KEY TAKEAWAYS
- Survival outcomes in newly diagnosed glioblastoma patients were analyzed using phase 2 clinical trial data.
- The main goal was to evaluate survival outcomes between non-trial patients and control arm participants in multiple multi-institutional trials.
- Cox regression with variables was used to evaluate patient-level data from 4 separate datasets.
- Clinical trial patients and those treated outside a clinical trial at a prominent academic center had similar survival rates.
- Trial participation did not increase life expectancy after controlling for critical clinical variables.
There may be doubts about the transferability of results from randomized clinical trials due to the tight eligibility criteria used to pick individuals. However, there has been a recent uptick in the potential use of real-world data sources as an external control in conjunction with innovative trial designs. Researchers compared the survival rates of patients with newly diagnosed glioblastoma across clinical trial data and data from the general population. Data on individual patients were analyzed from four different sources. The non-trial results were from a big academic facility that treated 453 patients with a newly diagnosed case of glioblastoma using conventional radiation therapy with concurrent and adjuvant temozolomide. Patients enrolled in studies ranged from those in their institution (n = 211) to those at other institutions (n = 273; n = 89, n = 459). Cox regression analysis controlling for age, sex, degree of resection, KPS, and MGMT status compared non-trial patients with each of the 4 clinical trial datasets.
About 1,484 individual patients’ records were evaluated. Patients outside of studies tended to be older than those participating in randomized controlled trials conducted at many institutions (mean 58.6 vs 56.1 years (, 53.9 years, 55.7 years ; p < = 0.001). There were fewer women in the non-trial group (43.5% vs 35.9%, p = 0.02), more patients with KPS scores below 47% KPS <90 vs 31% KPS <90, p < 0.001), and more patients with MGMT methylation (49% vs 33%, p < 0.001) in the non-trial cohort. Patients in the 4 trial datasets and those outside the trials were similar in every other way. Multivariable analysis adjusting for clinical variables (non-trial patients vs each trial dataset, P > 0.05) found no association between trial participation and better survival. Researchers found no statistically significant differences in survival between patients with glioblastoma treated on multi-institutional clinical trials and those treated outside of a clinical trial at a large academic center when researchers controlled for relevant factors. Their results lend credence to using external control arms based on real-world data in future trials of newly diagnosed glioblastoma.
Source: https://meetings.asco.org/abstracts-presentations/208742
Clinical Trial: https://clinicaltrials.gov/ct2/show/NCT00943826
Rifaquat Rahman, Gilbert Youssef, Steffen Ventz, Robert Redd, Jon McDunn, William Louv, Brian Michael Alexander, Patrick Y. Wen, Lorenzo Trippa/Comparison of survival outcomes of patients with newly diagnosed glioblastoma treated with standard chemoradiation in and outside of clinical trials/J Clin Oncol 40, 2022 (suppl 16; abstr 2051) DOI10.1200/JCO.2022.40.16_suppl.2051
