Patient Characteristics Differ in Some Clinical Trials From Real-World Population

Get Permission

Patients with metastatic renal cell carcinoma treated with tyrosine kinase inhibitors in real-world clinical practice tend to be older and sicker than the patients enrolled in pivotal clinical trials of these agents. In addition, patients with metastatic renal cell carcinoma treated with the mTOR inhibitor temsirolimus (Torisel) in the real-world setting also were older than those in the pivotal clinical trial but were more likely to have favorable-disease risk factors. Applying clinical trial results to dissimilar patient populations could result in harm, according to a study in the Journal of Oncology Practice.

“Clinical research with more inclusive eligibility criteria is needed to appropriately guide real-world practice,” concluded Aaron P. Mitchell, MD, of Duke University Medical Center, Duke Clinical Research Institute, Durham, North Carolina, and colleagues.

Comparing Patients

Investigators from Duke Clinical Research Institute and ACORN Research, Memphis, Tennessee, conducted a cohort study to compare baseline characteristics of patients with metastatic renal cell carcinoma in phase III clinical trials of new targeted therapies with 438 patients in a retrospective registry composed of academic (Duke) and community (ACORN Network) practices. Applying the eligibility criteria from the phase III clinical trials, the researchers determined that 39% of registry patients would have been excluded from the clinical trial testing the agent they received as first-line treatment in the real-world practice setting.

The agents most commonly used as first-line therapy for patients in the registry study were the tyrosine kinase inhibitors sunitinib (Sutent) (289 patients), sorafenib (Nexavar) (59 patients), and pazopanib (Votrient; 31 patients) and the mTOR inhibitor temsirolimus (59 patients). Data were abstracted from the published results of phase III clinical trials for these four agents.

Differences in Study Populations

Registry patients receiving the tyrosine kinase inhibitors were older than the clinical trial participants by an average of 4.1 years and “significantly less likely” to have favorable-risk disease by Memorial Sloan Kettering Cancer Center (MSKCC) criteria, 30.1% vs 43.8% (P < .001) for clinical trial patients. Registry patients were more likely to have poor-risk disease by MSKCC criteria, 7.4% vs 2.9% (P < .001), and to have impaired performance status (Eastern Cooperative Oncology Group > 1), 11.1% vs 0.6% (P < .001).

Overall survival was better for patients receiving sunitinib in clinical trials, 26.4 months vs 21.3 months for registry patients, but registry patients receiving sorafenib had better overall survival, 30.1 months vs 19.3 months for trial participants. Median overall survival was 22.9 months for trial participants receiving pazopanib and had not been reached for registry patients.

Among patients receiving temsirolimus, patients from the registry were older than the clinical trial participants by an average of 7.8 years; more likely to have favorable-risk disease, 16.9% vs 0% (P < .001) for clinical trial patients; and less likely to have poor-risk disease, 10.2% vs 69.4% (P < .001). Further, in regard to performance status, 11.9% of registry patients were impaired vs 80.4% of trial participants (P < .001).

“Our findings suggest that temsirolimus is being used in practice to treat disease of lower risk than that in which it has been studied,” the authors noted. “Interestingly, overall survival was shorter for patients receiving temsirolimus (4.8 months) [outside of a clinical trial] than for corresponding trial participants (10.9 months), despite the more favorable MSKCC risk and performance status of these patients.” The authors called these “the most intriguing results of our survival analysis.”

External Validity

The authors noted that their analysis builds on studies comparing “patients with cancer treated in clinical trials vs those excluded from clinical trials who received different treatments; in contrast, we were able to compare patients with cancer receiving the same treatments within and outside of clinical trials. This higher level of granularity allowed us to demonstrate that concerns regarding trial generalizability are borne out in clinical practice, in that the targeted therapies for metastatic renal cell carcinoma are commonly used in patients not represented in the landmark trials,” the researchers wrote.

“The question of external validity—or generalizability—is increasingly important in the era of targeted therapies,” the authors observed. “These newer agents are more easily tolerated than the cytotoxic chemotherapeutics they are replacing, allowing for treatment of patients with more comorbidities and advanced disease and increasing the need to study such patients.”  ■

Mitchell AP, et al: J Oncol Pract. September 1, 2015 (early release online).