Characterization of renal masses with MRI-based radiomics: assessment of inter-package and inter-observer reproducibility in a prospective pilot study

OBJECTIVES

To evaluate radiomics features' reproducibility using inter-package/inter-observer measurement analysis in renal masses (RMs) based on MRI and to employ machine learning (ML) models for RM characterization.

METHODS

32 Patients (23M/9F; age 61.8 ± 10.6 years) with RMs (25 renal cell carcinomas (RCC)/7 benign masses; mean size, 3.43 ± 1.73 cm) undergoing resection were prospectively recruited. All patients underwent 1.5 T MRI with T2-weighted (T2-WI), diffusion-weighted (DWI)/apparent diffusion coefficient (ADC), and pre-/post-contrast-enhanced T1-weighted imaging (T1-WI). RMs were manually segmented using volume of interest (VOI) on T2-WI, DWI/ADC, and T1-WI pre-/post-contrast imaging (1-min, 3-min post-injection) by two independent observers using two radiomics software packages for inter-package and inter-observer assessments of shape/histogram/texture features common to both packages (104 features; n = 26 patients). Intra-class correlation coefficients (ICCs) were calculated to assess inter-observer and inter-package reproducibility of radiomics measurements [good (ICC ≥ 0.8)/moderate (ICC = 0.5-0.8)/poor (ICC < 0.5)]. ML models were employed using reproducible features (between observers and packages, ICC > 0.8) to distinguish RCC from benign RM.

RESULTS

Inter-package comparisons demonstrated that radiomics features from T1-WI-post-contrast had the highest proportion of good/moderate ICCs (54.8-58.6% for T1-WI-1 min), while most features extracted from T2-WI, T1-WI-pre-contrast, and ADC exhibited poor ICCs. Inter-observer comparisons found that radiomics measurements from T1-WI pre/post-contrast and T2-WI had the greatest proportion of features with good/moderate ICCs (95.3-99.1% T1-WI-post-contrast 1-min), while ADC measurements yielded mostly poor ICCs. ML models generated an AUC of 0.71 [95% confidence interval = 0.67-0.75] for diagnosis of RCC vs. benign RM.

CONCLUSION

Radiomics features extracted from T1-WI-post-contrast demonstrated greater inter-package and inter-observer reproducibility compared to ADC, with fair accuracy for distinguishing RCC from benign RM.

CLINICAL RELEVANCE

Knowledge of reproducibility of MRI radiomics features obtained on renal masses will aid in future study design and may enhance the diagnostic utility of radiomics models for renal mass characterization.

Recent innovations in renal replacement technology and potential applications to transplantation and dialysis patients: a review of current methods

The current standard of care for patients with end-stage renal disease (ERSD) is a kidney transplant or dialysis when a donor organ isnot available. The growing gap between patients who require a kidney transplant and the availability of donor organs as well as thenegative effects of long-term dialysis, such as infection, limited mobility, and risk of cancer development, drive the impetus to developalternative renal replacement technology. The goal of this review is to assess the potential of two of the most recent innovations inkidney transplant technology-the implantable bioartificial kidney (BAK) and kidney regeneration technology-in addressing the aforementionedproblems related to kidney replacement for patients with ERSD. Both innovations are fully implantable, autologous, personalizedwith patient cells, and can replace all aspects of kidney function. Not only do these new innovations have the potential toimprove the possibility of transplantation for more patients, they also have potential to improve the outcome of transplantation or dialysis-related renal cancer diagnosis. A major limitation of the current technology is that both implantable BAK and kidney regenerationtechnology are still in preclinical stages, and thus their potential effects cannot be comprehensively generalized to human patients.

Conversion to Open Radical or Partial Nephrectomy Associated with Unplanned Hospital Readmission After Attempted Minimally Invasive Approach

Introduction/Objectives: We endeavored to explore the rates of unplanned hospital readmission (UHR) in patients who underwent minimally invasive radical or partial nephrectomy who were subsequently converted to open surgery. Patients and Methods: The National Cancer Database (NCDB) was used for this study. Patients diagnosed with renal cancer and who had minimally invasive partial or radical nephrectomy from 2004 to 2016 were included in the study. Patients were categorized as converted or not converted to open surgery. UHR was the outcome of the study and was defined as UHR within 30 days of discharge. We conducted a one-to-one nearest-neighbor propensity-score matching using baseline clinical, tumor, and facility characteristics. To evaluate the relationship between conversion to open surgery and UHR, we conducted a multivariable logistic regression on the propensity-matched cohort, a propensity score-matched model without controlling for any covariate, and a propensity score-adjusted model, controlling for only the propensity score. Results: A total of 142,040 patients were identified, with a 2.98% conversion rate. There was an overall decrease in the rate of conversion to open surgery from 2010 (4.11%) to 2016 (2.43%). Laparoscopic radical nephrectomy remained a significant contributor to the rate of conversion (at least 50% per year). In the unmatched cohort, the UHR rate was higher among those who had a conversion to open surgery (n = 710, 0.52% versus n = 44, 1.04%; P < .001). Similarly, in the propensity score-matched cohort, the UHR rate was higher among those who had a conversion to open (n = 22, 0.52% versus n = 44, 1.04%; P = .007). After controlling for other factors, conversion to open remained independently associated with UHF. Conclusion: Conversion to open radical or partial nephrectomy from a minimally invasive approach is independently associated with an increased risk of 30-day UHR.

Should a Drain Be Routinely Required After Transperitoneal Robotic Partial Nephrectomy?

Introduction: Closed drains have traditionally been placed after partial nephrectomy because of risks of bleeding and urine leak. We sought to study the safety of a nonroutine drain (NRD) approach after transperitoneal robotic partial nephrectomy (RPN). Patients and Methods: From a multi-institutional database, we have analyzed the data of 904 patients who underwent RPN. Five hundred forty-six (60.40%) patients underwent RPN by a surgeon who routinely placed drains. Three hundred fifty-eight (39.60%) patients underwent RPN by a surgeon who did not routinely placed drains. Perioperative outcomes, length of stay (LOS), and readmission rates were compared between the two groups. Baseline characteristics, perioperative, and postoperative outcomes were compared using Mann-Whitney U test, chi-square test, and Fisher's exact test. Results: Patients in the NRD group were more likely to have higher body mass index (30.10 kg/m² vs 28.07 kg/m²; P < 0.001), higher tumor size (3.0 cm vs 2.5 cm; P = 0.001), and higher renal score (8 vs 7; P < 0.001). Rate of transfusion (0.00% NRD vs 0.56% RD; P = 0.157) and overall complication (7.33% NRD vs 7.82% RD; P = 0.782) were comparable. Median hospital stay is 1 day for both groups. Readmission rate was also similar (0.55% NRD vs 1.40% RD; P = 0.279). In a multivariable analysis, NRD approach was associated with shorter length of hospital stay (incidence rate ratio [IRR] - 0.72, P < 0.001). Conclusion: An NRD approach for RPN yielded a decreased LOS and similar perioperative outcomes. Placement of surgical drains should be based on individual circumstances, and not required on a routine basis.