Background

Risk stratification for localized renal cell carcinoma (RCC) relies heavily on retrospective models, limiting their generalizability to contemporary cohorts.

Objective

To introduce a contemporary RCC prognostic model, developed using prospective, highly annotated data from a phase III adjuvant trial.

Design, setting, and participants

The model utilizes outcome data from the ECOG-ACRIN 2805 (ASSURE) RCC trial.

Outcome measurements and statistical analysis

The primary outcome for the model is disease-free survival (DFS), with overall survival (OS) and early disease progression (EDP) as secondary outcomes. Model performance was assessed using discrimination and calibration tests.

Results and limitations

A total of 1735 patients were included in the analysis, with 887 DFS events occurring over a median follow-up of 9.6 yr. Five common tumor variables (histology, size, grade, tumor necrosis, and nodal involvement) were included in each model. Tumor histology was the single most powerful predictor for each model outcome. The C-statistics at 1 yr were 78.4% and 81.9% for DFS and OS, respectively. Degradation of the DFS, DFS validation set, and OS model's discriminatory ability was seen over time, with a global c-index of 68.0% (95% confidence interval or CI [65.5, 70.4]), 68.6% [65.1%, 72.2%], and 69.4% (95% CI [66.9%, 71.9%], respectively. The EDP model had a c-index of 75.1% (95% CI [71.3, 79.0]).

Conclusions

We introduce a contemporary RCC recurrence model built and internally validated using prospective and highly annotated data from a clinical trial. Performance characteristics of the current model exceed available prognostic models with the added benefit of being histology inclusive and TNM agnostic.

Patient summary

Important decisions, including treatment protocols, clinical trial eligibility, and life planning, rest on our ability to predict cancer outcomes accurately. Here, we introduce a contemporary renal cell carcinoma prognostic model leveraging high-quality data from a clinical trial. The current model predicts three outcome measures commonly utilized in clinical practice and exceeds the predictive ability of available prognostic models.

Purpose

To validate currently used recurrence prediction models for renal cell carcinoma (RCC) by using prospective data from the ASSURE (ECOG-ACRIN E2805; Adjuvant Sorafenib or Sunitinib for Unfavorable Renal Carcinoma) adjuvant trial.

Patients and methods

Eight RCC recurrence models (University of California at Los Angeles Integrated Staging System [UISS]; Stage, Size, Grade, and Necrosis [SSIGN]; Leibovich; Kattan; Memorial Sloan Kettering Cancer Center [MSKCC]; Yaycioglu; Karakiewicz; and Cindolo) were selected on the basis of their use in clinical practice and clinical trial designs. These models along with the TNM staging system were validated using 1,647 patients with resected localized high-grade or locally advanced disease (≥ pT1b grade 3 and 4/pTanyN1Mo) from the ASSURE cohort. The predictive performance of the model was quantified by assessing its discriminatory and calibration abilities.

Results

Prospective validation of predictive and prognostic models for localized RCC showed a substantial decrease in each of the predictive abilities of the model compared with their original and externally validated discriminatory estimates. Among the models, the SSIGN score performed best (0.688; 95% CI, 0.686 to 0.689), and the UISS model performed worst (0.556; 95% CI, 0.555 to 0.557). Compared with the 2002 TNM staging system (C-index, 0.60), most models only marginally outperformed standard staging. Importantly, all models, including TNM, demonstrated statistically significant variability in their predictive ability over time and were most useful within the first 2 years after diagnosis.

Conclusion

In RCC, as in many other solid malignancies, clinicians rely on retrospective prediction tools to guide patient care and clinical trial selection and largely overestimate their predictive abilities. We used prospective collected adjuvant trial data to validate existing RCC prediction models and demonstrate a sharp decrease in the predictive ability of all models compared with their previous retrospective validations. Accordingly, we recommend prospective validation of any predictive model before implementing it into clinical practice and clinical trial design.

Purpose

We describe what is to our knowledge a novel classification system for local recurrence after surgery of renal cell carcinoma. We assessed its prognostic implications using prospective, randomized controlled data.

Materials and methods

We queried the ASSURE (Sunitinib Malate or Sorafenib Tosylate in Treating Patients With Kidney Cancer That Was Removed By Surgery) (ECOG-ACRIN [Eastern Cooperative Oncology Group-American College of Radiology Imaging Network] E2805) trial data for patients with fully resected, intermediate-high risk, nonmetastatic renal cell carcinoma with local recurrence. We used certain definitions, including type I-single recurrence in a remnant kidney or ipsilateral renal fossa, type II-single recurrence in the ipsilateral vasculature, the ipsilateral adrenal gland or a lymph node, type III-single recurrence in other intra-abdominal soft tissues or organs and type IV-any combination of types I-III or multiple recurrences of a single type. Multivariable logistic regression and the log rank test were performed to identify clinicopathological predictors and compare survival, respectively.

Results

Of the 1,943 patients 300 (15.4%) had local recurrence, which was type I, II, III and IV in 66 (22.0%), 97 (32.3%), 87 (29.0%) and 50 (16.7%), respectively. Surgical modality (minimally invasive vs open) and type of surgery (partial vs radical) did not predict any local recurrence. Five-year cancer specific survival and overall survival were worse in patients with type IV recurrence (each p <0.001). There was no difference in survival among patients with types I to III recurrence.

Conclusions

In patients with intermediate-high risk nonmetastatic renal cell carcinoma local recurrence appears to be a function of biology more than of surgical modality or surgery type. The prognosis for solitary intra-abdominal local recurrences appear similar regardless of location (types I-III). Local recurrences involving multiple sites and/or subdivisions are associated with worse survival (type IV).