators could not exclude a clinically meaningful beneficial or harmful effect of routine specialist referral on clinical outcomes. The eligibility criteria, protocol-directed statin approach, and pragmatic care setting further define the bounds of the evidence: the trial directly evaluated this systematic referral strategy in enrolled patients with recently diagnosed prostate cancer or prostate cancer newly treated with androgen-deprivation therapy. Its findings should therefore be kept distinct from claims about other populations, different referral models, or individual components of the strategy. (pmid:42669035)