Retrospective Data Shed Light on Prognostic Value of PD-L1 Expression in Metastatic Papillary RCC

Renal Cell Carcinoma -
stock.adobe.com

PD-L1 positivity was linked with shortened overall survival (OS) in patients with metastatic papillary renal cell carcinoma (pRCC), according to data from a retrospective study published in the European Journal of Cancer.

Findings demonstrated that patients with PD-L1–positive pRCC (n = 19) experienced significantly shorter OS from the initiation of first-line treatment with sunitinib (Sutent) or everolimus (Afinitor) compared with those with PD-L1–negative disease (n = 48; log-rank P = .044).

Additionally, findings from a multivariate analysis from the initiation of first-line treatment (n = 30) showed that PD-L1 expression was significantly associated with OS (HR, 4.0; 95% CI, 1.39-11.39; P = .01).

“These results reinforce clinical data on the expected benefit of immune checkpoint inhibitors in metastatic pRCC treatment, as PD-L1 expression is a factor of poor prognosis in this multicenter cohort,” lead study author Dr Jérémie Naffrichoux, of the Department of Medical Oncology at University Hospital in Tours, France, and colleagues, wrote in a publication of the data.

pRCC is a rare and aggressive tumor associated with a poorer prognosis compared with clear cell RCC (ccRCC), and VEGF TKIs and mTOR inhibitors have demonstrated reduced efficacy in the treatment of patients with pRCC.

In ccRCC, PD-L1 expression has been shown to be prognostic and predictive of response to immune checkpoint inhibitors. Although prior anterior studies done in pRCC did not show any prognostic value for PD-L1 status, these studies focused on those with localized tumors. This retrospective study sought to analyze the prevalence and level of PD-L1 expression in metastatic pRCC and evaluate its association with OS.

To do this, investigators collected data from a subset of patients from a multicenter, retrospective Groupe d′Etude des Tumeurs Uro-Génitales cohort. These patients were at least 18 years of age with metastatic pRCC per RECIST 1.1 criteria and received first-line treatment with sunitinib or everolimus. Out of 138 patients identified, 75 had tumor samples available.

Along with the primary objective of evaluating the prevalence and prognostic effect of PD-L1 in pRCC, secondary end points included the evaluation of the association between PD-L1 and other immune and angiogenic markers.

Combined positive score (CPS) was used to evaluate PD-L1 and PD-1 status. Patients with a PD-L1 CPS of at least 1% were considered PD-L1 positive. Other immune markers, including PD-L2, LAG-3, CAIX, and c-MET, received scores of 0 (no stained cells per immunohistochemistry [IHC]), 1 (low intensity), or 2 (high intensity). Fumarate hydratase (FH) was classified as either positive or negative.

Of the 75 tumor samples analyzed, investigators excluded 5 patients due to negative FH, and 2 more patients were excluded due to inconclusive IHC testing.

Among the 68 patients included, the median age was 62 years (interquartile range [IQR], 55-71). The majority of patients were male (87%), had non–type 1 histology (81%), underwent prior radical nephrectomy (82%), received first-line sunitinib (90%), and had a Karnofsky performance status of at least 80 (82%). The median lactate dehydrogenase level (n = 44) was 216 UI/L (IQR, 172-287). Patients had a favorable (22%), intermediate (38%), poor (18%), or unknown (22%) Heng score.

Furthermore, 27.9% of patients had a PD-L1 CPS of at least 1%, and in those who expressed PD-L1, the median CPS was 7.5% (IQR, 1.0%-30.0%). PD-L1 status was unknown for 1 patient. PD-1 was expressed in 17.6% of patients; 30.9% and 14.7% of patients had low and high expression of PD-L2, respectively; and 7.3% and 11.8% of patients had low and high expression of LAG-3, respectively. Regarding angiogenic markers, the respective rates of low and high expression of CAIX were 30.9% and 19.1%, and the respective rates of low and high expression of c-MET were 26.5% and 16.2%.

Additional data showed that among 60 patients treated with first-line sunitinib, there was a trend in shorted OS for patients with PD-L1–positive pRCC (n = 18) vs those with PD-L1–negative disease (n = 42); however, this OS difference was not significantly different (P = .091).

The multivariate analysis also showed that a poor Heng score was significantly associated with OS (HR, 10.6; 95% CI, 1.75-64.0; P = .01).

A correlative biomarker study demonstrated a positive and strong correlation was observed for LAG-3 and PD-L2 expression, as a higher LAG-3 expression was associated with a higher PD-L2 expression. However, no other positive or negative correlations between biomarkers were identified.

“In our cohort, we found no association between immune and angiogenic pathways, as PD-L1 expression was not associated with CAIX [P = .677] nor c-MET [P >.99] expressions,” the study authors wrote.

Naffrichoux and colleagues noted that the study was limited by its retrospective nature. Additionally, missing clinical data affected the multivariate OS analysis, and given the small subset of patients with PD-L1–positive metastatic pRCC, additional investigation is required to validate the prognostic value of PD-L1 in this patient population.

Reference

Naffrichoux J, Poupin P, Pouillot W, et al. PD-L1 expression and its prognostic value in metastatic papillary renal cell carcinoma: Results from a GETUG multicenter retrospective cohort. Eur J Cancer. Published online May 13, 2024. doi:10.1016/j.ejca.2024.114121