ESMO 2018: Assessment of Renal Cell Carcinoma Asymptomatic Brain Metastases Patients from GETUG-AFU 26 (NIVOREN) Trial

Munich, Germany (UroToday.com) Brain metastases continue to portend a worse prognosis for patients with metastatic renal cell carcinoma (mRCC). These patients are typically treated with local therapy (stereotactic radiation or surgery) independent of traditional targeted therapies due to the poor efficacy of these agents for intracranial lesions. However, the advent of immune checkpoint inhibitors (ICI’s) begs the question of efficacy in this unique patient population.

In this abstract, the authors look at a prespecified subset of the larger prospective phase 2 GETUG-AFU 26 (NIVOREN) trial, which assesses the real-world efficacy of nivolumab in patients who have progressed on VEGF-targeted 1st line therapy. As this subset is relatively rare, it was evaluated as a subset of a larger prospective study.

They specifically assessed two different subsets of patients with asymptomatic brain metastases:
1) Cohort A – patients without any prior brain focal therapy
2) Cohort B – patients with prior brain focal therapy (primarily stereotactic radiosurgery)

An important inclusion criterion was that these brain lesions had to be asymptomatic with no surrounding edema, not require steroids, surgery or radiation at the time of study entry

Treatment was with nivolumab, as previously discussed. The primary endpoint was the best intracranial response in Cohort A according to modified RECIST criteria allowing target lesions >= 5 mm.

Of the 729 patients in the entire study, 73 (10%) were included in the brain metastases study. Of these, 39 patients were in Cohort A and 34 in Cohort B – this was not randomized as patients had been pre-treated at the discretion of their primary oncologist. The cohorts were enriched with poor risk IMDC patients (~32-34%) compared to the entire cohort. N was given in 2nd or 3rd line in 74% of pts. Median follow up was 16.1 months (m). Among 72 pts, 33 (46%) had no prior local therapy for BM, and 39 (54%) had prior local therapy: 29 had radiation therapy (RT) including 25 pts with stereotactic and 4 with whole brain RT, and 10 had prior surgery plus RT. Interestingly, 5 patients in Cohort A had had prior stereotactic radiation for non-target lesions. The number of BM at baseline was 1 in 45 pts (62.5%), 2 or 3 in 17 (23.6%), >3 in 10 (13.9%).

When looking at outcomes, median PFS was 2.8m [CI95% 2.5-4.2] and 12m-OS was 61.7% [CI95% 48.5-72.5]. Response assessment on BM was available in 64 pts, among which 11 (17.2%) had an objective response.

  Untreated (N = 33)

Prior focal treatment (N = 39)

Overall (N = 72)

Missing

3**

5**

8**

CR

4* (13.3%)

3 (8.8%)

7 (10.9%)

PR

0 (0.0%)

4 (11.8%)

4 (6.3%)

SD

12 (40.0%)

16 (47.1%)

28 (43.8%)

PD

14 (46.7%)

11 (32.4%)

25 (39.1%)

(Table 1)

The CR rate was 4 (13.3%) in Cohort A and 7 (10.9%) in Cohort B. All of the four patients with complete response in Cohort A had unique lesions <= 10 mm at baseline. Neurologic deterioration was observed in 29 pts (41%) and 28 (39%) required steroids. Overall, 35/72 (49%) pts required focal treatment: 27/33 (82%) in Cohort  B and 8/39 (20.5%) in Cohort A.

In terms of intracranial PFS, a better response was seen in Cohort B than Cohort A (HR 0.51, CI 0.27-0.96).  Median extracranial PFS was no difference between the groups.

Ultimately, the authors concluded that In pts without prior BM focal therapy, nivolumab efficacy was poor, and most patients eventually required local treatment for brain progression. Patients receiving nivolumab should receive local therapy for brain metastases prior to systemic therapy.

Discussant Presentation Points

With regards to this trial, Dr. Jones made the following points:

1) This is the first prospective trial of ICI’s in patients with ccRCC brain metastases
- Important group in clinical practice
- BUT the study excluded patients with symptomatic brain metastases, which are probably a more relevant clinical group

2) There is a lower response rate (RR) seen in this study than in melanoma or lung-related brain metastases treated with PD-L1 inhibitors single-agent (20-33%) or with dual-agent therapy in melanoma (46-56%) – unclear why

3) Lower intracranial response rates are seen in TKI’s – so it shouldn’t be discounted altogether

4) Peri-tumoral edema (which resulted in exclusion in the current study) may represent higher tumor infiltration – and may provide benefit

5) This is a small patient population – and the results may be a statistical outlier, either positively or negatively


Presented by: Ronan Flippot, MD, Department of Medical Oncology, Gustave Roussy Institute, Villejuif, France
Invited Discussant: Rob Jones, MD, Professor, the University of Glasgow, Glasgow, Great Britain

Written by: Thenappan Chandrasekar, MD, Clinical Instructor, Thomas Jefferson University, twitter: @tchandra_uromd, @TjuUrology at the 2018 European Society for Medical Oncology Congress (#ESMO18), October 19-23,  2018, Munich Germany