Role of Cytoreductive Nephrectomy Unclear for Dedifferentiated Kidney Cancer

AUSTIN, Texas -- Cytoreductive nephrectomy improved overall survival (OS) by more than a year for patients with metastatic renal cell carcinoma (mRCC) with sarcomatoid or rhabdoid (S/R) dedifferentiation treated with immunotherapy in an observational study, but the difference did not achieve statistical significance.

Median OS was 30 months with surgery versus 17 months without, representing a 21% reduction in the survival hazard (95% CI 0.47-1.33, P=0.37). Statistical significance remained elusive due to multiple confounders.

"Our interpretation of the data is that it's not sufficient to circumvent [nephrectomy] unless we've also seen that systemic therapy is going to be a new driver," said Andrew Hahn, MD, of the University of Texas MD Anderson Cancer Center, during a poster presentation of the findings at the International Kidney Cancer Symposium.

"I still think there's going to a role for cytoreductive nephrectomy; it just has to be teased out further" he told MedPage Today. "Patient selection is going to be challenging, ultimately choosing high-volume centers that have the ability to tease out [the impact of nephrectomy] to a significant level."

RCC with S/R dedifferentiation is highly aggressive and associated with poor prognosis. However, several studies have shown that the tumors often respond to immune checkpoint inhibition (ICI), which has become the cornerstone of first-line treatment of mRCC, Hahn noted. The role of cytoreductive nephrectomy remains unclear, as previous studies yielded inconsistent or inconclusive results.

To continue the investigation of cytoreductive nephrectomy in mRCC with S/R dedifferentiation, investigators at MD Anderson and Memorial Sloan Kettering Cancer Center in New York City retrospectively reviewed records for 157 patients treated with ICI. The study population included 85 patients with intermediate-risk disease and 57 with high-risk disease. Additionally, 78 patients had sarcomatoid dedifferentiation of the primary tumor at diagnosis of metastatic disease, 49 had rhabdoid dedifferentiation, and 30 patients had both.

All but 39 patients underwent cytoreductive nephrectomy. Of the 118 patients who had surgery, resection was delayed in 29 (24.6%) cases.

The primary outcomes were ICI treatment duration (and relationship to cytoreductive nephrectomy status) and OS. Investigators adjusted for the timing of nephrectomy (upfront or delayed), comorbidity index, type of dedifferentiation, risk score, number of prior therapies, and epithelial histology. Such adjustment would provide for a more accurate estimate of the presumed impact of nephrectomy status on the outcomes, Hahn explained.

Analysis of baseline characteristics showed that almost 90% of the patients had clear cell RCC, some 52.2% had two or three metastatic sites, and 65% had received no systemic therapy prior to ICI. The most frequently used ICI regimens were nivolumab (Opdivo) plus ipilimumab (Yervoy), at 40.8%; ICI plus a tyrosine kinase inhibitor, for 27.4%, and single-agent ICI, in 28.0% of patients.

The primary analysis showed that patients who did not undergo cytoreductive nephrectomy had a median time on ICI therapy of 3.7 months, compared with 5.9 months with surgery, a nonsignificant difference (HR 0.98, 95% CI 0.65-1.47).

"Further studies which explore the impact of CN [cytoreductive nephrectomy] in this patient population are needed, including improved tools for patient selection in this setting," Hahn and colleagues concluded.

"This retrospective analysis is based on the assumption that dedifferentiation subtype does not change the effect of CN on outcomes, but we accounted for dedifferentiation type as a potential confounder in our multivariable model," they added. "There is risk for selection bias when retrospectively evaluating CN at two high-volume cancer centers, and we attempted to account for this bias by reporting Charlson Comorbidity Index and IMDC [International Metastatic RCC Database Consortium] risk score across groups."

Noting the large imbalance in nephrectomy status, poster session moderator Shawn Dason, MD, of Ohio State University in Columbus, questioned whether the study had statistical power to detect a true difference in OS. He also expressed skepticism about the reliability of pathologic assessment to rule out dedifferentiation, alluding to patients without dedifferentiation who nonetheless have poor outcomes.

"Pathology is challenging," said Hahn. "In our hands, about 30% of the time we find sarcomatoid or rhabdoid, and the rest come out on nephrectomy. It's a huge bias, and you're right: You're going to end up missing a lot of patients with no cytoreductive nephrectomy that maybe did have poor outcomes and maybe did have some dedifferentiation. It's an argument for molecular markers or blood markers of dedifferentiation."

With respect to the analysis' statistical power, Hahn said the analysis was unplanned, and investigators did not specify numbers for nephrectomy status. He pointed to the substantial number of patients who had delayed nephrectomy as a key confounder in the analysis.

Despite the study failing to find significant advantage, "we're not saying 'no cytoreductive nephrectomy in sarcomatoid or rhabdoid dedifferentiation,'" said Hahn. "You just have to be careful, really thoughtful. It's likely that IO [immuno-oncology] combinations are driving this."

Comment