2.1

Study population

MUSIC-KIDNEY is a subset of the MUSIC collaborative, with over 100 providers from 20 practices contributing cases [ ]. Data abstractors reviewed the primary medical record and entered clinical, radiographic, treatment, and laboratory parameters at fix intervals into a web-based registry. We retrospectively reviewed all patients 18 years and older who underwent an intervention (radical nephrectomy (RN), partial nephrectomy (PN), tumor ablation (TA), or stereotactic body radiation therapy (SBRT)) for a cT1RM in MUSIC from 05/2017 to 09/2023. We defined NSI as any of the following: PN, TA, or SBRT. As the primary study objective was to assess the association of treatment delay while on AS with the use of NSI, patients were either classified as undergoing immediate intervention, defined as intervention within 90 days of the index encounter with their MUSIC urologist, or delayed intervention, defined as intervention after at least a 90-day period of active surveillance before transitioning to treatment. Patients that were listed as undergoing immediate intervention in the registry, but had surgery greater than 90 days after the index encounter were excluded from this analysis as they did not accurately reflect the clinical question of how active surveillance may influence future treatment decisions. AS for T1RM in the MUSIC KIDNEY registry was defined as the affirmative selection of active surveillance and the absence of treatment for at least 90 days.

Patients with radiographically benign lesions (Bosniak I, II, IIF cysts and angiomyolipomas) were excluded, such that only patients suspicious or indeterminate for RCC were included. Patients were also excluded if they had previous systemic therapy, nephroureterectomy, metastasis at the time of diagnosis, treatment of bilateral kidney masses, familial kidney cancer syndrome, kidney metastasis from another primary tumor, and/or solitary kidney, as these may influence the decision for RN vs NSI.

2.2

Study objectives

The primary objective was to test for the association of treatment type (RN vs. NSI) with timing of treatment (immediate vs. delayed intervention). The primary independent variable was timing of treatment and the main dependent variable was treatment type: RN vs. NSI. Secondary objectives were to assess for the association of treatment timing with measures of perioperative morbidity, oncological, and renal function outcomes. Perioperative and oncological outcomes were limited to patients undergoing surgery (open/robotic/laparoscopic RN/PN). Perioperative outcomes of interest included prolonged length of stay (LOS: defined as ≥3 days for laparoscopic/robotic surgery or ≥5 days for open surgery), estimated blood loss (EBL) >500 ml, warm ischemia time (WIT) >30 minutes, emergency room visit, and readmission. Oncologic outcomes of interest included surgical pathology (benign vs. malignant) and presence of positive surgical margin (PSM). Survival outcomes included metastasis-free survival, overall survival, and CKD stage G3b-free survival (eGFR <45 ml/min/1.73 m ² ).

2.3

Statistical analysis

Patient clinical and demographic characteristics were summarized and compared between patients undergoing immediate vs. delayed intervention with the Chi-square test or Fisher’s exact test for categorical variables, and Wilcoxon rank sum test for continuous variables. For the primary objective, we fit a mixed-effects multivariable logistic regression model to assess for the adjusted association of immediate vs. delayed intervention with the receipt of NSI. A random intercept for surgeon nested within practice was included to account for intra-surgeon correlation. This model was then used to estimate the adjusted proportion of NSI among each group. The model was adjusted for age (continuous), body mass index (BMI, continuous per 5 unit change), sex (male/female), race (white /African American/other/unknown), family history of renal cancer (yes/no), Charlson Comorbidity Index (CCI, 0/1/≥2), preoperative glomerular filtration rate (GFR), insurance type (private/public/none or unknown), practice type (academic/hybrid/private), multiple lesions, radiographic tumor type (complex cyst/indeterminate/solid mass), tumor size (cm, per 0.5 cm change) and RENAL nephrometry score (low [4–6]/intermediate [7–9]/high complexity [10–12]). To test if the association of treatment timing (immediate vs. delayed intervention) and treatment type (RN vs. NSI) was different among patients with cT1a vs. cT1b renal masses, we fit a multivariable model with the same covariates described previously and included an interaction term between clinical stage and treatment timing.

For the secondary objectives, we compared the proportion of patients with the outcomes of interest stated above based on treatment timing and compared between groups with Chi-squared and Fisher’s exact tests. For all time to event analysis, the date of intervention was considered time zero. For overall survival, patients were censored at the date of their last follow-up, and for CKD stage G3b-free survival, patients were censored at their last GFR measurement. The Kaplan Meier estimator and log rank test was used to compare the time to event outcomes of overall survival (OS) and CKD stage G3b-free survival between immediate and delayed intervention cohorts. For time to the development of CKD stage G3b (eGFR <45), we included only patients with a baseline eGFR >45 prior to the date of intervention. Statistical analysis was performed with SAS v. 9.4 (SAS Institute Inc., Cary, NC, USA) with significance set at P <0.05. This study was deemed exempt from review by the Wayne State University IRB.