One of the most common cancers in sub-Saharan Africa can be diagnosed more than seven times faster in those countries with use of liquid biopsies, according to new research.

The cancer, Epstein-Barr virus (EBV)-positive Burkitt lymphoma, is most common in children and young adults in malaria-endemic areas and results in more than 3000 deaths a year in Africa. Though it’s an aggressive and fast-moving cancer, it has a high cure rate when the cancer is identified and treated appropriately early enough.

Compared with gold standard immunohistology with tissue biopsies, liquid biopsy reduced the median time to diagnosis from 46.8 days to 6.5 days, found a study published in Nature Medicine in March.

The much faster ability to diagnose EBV-positive Burkitt lymphoma could transform care of this cancer in youth if the methodology can be adopted clinically, though doing so requires additional investment and, for broader adoption in Africa, substantial logistical and training investment.

“Our study demonstrates the feasibility of integrating liquid biopsy into routine clinical workflow in under-resourced African hospitals using an multidisciplinary team approach,” lead author Clara Chamba, MD, head of hematology at Muhimbili University of Health and Allied Sciences, Dar es Salaam, Tanzania, and her colleagues wrote.

The findings also have wider implications for how African countries could, with adequate investment, shape future screening programs for other cancers, Anna Schuh, MD, PhD, senior author of the study, told Medscape Medical News.

Further, although EBV-positive Burkitt lymphoma is extremely rare in the US and other high-income countries, the findings provide proof-of-concept for faster diagnosis with liquid biopsy for Burkitt and other non-Hodgkin lymphomas in those countries, said Schuh. Tissue biopsies in the US currently take around 1-2 weeks for lymphoma. It’s theoretically possible to reduce that time significantly with a liquid biopsy if appropriate panels and models with appropriate characteristics were developed, Schuh said.

Huge Need for Faster Accurate Diagnosis

Currently, to distinguish EBV-positive Burkitt lymphoma from other non-Hodgkin lymphomas, the standard of care requires complex pathology that is not available in the mostly rural areas where the disease is most prevalent. Rural areas are so affected because the disease’s epidemiology closely tracks that of malaria. EBV infects more than 90% of the global population, and malaria’s suppressive effect on the immune system, particularly with repeated infections, allows the virus’s reactivation, which can lead to this cancer, Rosemary Rochford, PhD, a professor emeritus of immunology and microbiology at the University of Colorado Anschutz School of Medicine, Aurora, Colorado, told Medscape Medical News. “The need to improve diagnostic accuracy is huge,” said Rochford, who was not involved in this research.

A shortage of pathologists and reagents limits how much immunohistology can be done even in relatively well-resourced African cities. Most EBV-positive Burkitt lymphoma is therefore diagnosed based on clinical symptoms or with a hematoxylin and eosin (H&E) morphology that cannot fully distinguish it from certain other types of cancers, the authors explained. These barriers to care mean the median time from first onset of symptoms to diagnosis currently 91 days, with a 51% probability of treatment not starting until 90 days after the child presents with their symptoms, they wrote.

With liquid biopsy, however, the researchers diagnosed 93.3% of Burkitt lymphoma cases (14 out of 15) within a week compared with only 40% using only tissue biopsy (6 out of 15), in the patient group they used for validating their methods.

“These tumors can double in size over a frighteningly short interval of time, so the [difference in the] number of days required to make a liquid-biopsy-based diagnosis vs a gold-standard histopathologic has enormous implications because a lot of patients die really quickly,” Edus Warren, MD, PhD, an oncologist and professor of immunology and translational science at Fred Hutch Cancer Center in Seattle, told Medscape Medical News.

“A lot of patients need to be treated very quickly even with some temporizing chemotherapy just to keep them alive long enough for the time that it might take to receive a conclusive, confident diagnosis,” said Warren, who was not involved in this research. “When a diagnosis takes a month or more, that is a life-or-death event.”

Real-World Use After Validation

For the new study, the researchers first developed a limited immunohistology panel for standard tissue biopsy that they could use as a gold-standard comparison for liquid biopsy. That panel is optimized for resource-limited settings and has shown in past work to have 81% concordance with standard-of-care histopathology.

Then they enrolled 313 children and young adults from four hospitals in Uganda and Tanzania who had clinically suspected lymphoma. They collected blood samples from these children for liquid biopsy analysis and attempted to collect tissue biopsy from as many of these individuals as possible.

The researchers were able to sequence liquid biopsy samples from 78.9% of the 313 patients. They collected tissue biopsies and did H&E staining for 89.5% of the 313; the others died before biopsy, were lost to follow-up, declined the procedure, or had inadequate samples. Tissue samples from only 212 (67.7%) of the study participants could be diagnosed with gold-standard pathology. To validate the liquid biopsy diagnosis, the researchers compared the liquid biopsy and the tissue biopsy diagnoses from those 212 patients.

They found 38.2% had Burkitt lymphoma and 61.8% had a different diagnosis, either a benign tumor or another cancer. The liquid biopsy panel detected MYC-IGH translocations in only 48% cases of Burkitt lymphoma. This translocation is present in nearly all cases of Burkitt lymphoma, but the panel likely did not detect it more often because it’s distributed across DNA snippets outside the area targeted by the panel.

Expanding the target area would be costly, so the researchers used surrogate markers for the MYC translocation and developed six diagnostic models to compare to gold-standard tissue biopsy. The most accurate model was a comprehensive one incorporating age, sex, symptoms duration, tumor site, lab results, and liquid biopsy. It had an area under the curve (AUC) of 95%, a sensitivity of 86%, and a specificity of 95%.

The researchers subsequently validated the comprehensive model with liquid biopsy in 62 patients, resulting in an AUC of 97%, a sensitivity of 93%, and a specificity of 90%. They also compared turnaround time for the tissue biopsy and liquid biopsy in this second cohort. Liquid biopsy provided a diagnosis 40 days earlier, though the researchers acknowledged that their turnaround times might vary in real-world practice.

Implementation Challenges Remain Substantial

The speed of diagnosis is only part of the benefit of liquid biopsy, said Schuh, who is a professor of molecular diagnostics at the University of Oxford in Oxford, England, and at Muhimbili University of Health and Allied Sciences. Diagnosis of lymphoma is tricky even with highly trained pathologists, so the comprehensive model’s accuracy is key, particularly given current heavy reliance on the less accurate H&E staining.

Tackling the challenge of better and more timely diagnostic accuracy of Burkitt lymphoma is tremendously important, Rochford said, “but there are some real challenges with what they have done” that limit their goals and findings, she added. She noted the low recognition rate of MYC translocations, though using the comprehensive model helps address that. Rochford’s skepticism primarily arises from the expense and difficulty of establishing capacity for collection of cell-free DNA and conducting next generation sequencing (NGS) in such low-resource settings.

“There’s a huge problem with just getting reagents,” she said, but even if that problem can be solved, “how do you get the training there to support the technicians?”

The authors acknowledged in their paper that scientists in Tanzania involved in the study already had the lab skills to be rapidly trained to analyze cell-free DNA. Additionally, Schuh agreed with Rochford that establishing capacity for sample analysis in other locations would take years. However, if that capacity can eventually be established at multiple major population centers, cell-free DNA testing could complement the limited histopathology already occurring with fast, minimally invasive diagnosis at community hospitals, the authors wrote. Sequencing could occur at central hubs, perhaps relying on existing networks used for HIV or tuberculosis testing. Multiple pilots and pragmatic trials could determine how scalable, cost-effective, and sustainable broader use might be, they wrote.

Warren agreed that a “hub-and-spoke” approach could be feasible if capacity is built at enough hubs. “The principles established in this work are extremely important,” Warren said. He noted that liquid biopsy can be done with a very small amount of blood — just 1 mL — which is particularly important given how dehydrated many patients often are. Another advantage is that liquid biopsy is much less invasive than tissue biopsy, especially if the patient has an abdominal tumor instead of a jaw tumor.

The ability to accurately diagnose abdominal disease “is probably ten times more important than making the diagnosis from a jaw tumor,” Warren said, because of the difficulty of getting an abdominal biopsy. “A belly tumor in a young child could be ten different things, and timely diagnosis and initiation of treatment is really critical because this tumor progresses so fast,” he said.

There often isn’t time to wait for pathology because children often presenting late, many times after having been to faith healers.

“They come from rural areas, often many hours away by long, painful, slow bus rides,” an “incredible undertaking” that also requires an adult to take time off work, he said. Drawing a mL of blood and shipping it to a central facility in a large urban area, with results sent via text message, is much more efficient.

That said, Warren acknowledged Rochford’s concern about the feasibility of building capacity for liquid biopsy analysis. “Implementation will not be trivial,” Warren said.

Stepping Stone to Broader Cancer Screening

The biggest barrier to broader adoption of this method is the cost of NGS. The authors estimated the cost of liquid biopsy at $710.15 per patient, the largest single component of which is reagents, costing $175.48. Histopathology of tissue biopsies, meanwhile, cost much less at about $185.01 per patient. But the improved clinical outcomes could justify the higher upfront costs, the authors wrote, and broader adoption of NGS for other tests could also eventually drive down costs.

The training and initial capacity for NGS in Tanzania and Uganda was funded by a UK development fund and took several years, Schuh said. But about 5 years after that process began this study shows the potential payoff if training and logistics can be expanded to other diseases and in other parts of Africa.

“Once we’ve got this technology, we can also then start to evaluate whether it would be useable for other types of cancer in situations where screening programs are not available,” Schuh said.

The research was funded by the National Institute for Health and Care Research and the National Cancer Institute. The authors reported no disclosures. Warren and Rochford had no disclosures.

Tara Haelle is a science/health journalist based in Dallas.