2.1

Patient population

After IRB approval (Protocol Rene-2007), within a prospectively maintained database, consecutive patients with indeterminate renal masses who underwent 18-gauge RTB between 2015 and 2022 at a single tertiary referral center were identified. The database comprises patients 1) who underwent RTB before clinical decision or 2) RTB in the context of an ablative treatment or 3) before systemic therapy in metastatic patients.

Patients were assessed via a detailed medical history including data on health-significant comorbidities as scored using the Charlson comorbidity index (CCI) [ ]. Body mass index (BMI; kg/m ² ) was measured. For each patient, we considered RTB approach (US vs. CT), number of biopsy cores (NoC) and length of cores (LoC). Malignancy, histological type, and grading were recorded. All specimens were evaluated by dedicated uro-pathologists with extensive experience in renal cancer and categorized according to ISUP [ ].

Subsequently, patients were annually followed to track the type of subsequent treatment (surgery vs. AS vs. ABL vs. systemic therapy) and to assess survival. For patients surgically treated, RTB data were compared with pathological histology and grading. Postprocedural complications were identified using the Clavien-Dindo classification for both RTB and renal surgeries [ , ].

2.2

Variable definition

The outcomes of the study were:

• 1.
  

  A diagnostic biopsy with defined as a RTB allowing a conclusive pathology report with specific information of RCC histology and grade defined according to ISUP. [ ]

  
  
• 2.
  

  Defining the degree of concordance between RTB histology and grade, and final pathology report after surgery.

The variables of interest were modifiable features of RTB, under the assumption that unmodifiable factors cannot have a direct impact on clinical decision-making per se. Covariates were age, sex, CCI, BMI, lesion side and clinical size.

2.3

Statistical analysis

We have described study methodology in Fig. 1 . Statistical descriptive analyses were used to characterize the patient demographics and clinical features of those undergoing renal biopsies, as well as short-term and long-term outcomes. Additionally, a Sankey diagram was utilized to visually represent the subsequent clinical decisions after RTB.

Study workflow.

RTBs that reached a thorough level of identification, covering both histology and grading, were considered diagnostic.

To predict diagnostic RTB, the Boruta algorithm was employed to ascertain the most significant features in predicting diagnostic RTB. The Boruta algorithm is an all-relevant feature selection method that identifies important variables by creating shadow features and comparing their importance to the original features, ensuring that only those that have a significant contribution to the predictive model are retained through a comparison of the Z-values of each feature with those of “shadow features.” This involved duplicating all actual features and shuffling them in sequence. In each iteration, the Z-value of each attribute was derived from a random forest model, while the Z-value of the shadow was created by randomly shuffling the actual features. A real feature was deemed “important” if its Z-value exceeded the maximum Z-value of shadow features in multiple independent trials [ , ]. Type of imaging used, NoC and LoC were used in the Boruta algorithm model. Subsequently, the K-Nearest Neighbors (KNN), a nonparametric supervised machine learning model, predicted the probability of obtaining a diagnostic RTB based on the chosen variables [ , ]. The KNN algorithm is a machine learning technique that classifies or predicts the category of a data point by considering the majority class among its k nearest neighbours in a feature space [ ]. The dataset was split into training and testing sets using a 90–10 ratio for the KNN model. Specifically, 90% of the data was allocated to the training set, which was utilized for building the predictive model. The remaining 10% constituted the testing set, serving as independent data for evaluating the model’s performance and assessing its ability to generalize to new, unseen observations. The KNN model was employed using a 5-fold cross-validation repeated 3 times, with the optimal K value being determined automatically during the modelling process. Decision boundaries from the KNN model guided the creation of diagnostic recommendations [ ], which were tested using multivariable logistic regression predicting the probability of diagnostic RTB, both in terms of histology and grade. The association between RTB’s complications and NoC, Loc, or the fact that the biopsies followed the final recommendations from this paper was investigated using a multivariate logistic regression.

3.1

Population characteristics and RTB outcomes

Table 1 depicts patient and RTB characteristics. Of the 197 patients included in the study, 17% (n=33) proceeded to surgery. Overall, 89.8% (n=177) and 44.7% (n=88) of biopsies were diagnostic with respect to histotype and grading, respectively. In respect to specific histology results, ccRCC, papillary, and chromophobe were found in 42.1% (n=83), 15.7 (n=31), and 3.6% (n=7) patients and 35% (n=69) had a benign SRM. The median (IQR) NoC and LoC were 2 (1–3) and 1.2 (0.6–1.6) cm, respectively. Only 1% (n=2) of patients reported grade III complications and no grade IV-V complications were reported. The recorded grade III complications encompassed a pleural effusion, managed through thoracic drainage, and bleeding from a costal vessel, addressed through embolization. Table 1B

Table 1A

Population Characteristic (N=197)

Variable	levels	Value
Patients Demographic
Age at biopsy (Years) Median [IQR]		68 [57, 76]
Sex (N., %)	F	60 (30.5)
	M	137 (69.5)
CCI (%)	0	20 (10.2)
	1	26 (13.2)
	2	35 (17.8)
	3+	116 (58.8)
BMI (Kg/m 2 ) Median [IQR]		25.6 [23.3, 29.0]
Lesions and RTBs Characteristics
Lesion Side (%)	L	100 (50.8)
	R	97 (49.2)
Lesion size (Cm) Median [IQR]		2.5 [2.0, 3.2]
N. of biopsy cores (%)	1	67 (34.0)
	2	57 (28.9)
	3	36 (18.3)
	4+	37 (18.8)
Total Biopsy Length (Cm) Median [IQR]		1.2 [0.6, 1.6]
Type of Biopsy (%)	CT – guided	82 (41.6)
	US – guided	115 (58.4)

Only gold members can continue reading. Log In or Register to continue

Related posts:

Tumor involvement of the trigone and urethra at the time of robot-assisted radical cystectomy is associated with adverse oncological outcomes Diet and the microbiome as mediators of prostate cancer risk, progression, and therapy response 5α-reductase inhibitors with or without alpha-blockers and risk of incident upper tract urothelial carcinoma in men with benign prostatic hyperplasia: Analysis of US insurance claims data Quantification of micropapillary component on transurethral resection is associated with likelihood of occult lymph node metastasis at radical cystectomy Laparoscopic suture-free partial nephrectomy using argon-beam-coagulator: Surgical technique and outcomes of a single-center, open-label randomized controlled trial Causal relationship between bladder cancer and gut microbiota contributes to the gut-bladder axis: A two-sample Mendelian randomization study

Stay updated, free articles. Join our Telegram channel

Join