3 Ways Single-Cell Sequencing and AI are Transforming Kidney Transplantation
Posted 27th November 2019 by Liv Sewell
Transplantation is now a successful therapy for end-stage renal disease (ESRD). The first successful kidney transplantation happened in Boston in 1954 and the procedure is now routine clinical practice in more than 80 countries worldwide.
Ishita Moghe and Kim Solez outline some of the major strides taken in kidney pathology and transplantation through the use of scRNA-seq and Artificial Intelligence (AI).
Official statistics demonstrate the success of kidney transplantation as a therapy: 97% of kidney transplants are working at the end of a month; 93% are working at the end of a year; and 83% are working at the end of 3 years.
Yet on the flip-side, 7% of patients’ transplants have failed within a year and the 17% of patients’ transplants have failed within 3 years.10 years after therapy only 54% of transplant kidneys are still working. 20% of kidney transplants every year are re-transplants. The main causes of graft failure are rejection, infection, and disease relapse. Some research points to these outcomes being difficult to reduce and improvements in solid-organ transplantation therapies stagnating since the late eighties.
Ishita Moghe (BSc) and Kim Solez (MD) have demonstrated that the combination of single-cell RNA sequencing and AI programs offers huge potential for kidney pathology and improved outcomes for kidney transplant patients. Here are just three of those ways:
Whole-slide images can now be easily annotated and analysed via a variety of algorithms. The computer-aided pathology is enabling better diagnosis because of the ability to examine sub-visual features which can enable precise disease characterization and prognostication for kidney transplants, ultimately decreasing chances of rejection.
Predicting Allograft Survival
Transplantation outcomes are difficult to predict because of the extensive and diverse factors at play in both donor and recipient’s history. Machine Learning (ML) can be applied and find connections and risk factors that scientists might not have been able to identify. For example, ML has been applied to large-scale retrospective data and identified that early acute rejection combined with serum creatinine levels were important indicators for long-term survival.
Identifying novel cell types
Single-cell RNA sequencing has been shown to identify 3 novel renal cell types and demonstrated that specific kidney diseases are linked to specific cell types. With better understanding of cell subtypes, subtle pathological mechanisms, and low-level gene expression will enable more precision and better outcomes in transplantations.
Single-Cell Sequencing & Artificial Intelligence: Bringing Kidney Transplantation to the Digital Age & Beyond
Ishita Moghe and Kim Solez review the potential of scRNA-seq and AI for improved patient outcomes in kidney transplantation.
Although the 6th Digital Pathology and AI Congress: Europe is now sold out, there are still places at the workshop hosted by Akoya Biosciences exploring Proxima, their newest digital pathology platform. Apply for your free pass.
Lara E. Tushla, ‘When a Transplant Fails’, National Kidney Foundation Website, available at https://www.kidney.org/transplantation/transaction/TC/summer09/TCsm09_TransplantFails(accessed on 20/11/2019). U.S. Department of Health & Human Services, Organ Procurement and Transplantation Network, Kaplan-Meier Graft Survival Rates For Transplants Performed: 2008 – 2015, available at https://optn.transplant.hrsa.gov/data/view-data-reports/national-data/#(accessed 20/11/2019)
Tushla, ‘When a Transplant Fails’.
Abbas Rana & Elizabeth Louise Godfrey. “Outcomes in Solid-Organ Transplantation: Success and Stagnation.” Texas Heart Institute journal, vol. 46 (1), pp. 75-76. Available at https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6379008/#i1526-6702-46-1-75-b2(accessed on 20/11/2019)
Leave a Reply