Projected Impact of Omidubicel-onlv on Racial/Ethnic Disparities in Allogeneic Hematopoietic Cell Transplantation (Allo-HCT) Outcomes in Hematologic Malignancies.

INTRODUCTION: In a phase III clinical trial (NCT02730299), omidubicel-onlv, a nicotinamide-modified allogeneic hematopoietic progenitor cell therapy, showed rapid hematopoietic and immune recovery compared with standard umbilical cord blood (UCB) transplant across all racial/ethnic groups. METHODS: A decision-tree model was used to project the effect of omidubicel-onlv availability on addressing health disparities in allogeneic hematopoietic cell transplantation (allo-HCT) access and outcomes for patients with hematologic malignancies. The model used a hypothetical population of 10,000 allo-HCT-eligible US adults, for whom matched related donors were not available. Patients received matched or mismatched unrelated donor, haploidentical, UCB transplant, or no transplant. Scenarios with omidubicel-onlv use of 0% (status quo), 10%, 15%, 20%, and 30% were modeled on the basis of proportional reductions in other allo-HCT sources or no transplant by racial/ethnic group. RESULTS: Increased omidubicel-onlv use was associated with a higher proportion of patients undergoing allo-HCT, decreased time to allo-HCT, decreased 1-year non-relapse mortality, and increased 1-year overall survival, particularly among racial minorities. In the scenario modeling 20% omidubicel-onlv use, the proportion of Black patients receiving allo-HCT increased by 129%; increases were also observed in Asian (64%), Hispanic (45%), and other (42%) patient groups. Modeled time to allo-HCT improved among transplanted patients (23%) from 11.4 weeks to 8.8 weeks. One-year OS in the overall population increased by 3%, with improvements ranging from 3% for White patients to 5% for Black patients. CONCLUSION: This study demonstrates that broad access to omidubicel-onlv could increase access to allo-HCT and improve outcomes for patients, with the greatest benefits seen among racial/ethnic minority groups.

Real-world use and clinical impact of an electronic patient-reported outcome tool in patients with solid tumors treated with immuno-oncology therapy.

BACKGROUND: Utilization of electronic patient-reported outcome (ePRO) tools to monitor symptoms in patients undergoing cancer treatment has shown clinical benefits. Tennessee Oncology (TO) implemented an ePRO platform in 2019, allowing patients to report their health status online. We conducted a real-world, multicenter, observational, non-interventional cohort study to evaluate utilization of this platform in adults with solid tumors who initiated immuno-oncology (IO) therapy as monotherapy or in combination at TO clinics. METHODS: Patients initiating IO therapy prior to platform implementation were included in a historical control (HC) cohort; those initiating treatment after implementation were included in the ePRO cohort, which was further divided into ePRO users (platform enrollment ≤ 45 days from IO initiation) and non-users. Data were extracted from electronic medical records; patients were followed for up to 6 months (no minimum follow up). Outcomes included patient characteristics, treatment patterns, duration of therapy (DoT), and overall survival (OS). RESULTS: Data were collected for 538 patients in the HC and 1014 in the ePRO cohort; 319 in the ePRO cohort were ePRO users (uptake rate 31%). Baseline age was higher, more patients had stage IV disease at diagnosis, and more received monotherapy (82 vs 52%, respectively) in the HC vs the ePRO cohort. Median follow-up was 181.0 days (range 0.0-182.6) in the HC and 175.0 (0.0-184.0) in the ePRO cohort. Median DoT of index IO regimen was 5.1 months (95% confidence interval [CI], 4.4-NE) in the HC cohort vs not estimable (NE) in the ePRO cohort. Multivariable regression adjusting for baseline differences confirmed lower risk of treatment discontinuation in the ePRO vs HC cohort: hazard ratio (HR) 0.83 (95% CI, 0.71-0.97); p < 0.05. The estimated 6-month OS rate was 65.5% in the HC vs 72.4% in the ePRO cohort (p < 0 .01). Within the ePRO cohort, DoT of index IO regimen and OS did not differ between users and non-users. In ePRO users, patient platform use was durable over 6 months. CONCLUSION: Improvements in DoT and OS were seen after ePRO platform implementation. Conclusions are limited by challenges in separating the impact of platform implementation from other changes affecting outcomes.

Exploring fathers' experiences of caring for a child with complex care needs through ethnography and arts-based methodologies.

BACKGROUND: Although the number of children living with complex care needs (CCN) is increasing worldwide, there is limited data on the experience of fathers caring for children with CCN. This paper reports on findings specific to fathers' experiences of caring for their child with CCN and highlights recommendations provided for parents of children with CCN, service providers, and policymakers. The findings emerged from a larger study designed to examine how Canadian families of children with CCN participate in society. METHODS: We used the qualitative research approach of ethnography and arts-based methodologies (ecomaps and photovoice) as well as purposive and snowball sampling techniques. Four parents were engaged as advisors and twenty-nine fathers participated in interviews (all were married or in a relationship; age range of 28 to 55 years). In line with an ethnographic approach, data analysis involved several iterative steps including comparing data from the first, second, and third set of interviews and refining themes. RESULTS: One overarching theme, striving to be there for the child with CCN, was identified. Five supporting themes further exemplified how fathers strived to be there for their child: 1) contributing to the parental team through various roles; 2) building accessibility through adaptation; 3) engaging in activities with the child; 4) expressing admiration and pride in their children; and 5) meaning making. Recommendations for parents included making and nurturing connections and asking for help while recommendations for healthcare and social service providers included communicating authentically with families and listening to parents. Fathers also indicated that leadership and funding for programs of families of children with CCN should be priorities for policymakers. CONCLUSIONS: In addition to documenting fathers' active involvement in their child's care and development, our findings provide new insights into how fathers make participation in everyday life accessible and inclusive for their children. Study findings also point to 1) priority areas for policymakers (e.g., accessible physical environments); 2) factors that are critical for fostering collaborative care teams with fathers; and 3) the need for complex care teams in the adult health care system. Implications for those providing psychosocial support for these families are noted as well as knowledge gaps worthy of future exploration such as the role of diversity or intersectionality in fathering children with CCN.

Health care costs among patients with hematologic malignancies receiving allogeneic transplants: a US payer perspective.

ABSTRACT: Patients with hematologic malignancies undergoing allogeneic hematopoietic cell transplant (allo-HCT) require extensive care. Using the Merative MarketScan Commercial Claims and Encounters database (2016 Q1-2020 Q2), we quantified the costs of care and assessed real-world complication rates among commercially insured US patients diagnosed with a hematologic malignancy and aged between 12 and 64 years undergoing inpatient allo-HCT. Health care resource use and costs were assessed from 100 days before HCT to 100 days after HCT. Primary hospitalization was defined as the time from HCT until first discharge date. Incidence of complications was assessed using medical billing codes from HCT date to 100 days after HCT. Among the 1082 patients analyzed, allo-HCT grafts included peripheral blood (79%), bone marrow (11%), and umbilical cord blood (3%). In the 100 days after HCT, 52% of the patients experienced acute graft-versus-host disease; 21% had cytomegalovirus infection. The median primary hospitalization length of stay (LOS) was 28 days; 31% required readmission in first 100 days after HCT. Across the transplant period (14 days pretransplant to 100 days posttransplant), 44% of patients were admitted to the intensive care unit with a median LOS of 29 days. Among those with noncapitated health plans (n = 937), median cost of all-cause health care per patient during the transplant period was $331 827, which was driven by primary hospitalization and readmission. Additionally, the predicted median incremental costs per additional day in an inpatient setting increased with longer LOS (eg, $3381-$4071, 10th-20th day.) Thus, decreasing length of primary hospitalization and avoiding readmissions should significantly reduce the allo-HCT cost of care.

Real-World Outcomes of Trilaciclib Among Patients with Extensive-Stage Small Cell Lung Cancer Receiving Chemotherapy.

INTRODUCTION: Trilaciclib was recently approved in the USA for reducing chemotherapy-induced myelosuppression (CIM) among adults with extensive-stage small cell lung cancer (ES-SCLC) when administered prior to chemotherapy. There is limited understanding of real-world outcomes of trilaciclib. METHODS: A comprehensive literature review was conducted using a keyword search in the MEDLINE, Embase, and conference abstracts. Additional studies were identified through communications with the authors of relevant studies. Published and unpublished real-world studies of trilaciclib- and comparable non-trilaciclib-treated patients with ES-SCLC were included. Evidence on myelosuppressive hematologic adverse events (HAEs), cytopenia-related healthcare utilization, and other reported outcomes (e.g., hospitalizations, dose reduction, and treatment delay) were synthesized. If feasible, outcomes were compared qualitatively between the trilaciclib and historical reference groups, and between first-line trilaciclib initiators and the overall trilaciclib population. Weighted averages were estimated for selected outcomes using sample size as the weight. RESULTS: The literature search identified five unique studies based on eight records-two included trilaciclib only, two non-trilaciclib only, and one both. In trilaciclib cohorts, the weighted average prevalence of grade ≥ 3 myelosuppressive HAEs in ≥ 1 lineage, ≥ 2 lineages, and all three lineages was 40.5%, 14.5%, and 7.5%, respectively. All rates were numerically lower compared to the historical non-trilaciclib cohorts (58.8%, 28.0%, 13.0% respectively). Cytopenia-related healthcare utilization was also lower in the trilaciclib cohorts. In general, first-line trilaciclib initiators had numerically lower myelosuppressive HAEs and cytopenia-related healthcare utilization than the overall trilaciclib patients. CONCLUSIONS: The existing evidence suggests that trilaciclib may reduce single and multilineage grade ≥ 3 myelosuppressive HAEs and cytopenia-related healthcare utilization among patients with ES-SCLC in the real world. It is a promising new treatment for CIM prevention in ES-SCLC and may bring greater benefits to first-line trilaciclib initiators. Future studies are recommended to further evaluate the real-world effectiveness of trilaciclib.

Economic Impact of Recurrent Clostridioides difficile Infection in the USA: A Systematic Literature Review and Cost Synthesis.

INTRODUCTION: Up to 35% of patients with a first episode of Clostridioides difficile infection (CDI) develop recurrent CDI (rCDI), and of those, up to 65% experience multiple recurrences. A systematic literature review (SLR) was conducted to review and summarize the economic impact of rCDI in the United States of America. METHODS: English-language publications reporting real-world healthcare resource utilization (HRU) and/or direct medical costs associated with rCDI in the USA were searched in MEDLINE, MEDLINE In-Process, Embase, and the Cochrane Library databases over the past 10 years (2012-2022), as well as in selected scientific conferences that publish research on rCDI and its economic burden over the past 3 years (2019-2022). HRU and costs identified through the SLR were synthesized to estimate annual rCDI-attributable direct medical costs to inform the economic impact of rCDI from a US third-party payer's perspective. RESULTS: A total of 661 publications were retrieved, and 31 of them met all selection criteria. Substantial variability was found across these publications in terms of data source, patient population, sample size, definition of rCDI, follow-up period, outcomes reported, analytic approach, and methods to adjudicate rCDI-attributable costs. Only one study reported rCDI-attributable costs over 12 months. Synthesizing across the relevant publications using a component-based cost approach, the per-patient per-year rCDI-attributable direct medical cost was estimated to range from $67,837 to $82,268. CONCLUSIONS: While real-world studies on economic impact of rCDI in the USA suggested a high-cost burden, inconsistency in methodologies and results reporting warranted a component-based cost synthesis approach to estimate the annual medical cost burden of rCDI. Utilizing available literature, we estimated the average annual rCDI-attributable medical costs to allow for consistent economic assessments of rCDI and identify the budget impact on US payers.

Clinical Implementation of an Artificial Intelligence Algorithm for Magnetic Resonance-Derived Measurement of Total Kidney Volume.

OBJECTIVE: To evaluate the performance of an internally developed and previously validated artificial intelligence (AI) algorithm for magnetic resonance (MR)-derived total kidney volume (TKV) in autosomal dominant polycystic kidney disease (ADPKD) when implemented in clinical practice. PATIENTS AND METHODS: The study included adult patients with ADPKD seen by a nephrologist at our institution between November 2019 and January 2021 and undergoing an MR imaging examination as part of standard clinical care. Thirty-three nephrologists ordered MR imaging, requesting AI-based TKV calculation for 170 cases in these 161 unique patients. We tracked implementation and performance of the algorithm over 1 year. A radiologist and a radiology technologist reviewed all cases (N=170) for quality and accuracy. Manual editing of algorithm output occurred at radiology or radiology technologist discretion. Performance was assessed by comparing AI-based and manually edited segmentations via measures of similarity and dissimilarity to ensure expected performance. We analyzed ADPKD severity class assignment of algorithm-derived vs manually edited TKV to assess impact. RESULTS: Clinical implementation was successful. Artificial intelligence algorithm-based segmentation showed high levels of agreement and was noninferior to interobserver variability and other methods for determining TKV. Of manually edited cases (n=84), the AI-algorithm TKV output showed a small mean volume difference of -3.3%. Agreement for disease class between AI-based and manually edited segmentation was high (five cases differed). CONCLUSION: Performance of an AI algorithm in real-life clinical practice can be preserved if there is careful development and validation and if the implementation environment closely matches the development conditions.

Physician perceptions of drug utilization management: Results of a national survey.

The use of drug utilization management techniques such as formulary exclusions, prior authorizations, and step edits has risen sharply during the last decade, contributing to the growing burden on physicians and patients. Limited quantitative data exist, however, on physician perceptions of drug utilization management. A national survey was conducted between February 9 and March 30, 2021, targeting office-based physicians working in the United States to assess their perceptions on drug utilization management in their practice. Of the 742 physicians that participated in the study, over 80% reported deciding against prescribing certain treatments in anticipation of drug utilization management at least sometimes (>50% of the time). Despite utilization management having an impact on prescribing decisions, about half of physicians said that the utilization management policies they encounter rarely or never (0-25% of the time) align with clinical evidence.

The Physician and Administrator-Reported Cost of Drug Utilization Management to Physician Practices: A Cross-Sectional Survey.

BACKGROUND: The use of drug utilization management techniques such as formulary exclusions, prior authorizations, and step edits has risen sharply during the last decade, contributing to growing administrative costs for physician practices. However, limited data exist on the extent of these administrative costs, with previous studies relying on data from over a decade ago. OBJECTIVE: The aim of this study was to assess physician and practice administrator experiences with drug utilization management. METHODS: A national survey was conducted between 9 February and 30 March 2021, targeting 925 physicians and administrators working at medical practices in the US. Time spent by physicians and their staff on tasks related to drug utilization management for prescription medications was collected and used to calculate the dollar value of that time. RESULTS: We estimated that physicians spent a median of 4.0 h per week on drug utilization management, while nurses spent 15.0 h and other staff spent between 3.6 and 10.0 h on drug utilization management per physician per week. This time was associated with a calculated median dollar value of $75,927 per physician per year. Extrapolating this estimate to a national scale suggests that time spent annually by physician practices on drug utilization management could be valued at more than $43 billion. CONCLUSIONS: Drug utilization management results in significant time spent by US physician practices, which in turn, results in meaningful costs to these practices. As the prevalence of drug utilization management continues to grow, the impact on physician practices will remain an important topic.

Impact of clinical pathways on treatment patterns and outcomes for patients with non-small-cell lung cancer: real-world evidence from a community oncology practice.

Introduction: The evolving treatment landscape for non-small-cell lung cancer (NSCLC) and complexities of regulations and reimbursement present challenges to community oncologists. Clinical pathways are tools to optimize care, but information on their value in the real world is limited. This retrospective study assessed treatment patterns and clinical outcomes in patients with stage I-III NSCLC pre- and post-pathways implementation at Tennessee Oncology, a large, community-based oncology practice in the USA. Methods & Materials: Chart data were abstracted for adults diagnosed with stage I-III NSCLC who received systemic treatment. Patients were divided into pre-pathways (treatment initiation 2014-2015) and post-pathways (treatment initiation 2016-2018) cohorts. Patient characteristics, treatment patterns and outcomes were summarized descriptively. Kaplan-Meier curves were used to assess time-dependent outcomes, and log-rank test was used to compare the cohorts. Results: 291 patients were included (stage I-II: 38 pre-pathways, 55 post-pathways; stage III: 105 pre-pathways, 93 post-pathways). Duration on first-line (1L) therapy was similar for stage I-II patients pre- and post-pathways (median 1.9 months vs 2.1 months; p = 0.75), but increased for stage III patients post-pathways (2.1 months vs 1.4 months pre-pathways; p < 0.01). Achievement of a complete or partial response with 1L therapy was similar post-pathways among stage I-stage -IIII patients (60.0% vs 55.2% pre-pathways), but increased for stage III patients (56.0% vs 35.2% pre-pathways). Conclusion: Given that improvements in rates of treatment response post-pathways occurred only for patients diagnosed with stage III NSCLC, among whom immunotherapy uptake increased post-pathways, such improvements may be attributable to evolving practices in cancer care, including advances in treatment and care delivery, rather than clinical pathways implementation. Further research is warranted to assess the impact of clinical pathways in the current treatment era, given that immunotherapy has now become the standard of care in NSCLC.

Automated measurement of total kidney volume from 3D ultrasound images of patients affected by polycystic kidney disease and comparison to MR measurements.

PURPOSE: Total kidney volume (TKV) is the most important imaging biomarker for quantifying the severity of autosomal-dominant polycystic kidney disease (ADPKD). 3D ultrasound (US) can accurately measure kidney volume compared to 2D US; however, manual segmentation is tedious and requires expert annotators. We investigated a deep learning-based approach for automated segmentation of TKV from 3D US in ADPKD patients. METHOD: We used axially acquired 3D US-kidney images in 22 ADPKD patients where each patient and each kidney were scanned three times, resulting in 132 scans that were manually segmented. We trained a convolutional neural network to segment the whole kidney and measure TKV. All patients were subsequently imaged with MRI for measurement comparison. RESULTS: Our method automatically segmented polycystic kidneys in 3D US images obtaining an average Dice coefficient of 0.80 on the test dataset. The kidney volume measurement compared with linear regression coefficient and bias from human tracing were R2 = 0.81, and - 4.42%, and between AI and reference standard were R2 = 0.93, and - 4.12%, respectively. MRI and US measured kidney volumes had R2 = 0.84 and a bias of 7.47%. CONCLUSION: This is the first study applying deep learning to 3D US in ADPKD. Our method shows promising performance for auto-segmentation of kidneys using 3D US to measure TKV, close to human tracing and MRI measurement. This imaging and analysis method may be useful in a number of settings, including pediatric imaging, clinical studies, and longitudinal tracking of patient disease progression.

Automated Segmentation of Kidney Cortex and Medulla in CT Images: A Multisite Evaluation Study.

BACKGROUND: In kidney transplantation, a contrast CT scan is obtained in the donor candidate to detect subclinical pathology in the kidney. Recent work from the Aging Kidney Anatomy study has characterized kidney, cortex, and medulla volumes using a manual image-processing tool. However, this technique is time consuming and impractical for clinical care, and thus, these measurements are not obtained during donor evaluations. This study proposes a fully automated segmentation approach for measuring kidney, cortex, and medulla volumes. METHODS: A total of 1930 contrast-enhanced CT exams with reference standard manual segmentations from one institution were used to develop the algorithm. A convolutional neural network model was trained (n=1238) and validated (n=306), and then evaluated in a hold-out test set of reference standard segmentations (n=386). After the initial evaluation, the algorithm was further tested on datasets originating from two external sites (n=1226). RESULTS: The automated model was found to perform on par with manual segmentation, with errors similar to interobserver variability with manual segmentation. Compared with the reference standard, the automated approach achieved a Dice similarity metric of 0.94 (right cortex), 0.90 (right medulla), 0.94 (left cortex), and 0.90 (left medulla) in the test set. Similar performance was observed when the algorithm was applied on the two external datasets. CONCLUSIONS: A fully automated approach for measuring cortex and medullary volumes in CT images of the kidneys has been established. This method may prove useful for a wide range of clinical applications.

Prescribed Water Intake in Autosomal Dominant Polycystic Kidney Disease.

BACKGROUND: Arginine vasopressin promotes kidney cyst growth in autosomal dominant polycystic kidney disease (ADPKD). Increased water intake reduces arginine vasopressin and urine osmolality and may slow kidney cyst growth. METHODS: In this randomized controlled 3-year clinical trial, we randomly assigned adults with ADPKD who had a height-corrected total kidney volume in Mayo imaging subclass categories 1B to 1E and an estimated glomerular filtration rate of 30 ml/min/1.73 m2 or greater to (1) water intake prescribed to reduce 24-hour urine osmolality to 270 mOsmol/kg or less or (2) ad libitum water intake irrespective of 24-hour urine osmolality. The primary end point was the percentage annualized rate of change in height-corrected total kidney volume. RESULTS: A total of 184 patients participated in either the ad libitum water intake group (n=92) or the prescribed water intake group (n=92). Over 3 years, there was no difference in the annualized rate of change in height-corrected total kidney volume between the ad libitum (7.8% per year; 95% confidence interval [CI], 6.6 to 9.0) and prescribed (6.8% per year; 95% CI, 5.8 to 7.7) water intake groups (mean difference, −0.97% per year; 95% CI, −2.37 to 0.44; P=0.18). The difference in mean 24-hour urine osmolality between the ad libitum and prescribed water intake groups was −91 mOsmol/kg (95% CI, −127 to −54 mOsmol/kg), with 52.3% of patients achieving adherence to the target 24-hour urine osmolality and no reduction in serum copeptin over 3 years. The frequency of adverse events was similar between groups. CONCLUSIONS: For patients with ADPKD, prescribed water intake was not associated with excess adverse events and achieved the target 24-hour urine osmolality for half of the patients but did not reduce copeptin or slow the growth of total kidney volume over 3 years compared with ad libitum water intake. (Funded by the National Health and Medical Research Council of Australia [grant GNT1138533], Danone Research, PKD Australia, the University of Sydney, and the Westmead Medical Research Foundation; Australian New Zealand Clinical Trials Registry number, ACTRN12614001216606).

A randomized phase 1b cross-over study of the safety of low-dose pioglitazone for treatment of autosomal dominant polycystic kidney disease.

BACKGROUND: Autosomal dominant polycystic kidney disease (ADPKD) is one of the most common monogenetic disorders in humans and is characterized by numerous fluid-filled cysts that grow slowly, resulting in end-stage renal disease in the majority of patients. Preclinical studies have indicated that treatment with low-dose thiazolidinediones, such as pioglitazone, decrease cyst growth in rodent models of PKD. METHODS: This Phase 1b cross-over study compared the safety of treatment with a low dose (15 mg) of the peroxisome proliferator-activated receptor-γ (PPAR-γ) agonist pioglitazone or placebo in PKD patients, with each treatment given for 1 year. The study monitored known side effects of PPAR-γ agonist treatment, including fluid retention and edema. Liver enzymes and risk of hypoglycemia were assessed throughout the study. As a secondary objective, the efficacy of low-dose pioglitazone was followed using a primary assessment of total kidney volume (TKV), blood pressure (BP) and kidney function. RESULTS: Eighteen patients were randomized and 15 completed both arms. Compared with placebo, allocation to pioglitazone resulted in a significant decrease in total body water as assessed by bioimpedance analysis {mean difference 0.16 Ω [95% confidence interval (CI) 0.24-2.96], P = 0.024} and no differences in episodes of heart failure, clinical edema or change in echocardiography. Allocation to pioglitazone led to no difference in the percent change in TKV of -3.5% (95% CI -8.4-1.4, P = 0.14), diastolic BP and microalbumin:creatinine ratio. CONCLUSIONS: In this small pilot trial in people with ADPKD but without diabetes, pioglitazone 15 mg was found to be as safe as placebo. Larger and longer-term randomized trials powered to assess effects on TKV are needed.

Semantic Instance Segmentation of Kidney Cysts in MR Images: A Fully Automated 3D Approach Developed Through Active Learning.

Total kidney volume (TKV) is the main imaging biomarker used to monitor disease progression and to classify patients affected by autosomal dominant polycystic kidney disease (ADPKD) for clinical trials. However, patients with similar TKVs may have drastically different cystic presentations and phenotypes. In an effort to quantify these cystic differences, we developed the first 3D semantic instance cyst segmentation algorithm for kidneys in MR images. We have reformulated both the object detection/localization task and the instance-based segmentation task into a semantic segmentation task. This allowed us to solve this unique imaging problem efficiently, even for patients with thousands of cysts. To do this, a convolutional neural network (CNN) was trained to learn cyst edges and cyst cores. Images were converted from instance cyst segmentations to semantic edge-core segmentations by applying a 3D erosion morphology operator to up-sampled versions of the images. The reduced cysts were labeled as core; the eroded areas were dilated in 2D and labeled as edge. The network was trained on 30 MR images and validated on 10 MR images using a fourfold cross-validation procedure. The final ensemble model was tested on 20 MR images not seen during the initial training/validation. The results from the test set were compared to segmentations from two readers. The presented model achieved an averaged R2 value of 0.94 for cyst count, 1.00 for total cyst volume, 0.94 for cystic index, and an averaged Dice coefficient of 0.85. These results demonstrate the feasibility of performing cyst segmentations automatically in ADPKD patients.

Characteristics of Patients with End-Stage Kidney Disease in ADPKD.

INTRODUCTION: Cystic expansion damaging the parenchyma is thought to lead to end-stage kidney disease (ESKD) in autosomal dominant polycystic kidney disease (ADPKD). Here we characterized genotypic and phenotypic attributes of ADPKD at time of ESKD. METHODS: This is a retrospective cross-sectional study of patients with ADPKD with ESKD evaluated at Mayo Clinic with available abdominal computed tomography (CT) or magnetic resonance imaging (MRI). Kidney volumes were measured (total kidney volume adjusted for height [HtTKV]), Mayo Image Class (MIC) calculated, ADPKD genotype determined, and clinical and laboratory features obtained from medical records. RESULTS: Differences in HtTKV at ESKD were associated with patient age and sex; older patients and women had smaller HtTKV at ESKD. HtTKV at ESKD was observed to be 12.3% smaller with each decade of age (P < 0.01); but significant only in women (17.8%, P < 0.01; men 6.9%, P = 0.06). Patients with onset of ESKD at <47, 47-61, or >61 years had different characteristics, with a shift from youngest to oldest in male to female enrichment, MIC from 1D/1E to 1B/1C, likely fully penetrant PKD1 mutations from 95% to 42%, and presence of macrovascular disease from 8% to 40%. Macrovascular disease was associated with smaller kidneys in female patients. CONCLUSION: HtTKV at ESKD was smaller with advancing age in patients with ADPKD, particularly in women. These novel findings provide insight into possible underlying mechanisms leading to ESKD, which differ between younger and older individuals. Cystic growth is the predominant mechanism in younger patients with ESKD, whereas aging-related factors, including vascular disease, becomes potentially important as patients age.

Automatic semantic segmentation of kidney cysts in MR images of patients affected by autosomal-dominant polycystic kidney disease.

PURPOSE: For patients affected by autosomal-dominant polycystic kidney disease (ADPKD), successful differentiation of cysts is useful for automatic classification of patient phenotypes, clinical decision-making, and disease progression. The objective was to develop and evaluate a fully automated semantic segmentation method to differentiate and analyze renal cysts in patients with ADPKD. METHODS: An automated deep learning approach using a convolutional neural network was trained, validated, and tested on a set of 60 MR T2-weighted images. A three-fold cross-validation approach was used to train three models on distinct training and validation sets (n = 40). An ensemble model was then built and tested on the hold out cases (n = 20), with each of the cases compared to manual segmentations performed by two readers. Segmentation agreement between readers and the automated method was assessed. RESULTS: The automated approach was found to perform at the level of interobserver variability. The automated approach had a Dice coefficient (mean ± standard deviation) of 0.86 ± 0.10 vs Reader-1 and 0.84 ± 0.11 vs. Reader-2. Interobserver Dice was 0.86 ± 0.08. In terms of total cyst volume (TCV), the automated approach had a percent difference of 3.9 ± 19.1% vs Reader-1 and 8.0 ± 24.1% vs Reader-2, whereas interobserver variability was - 2.0 ± 16.4%. CONCLUSION: This study developed and validated a fully automated approach for performing semantic segmentation of kidney cysts in MR images of patients affected by ADPKD. This approach will be useful for exploring additional imaging biomarkers of ADPKD and automatically classifying phenotypes.

Automated total kidney volume measurements in pre-clinical magnetic resonance imaging for resourcing imaging data, annotations, and source code.

The objective of this study was to validate a fully automated total kidney volume measurement method for pre-clinical rodent trials that is fast, accurate, reproducible, and to provide these resources to the research community. Rodent studies that involve imaging are crucial for monitoring treatment efficacy in diseases such as polycystic kidney disease. Previous studies utilize manual or semi-automated segmentations, which are time consuming and potentially biased. To develop our automated system, a total of 150 axial magnetic resonance images (MRI) from a variety of mouse models were manually segmented and used to train/validate an automated algorithm. To test the longitudinal application of the model, four mutant and four wild-type mice were imaged sequentially over three to twelve weeks via MRI. Segmentations of the kidneys (excluding the renal pelvis) were generated by the automated method and two different readers, with one reader repeating the measurements. Similarity metrics and longitudinal analysis were calculated to assess the performance of the automated compared to the manual methods. The automated approach required no user input, besides a final visual quality control step. Similarity metrics of the automated method versus the manual segmentations were on par with inter- and intra-reader comparisons. Thus, our fully automated approach described here can be safely used in longitudinal, pre-clinical trials that involve the segmentation of rodent kidneys in T2-weighted MRIs.

Development of a volumetric pancreas segmentation CT dataset for AI applications through trained technologists: a study during the COVID 19 containment phase.

PURPOSE: To evaluate the performance of trained technologists vis-à-vis radiologists for volumetric pancreas segmentation and to assess the impact of supplementary training on their performance. METHODS: In this IRB-approved study, 22 technologists were trained in pancreas segmentation on portal venous phase CT through radiologist-led interactive videoconferencing sessions based on an image-rich curriculum. Technologists segmented pancreas in 188 CTs using freehand tools on custom image-viewing software. Subsequent supplementary training included multimedia videos focused on common errors, which were followed by second batch of 159 segmentations. Two radiologists reviewed all cases and corrected inaccurate segmentations. Technologists' segmentations were compared against radiologists' segmentations using Dice-Sorenson coefficient (DSC), Jaccard coefficient (JC), and Bland-Altman analysis. RESULTS: Corrections were made in 71 (38%) cases from first batch [26 (37%) oversegmentations and 45 (63%) undersegmentations] and in 77 (48%) cases from second batch [12 (16%) oversegmentations and 65 (84%) undersegmentations]. DSC, JC, false positive (FP), and false negative (FN) [mean (SD)] in first versus second batches were 0.63 (0.15) versus 0.63 (0.16), 0.48 (0.15) versus 0.48 (0.15), 0.29 (0.21) versus 0.21 (0.10), and 0.36 (0.20) versus 0.43 (0.19), respectively. Differences were not significant (p > 0.05). However, range of mean pancreatic volume difference reduced in the second batch [- 2.74 cc (min - 92.96 cc, max 87.47 cc) versus - 23.57 cc (min - 77.32, max 30.19)]. CONCLUSION: Trained technologists could perform volumetric pancreas segmentation with reasonable accuracy despite its complexity. Supplementary training further reduced range of volume difference in segmentations. Investment into training technologists could augment and accelerate development of body imaging datasets for AI applications.