Clinical Prognostic Factors
Several investigators have attempted to define prognostic criteria for patients with mRCC.
The Memorial Sloan–Kettering Cancer Center (MSKCC; NY, USA) criteria, first published in 1999, defined pretreatment clinical features that were predictive of survival in untreated patients with mRCC who were going to receive cytokine therapy, particularly IFN-α. The MSKCC risk system stratifies patients with mRCC into poor-, intermediate- and favorable-risk categories based on the number of adverse clinical and laboratory parameters present. Poor prognostic factors include a Karnofsky performance status (KPS) of less than 80, time from diagnosis to treatment less than 12 months, serum lactate dehydrogenase (LDH) more than 1.5-times the upper limit of normal (ULN), corrected serum calcium >10.0 mg/dl and hemoglobin less than the lower limit of normal (LLN). Patients in the favorable-risk group have no poor prognostic factors, those in the intermediate-risk category have one or two adverse prognostic factors, and patients with poor-risk RCC have more than two poor prognostic factors. This model was externally validated and extended to include prior radiotherapy and the number of individual-organ metastatic sites (one organ involved vs two or more) among the poor prognostic factors that predicted decreased survival. Using either prognostic classification, median OS was in excess of 2 years in the favorable-risk group and approximately 14 months in the intermediate-risk group; the outcome of patients in the poor-risk group instead was disappointing, with a median OS of 4.9 months following treatment with IFN, a 1-year survival of 20% and a 12-month progression-free survival (PFS) of only 10% using the classical MSKCC prognostic score. Results were only slightly better using the Cleveland Clinic Foundation (CCF)–expanded criteria, which expanded the cohort of poor-risk patients to 28% of the population (as opposed to 11% using the MSKCC criteria), resulting in a 7.3-month median OS.
The MSKCC prognostic score and its modifications have subsequently become a standard by which RCC patients have been assessed and stratified prior to enrollment into clinical trials of targeted agents. Several authors have validated the MSKCC model in patients with mRCC treated with contemporary VEGF-targeted therapies, eventually adding neutrophil and platelet count to the mix of prognostic variables (Table 1). Heng and colleagues evaluated patients who had received first-line treatment with bevacizumab, sunitinib or sorafenib. Four of the five MSKCC-adverse prognostic factors maintained their independent prognostic value: hemoglobin lower than LLN; corrected calcium greater than ULN; KPS of less than 80% and time from diagnosis to treatment initiation of less than 1 year. The authors also identified absolute neutrophil count and platelet count greater than ULN as additional independent adverse prognostic factors. Patients were thus categorized as favorable-risk (no adverse factors; median OS: not reached), intermediate risk (one to two adverse factors; median OS: 27 months) or poor risk (three to six adverse factors; median OS 8.8 months). The International Kidney Cancer Working Group identified five independent prognostic variables (hemoglobin, white cell count, LDH, alkaline phosphatase and calcium) in a dataset encompassing 3748 mRCC patients. Median survival in the favorable-, intermediate- and poor-risk groups were 26.9, 11.5 and 4.2 months, respectively.
Heng et al.'s prognostic model is currently the most widely used and has recently been validated in a large international multicenter dataset; the reported versus predicted number of deaths at 2 years was most similar in the International Database Consortium (IDC) compared with the other models.
Although the definition of poor prognosis varies in published reports, certain features are common to all variants of this category: an interval of less than 1 year from the original diagnosis to the development of metastatic disease, metastases in multiple organs, a low KPS, anemia and elevated serum levels of LDH and calcium. These features are apparently more related to tumor burden and growth kinetics, rather than the underlying tumor biology, and indeed discriminate prognosis equally well, whether applied to cytokine-treated patients or to targeted agents, and regardless of the line of treatment.