Genomic annotation improves MSKCC risk model for renal carcinoma
medwireNews: Adding the mutation status of three genes linked to renal cell carcinoma improves the prognostic value of the Memorial Sloan Kettering Cancer Center (MSKCC) risk model for patients receiving tyrosine kinase inhibitors for metastatic disease, research shows.
Martin Voss (Memorial Sloan Kettering Cancer Center, New York, USA) and colleagues created the genomically annotated model using data from 357 participants of the COMPARZ trial and then independently validated it among 258 participants of the RECORD-3 trial.
They say their findings suggest that the annotated model “could be applied for prognostication of individual patients in future clinical trials.”
The researchers initially investigated the association between six genes of interest (BAP1, PBRM1, TP53, TERT, KDM5C, and SETD2) and cancer-specific outcomes among the patients in the validation cohort, who all had advanced or metastatic renal cell carcinoma and were receiving first-line treatment with sunitinib or pazopanib.
They found that patients who harbored any mutation in BAP1, TP53, or both had significantly better overall survival (OS), at a hazard ratio of 1.57, than patients who carried the wild-type form of both genes. Mutation frequencies were 15% for BAP1 and 11% for TP53.
By contrast, OS was significantly better among the 55% of patients with wild-type PBRM1 compared with those who carried PBRM1 mutations, at a hazard ratio of 1.58.
When the mutation status for these genes was added to the original MSKCC risk model, and a fourth risk group was created, risk categorization changed for around half of the participants.
Specifically, the patients’ risk changed from 24%, 61%, and 15% at favorable, intermediate, and poor risk, respectively, in the original model to 10%, 22%, 30%, and 38%, at favorable, good, intermediate, and poor risk, respectively, in the genomically annotated model.
Median OS was not reached for patients with favorable and good risk using the genomically annotated model but was 30.6 and 17.4 months for those with intermediate and poor risk, respectively. By comparison, median OS was not reached among patients with favorable risk in the original model and was 26.6 and 18.1 months in those with intermediate and poor risk, respectively.
Furthermore, the team found that adding the genomic information to the MSKCC model increased the predictive C-index for OS from 0.595 to 0.637, and also improved correlation with progression-free survival and the proportion of patients who achieved and objective response.
Similar results were observed for the patients in the validation cohort who were receiving treatment with everolimus and sunitinib. In this group, the C-index for OS increased from 0.658 to 0.670 with the addition of the genomic information.
In a comment that accompanies the study in The Lancet Oncology, Neeraj Agarwal (University of Utah, Salt Lake City, USA) and co-authors say that “Voss and colleagues have shown the value of genomic profiling in prognostic models.”
However, they caution that the findings “might come at a time when the MSKCC stratification tool is being phased out,” because allocation to treatment with newer regimens such as cabozantinib or nivolumab plus ipilimumab is based on International Metastatic Renal Cell Carcinoma Database Consortium (IMDC) criteria rather than MSKCC criteria.
Therefore, “the genomic markers assessed by Voss and colleagues could warrant reassessment in the context of the studies that are leading to approval of these regimens,” say Agarwal et al.
And they conclude: “Until this happens, clinical [implementation] of the IMDC criteria (without genomic markers) will most likely prevail.”
By Laura Cowen
medwireNews is an independent medical news service provided by Springer Healthcare. © 2018 Springer Healthcare part of the Springer Nature group