Introduction:
Current prognostic tools for identifying patients at risk of recurrence after treatment of their localized renal cell carcinoma (RCC) rely primarily on tumor pathology and have suboptimal predictive abilities. Patient-derived factors such as body composition and inflammation are gaining importance for prognostication in cancer. We hypothesized that patient factors such as pre-operative skeletal muscle mass and serum markers of systemic inflammation can have prognostic utility in patients with localized RCC. We investigated the individual and combined utility of pre-operative skeletal muscle index (SMI) and modified Glasgow Prognostic Score (mGPS) for estimating survival and recurrence in patients undergoing surgery for localized RCC.
Methods:
We performed a retrospective chart review of 352 patients with localized RCC who underwent nephrectomy 2005-2019. Skeletal muscle index (SMI) was measured using pre-operative computed tomography images. Patients met criteria for sarcopenia by BMI- and sex-stratified thresholdS: for BMI <30, male: SMI <47cm2/m2, female: SMI <38 cm2/m2; for BMI ≥30, male: SMI<54cm2/m2, female: SMI <47 cm2/m2. We performed multivariable and Kaplan Meier analysis of associations between sarcopenia, mGPS and overall (OS), recurrence-free (RFS), and cancer-specific survival (CSS). Variables were analyzed independently and combined into risk groups: low-risk: non-sarcopenic, low mGPS; medium-risk: sarcopenic, low mGPS; high-risk: non-sarcopenic, high mGPS; very-high risk: sarcopenic, high mGPS. Time-dependent receiver operating characteristic (ROC) curves analyzed risk groups compared to Stage, Size, Grade and Necrosis score, the current gold standard for prognostication in localized RCC.
Results:
In our cohort of majority T3 (63%), N0 (84%) patients, 39.5% were sarcopenic, 19.3% had elevated mGPS. Mean follow up time was 37.7 months. Sarcopenia was associated with inferior OS (HR 1.64, p=0.006), CSS (HR 2.01, p=0.009), trended towards worse RFS (HR 1.51, p=0.055). mGPS was associated with worse OS (HR 1.72, p=0.012), RFS (HR 1.85, p=0.018), not with CSS. Risk groups demonstrated incremental associations with outcomes. Compared to low-risk group, medium, high, and very high-risk groups had increasing association with worse OS, RFS, CSS (p-values from grouped multivariable analysis 0.087, 0.015, 0.004, respectively). In the very high-risk group, hazards ratios were 2.62 for OS (p<0.001),1.83 for RFS (p=0.069), 2.55 for CSS (p=0.034). ROC analyses demonstrated higher area under the curve for risk groups compared to SSIGN score at 5 years: OS (0.703 versus 0.676), RFS (0.790 versus 0.776), and CSS (0.837 versus 0.778); differences not statically significant.
Conclusion:
Preoperative sarcopenia and elevated mGPS were independently associated with decreased RFS, OS, and CSS in patients with localized RCC. Furthermore, the combination of SMI and mGPS predicted 5-year outcomes comparatively to the SSIGN score. This may have implications for prognostication and treatment decisions for urologists and medical oncologists, both for surgical planning and for clinical trial enrollment stratification for neoadjuvant and adjuvant RCC therapies.
Funding: N/A
Image(s) (click to enlarge):
SARCOPENIA AND MODIFIED GLASGOW PROGNOSTIC SCORE HAVE COMBINED PROGNOSTIC VALUE IN PREDICTING POST-SURGICAL OUTCOMES IN LOCALIZED RENAL CELL CARCINOMA
Category
Kidney Cancer > Localized
Description
Poster #110
-
Presented By: Michelle Irene Higgins
Authors:
Michelle Irene Higgins
Dylan J. Martini
Dattatraya H. Patil MBBS, MPH
Reza Nabavizadeh MD
Sean Steele
Milton Williams MD
Aarti Sekhar MD
Sarah P. Psutka MD, MS
Kenneth Ogan MD
Mehmet Asim Bilen MD
Viraj A. Master MD, PhD