Stanford Medicine student devises liver exchange, easing shortage of organs

A rare three-way exchange of liver transplants in Pakistan was made possible with a new algorithm developed by a Stanford Medicine student.

- By Nina Bai

In some countries, cultural norms limit the availability of organs from deceased donors. An algorithm devised by Stanford Medicine's Alex Chan can increase the number of transplants through a liver exchange.
Marko Aliaksandr/Shutterstock.com

On March 17 of this year, six operating rooms at the Pakistan Kidney and Liver Institute were prepped for six simultaneous surgeries. On three operating tables were patients with end-stage liver disease. On the other three were their relatives — a daughter, a son and a wife — who had agreed to donate a portion of their livers.

The donations, however, weren’t staying in the family. Due to blood type or size, each of the donors’ livers was incompatible with the relative in need of a transplant. A sophisticated algorithm had matched the three pairs of relatives, orchestrating a three-way liver exchange that allowed each patient to receive a compatible, life-saving liver from a stranger.

The transaction was one of the world’s first documented three-way liver exchanges and the first to use a new liver exchange algorithm that finds optimal matches from a pool of candidates. The algorithm, developed by Alex Chan, a graduate student economist working on health policy at Stanford Medicine, has the potential to increase the number of liver transplants, especially in countries like Pakistan, where living donors are the norm.

“Liver exchanges really benefit transplant systems that depend on living donor transplants,” Chan said. “In places like Pakistan, India or South Korea, more than 90% of organs come from living donors because of cultural preferences to keep the body intact after death.”

In the United States, where deceased donors are more common, only 6% of liver transplants in 2021 came from living donors.

To discourage the sale of organs, Pakistan further restricts transplants by requiring that transplant candidates co-register with a donor who is a close relative. As a result, some 30% to 50% of liver transplant candidates in Pakistan are unable to find a compatible donor, and about 10,000 people there die each year waiting for an organ.

Liver swap

At the time of the surgeries in March, the Pakistan Kidney and Liver Institute had 20 liver transplant candidates who had co-registered with a related, but incompatible donor. Chan’s algorithm was able to identify the three-way swap and two more two-way swaps, which allowed seven of the candidates to receive compatible livers. The procedures are described in a paper published Dec. 7 in JAMA Surgery.

“We’ve shown that you don’t need to have a huge number of candidates. Even with a sample of 20, this algorithm can find extra matches,” said Saad Salman, MD, a medical resident at Harvard and lead author of the paper. Chan is the author of a separate working paper detailing the algorithm.

The concept of transplant exchanges is not new. In fact, an algorithm for kidney exchanges developed by Nobel-prize winning economist Alvin Roth, PhD, the Craig and Susan McCaw Professor at Stanford, and collaborators has been the standard for living kidney transplantations in the United States.

Alex Chan

Roth, known for his work in game theory and market design, and for applying these concepts to real-world problems, has been a longtime mentor to Chan. “The liver exchange idea actually came out of a term paper in a first-year market design class at Stanford,” Chan said.

As he learned more about liver transplants, Chan realized there were important biological and ethical differences from kidney transplants. “It turns out that it’s a very different problem,” he said. “Using the kidney algorithm to match liver transplants would be like driving a monorail train on two tracks.”

For both organs, the algorithm needs to consider blood type compatibility, but livers present additional matching requirements, as the size of the organ and the severity of the patient’s condition must be considered.

Although the liver has the remarkable ability to regenerate itself, larger people require donations of larger liver lobes. And while patients waiting for a kidney transplant can stay alive on dialysis, sometimes for years, patients with end-stage liver disease have no similar treatment and are likely to die within months without a transplant.

Patients waiting for a liver are ranked by their model for end-stage liver disease, or MELD, score, which estimates their short-term risk of death. The sickest patients, with the highest mortality risk, are at the top of the waiting list.

“Instead of just finding compatible swaps, we want to find swaps that prioritize the most urgent patients first in order to prevent the most deaths,” Chan said.

He designed his algorithm to incorporate liver size, medical urgency and risk to the donor in determining optimal swaps.

“My being an economist in the medical school, with access to clinicians, was a huge advantage in helping formulate the matching problem correctly,” Chan said. “The algorithm follows the institutional and ethical constraints for liver transplantation, so it is practically relevant.”

Fulfilling a need in Pakistan

The next challenge was to put the algorithm into practice. Chan and Salman reached out to contacts in several Asian countries and eventually connected with Faisal Saud Dar, MBBS, a renowned transplant surgeon at the Pakistan Kidney and Liver Institute and co-author of the paper.

To convince the institute’s liver transplant team, they ran the algorithm on prior patients at the institute and found exchanges that would have saved many lives.

“The surgery team was extremely important, not just in doing the surgeries, but in helping us revise the algorithm by telling us what kind of clinical workup they use before they do these surgeries,” Salman said.

With the transplant team on board, approval from local authorities, and consent from the patients and donors, the first algorithm-matched liver transplants went ahead. A month afterward, all patients were doing well.  

Chan hopes that the success of these transplants will encourage more hospitals to adopt the algorithm. “Liver exchange is not a weird, scary thing. It has been done and it has helped people,” he said. “If more people are willing to do this, we can increase access to liver transplantations in general.”

About Stanford Medicine

Stanford Medicine is an integrated academic health system comprising the Stanford School of Medicine and adult and pediatric health care delivery systems. Together, they harness the full potential of biomedicine through collaborative research, education and clinical care for patients. For more information, please visit med.stanford.edu.

2023 ISSUE 3

Exploring ways AI is applied to health care