Dr. Fallah on limitations of pooled analysis of cytoreductive nephrectomy

In this video, Jaleh Fallah, MD, discusses limitations of the Journal of the National Cancer Institute paper, “Cytoreductive nephrectomy in the era of immune checkpoint inhibitors: a U.S. FDA pooled analysis.” Fallah is a medical oncologist with the FDA.

Transcription:

What are the major limitations of the pooled analysis?

Our pooled analysis had a couple of limitations. It was a retrospective analysis, and there were variations across trials in terms of previously collected information. For example, the studies did not indicate if nephrectomy was conducted with curative or cytoreductive intent. And the date of nephrectomy was not available in many of the cases. So, we only included patients who had metastatic disease at diagnosis in whom the indication for nephrectomy would be cytoreduction. This did not allow us to include studies that did not provide this information, such as Checkmate 214, which evaluated ipilimumab [Yervoy] and nivolumab [Opdivo] vs sunitinib [Sutent], and this issue limited the number of patients in our pooled analysis. Another important limitation of this analysis is the lack of randomization. We tried to account for the impact of potential imbalances in baseline characteristics by adjusting for factors that are commonly thought to be involved in patient selection for nephrectomy, such as patient's age and performance status. However, there are likely other factors that were not adequately captured in these trials and could not be fully controlled for, and in the absence of a pre-specified randomization, the results of any analysis on the effect of cytoreductive nephrectomy on outcomes should only be considered exploratory and hypothesis generating. This means that our results only found an association between prior cytoreductive nephrectomy and better outcomes. However, outside a randomized trial, we cannot make any conclusion about the causal relationship between conducting cytoreductive nephrectomy and improved outcomes.

This transcription was edited for clarity.