NEW YORK – Following the results of a validation study, startup Nephrosant plans to launch its urine-based QiSant assay — which provides a threshold score for kidney transplant rejection risk and a numerical score for the intensity of rejection — as a laboratory-developed test later this year.
The San Francisco-based firm believes the serial monitoring assay could guide the clinical use of immunosuppressive drugs and other therapies to prevent a recipient's body from rejecting a donated kidney.
To diagnose acute rejection in kidney transplant patients, clinicians usually extract several small transplanted kidney samples every few months post-transplant, in addition to measuring the patient's serum creatinine levels. However, kidney biopsies can be invasive and expensive, while serum creatinine levels are often inaccurate and not sensitive enough to detect transplant rejection.
"If you decide to do a protocol biopsy in the first year, every three to six months, you'll see some level of rejection in 30 percent of kidney biopsies, which could be mild, or very severe at the time," Minnie Sarwal, Nephrosant cofounder and UCSF professor of surgery, explained. "But a biopsy isn't something we can do every week, which is why we believe [QiSant] will give a more accurate readout of whether the kidney has immune quiescence or risk of rejection."
Initially spun out from UCSF in 2017 as KIT Bio, the startup was rebranded Nephrosant in 2019 after establishing an off-campus laboratory. Joshua Yang, Nephrosant cofounder and director of assay development, said the firm seeks to noninvasively assess kidney health in patients, starting with kidney transplant rejection prediction.
Adapting its previously developed kidney injury assay, Nephrosant's QiSant assay analyzes six biomarkers from 4ml of urine sample: the amount of cell-free DNA (cfDNA); the fraction of methylated cfDNA; the proteins clusterin and creatinine; the inflammation marker CXCL10; and total protein amount in the urine. The assay uses a proprietary enzyme-linked immunosorbent (ELISA)-based tool, including a 5' biotinylated oligonucleated immunoprobe to target cfDNA fragments, and artificial intelligence to estimate the likelihood of acute kidney rejection.
In the validation study, published Wednesday in Science Translational Medicine, Sarwal and her team at UCSF evaluated the performance of the QiSant's urine score (Q score) to diagnose biopsy-confirmed acute renal transplant rejection.
"Because the biomarkers were already developed for early spectrum of injury, we used a different algorithm for the sensitive and accurate detection kidney transplant rejection," Sarwal explained.
The researchers collected a total of 601 prospective urine samples from both pediatric and adult renal allograft recipients immediately before a renal allograft biopsy. Each sample was then paired with a renal transplant biopsy and classified into the following diagnoses: stable (170); acute rejection (AR, 103); borderline AR (bAR, 50); and BK virus nephropathy (9). The team also collected additional urine samples from 32 patients with AR before the rejection episode and paired them with biopsies.
After collecting patient samples, the researchers developed a composite Q score — ranging from 0 to 100 — on all six biomarkers in a training set of 39 AR and 72 STA patients, which Sarwal said allows for sensitive tracking of AR progression and recovery.
"By including the six biomarkers, as well as patient data such as demographics and age, we applied artificial intelligence to build a urine score that works in two different ways," Sarwal explained. "The score gives a binary cutoff for rejection, as well as a way to scale the intensity of injury and rejection."
Using logistic regression, the team established an optimal threshold of 32 to maximize sensitivity and specificity for detecting rejection in the urine sample paired with an allograft biopsy in the training set.
The researchers then applied the Q score threshold to two independent cohorts to determine its clinical performance for distinguishing between rejection and immune quiescence.
In the first validation set, which had 32 AR patients and 71 stable patients, the group found that the score between the patient types had about a 91 percent clinical sensitivity and a 92 percent clinical specificity. Meanwhile, in a second validation set of 32 AR patients and 27 stable patients, the team found that the scaled score had a 100 percent sensitivity and a 96 percent specificity.
Aggregating the training and validation sets together, the team found that the Q score performed better than proteinuria and eGFR. At the quiescence threshold of 32, the team found that the urine score had a 95 percent clinical sensitivity and 96 percent clinical specificity.
The researchers also saw that most patients with samples (159) with scores above the AR threshold had a clinical diagnosis of active AR, early AR, or went on to develop biopsy-confirmed AR up to 200 days after using the QiSant assay.
However, if a patient was scheduled for a tissue biopsy and the urine score was below the threshold, the chance of the patient having AR was only 5 percent. Therefore, the researchers determined that a score of lower than 32 could indicate "that the allograft is likely to be in state of immune quiescence," and would not necessarily require a protocol biopsy.
The study authors noted that the study had a limited assessment of the longitudinal samples, especially from stable patients, whose natural variation of the urine score could be examined over time. The authors believe that they will need to collect additional monthly samples from patients from ongoing clinical studies.
Because the urine test highly correlates with tissue biopsy results, Yang believes the Q-score offers utility for serial monitoring that may not be possible with invasive kidney biopsies. He also highlighted that the test was able to predict episodes of acute kidney allograft rejection in both pediatric and adult patients up to seven months earlier than serum creatinine testing.
The study authors highlighted that the findings "suggest that this urine score has the potential to replace the renal biopsy as the gold standard of posttransplant monitoring, and that such measurements can be used for the detection of rejection episodes … [and] to guide the use of immunosuppression, with the ultimate goal of controlling subclinical intragraft inflammation and thus prolong graft survival."
Commercial launch
Nephrosant now plans to launch the kidney transplant rejection assay as an LDT later this year. The test will be performed in a recently built CAP-accredited, CLIA-certified lab in South San Francisco, California. The firm expects to potentially either kit the test or offer a lateral flow assay for regions outside the US.
Sarwal envisions offering QiSant as part of an integrated benchtop workflow for serial urine monitoring and transplant rejection detection in the clinical space. While initially launching QiSant as a complementary tool for tissue biopsy, Nephrosant believes that clinicians will eventually begin using the test as an alternative or a replacement for the invasive procedure.
Clinicians aiming to determine a transplant's risk of rejection can collect a patient's urine sample in a universal urine cup, add a supplied preservative, and then mail the sample in a prepaid package to Nephrosant's lab. The firm then tests the sample with the QiSant assay, which Sarwal said can provide actionable results within two days.
Yang said that Nephrosant has raised about $6 million in funding through venture capital investments and through the issuance of convertible notes, with Silicon Valley, California-based venture capital firm DigitalDx Ventures acting as the firm's leading investor. He noted that Nephrosant anticipates launching the next funding round later this year to help further commercialize the assay.
Sarwal acknowledged that the firm has dealt with several challenges while developing the technology for kidney-related issues. She explained that the major problem involved simplifying the assay's collection process while maximizing results and ensuring that the biomarkers were not degraded prior to analysis. Nephrosant also needed to figure out how to collect a urine sample and avoid processing the sample at the site of collection. The group, therefore, developed a patented preservative to stabilize biomarkers, especially cfDNA, when the samples are unable to be processed right away.
Declining to comment on the anticipated commercial price of the QiSant assay, Sarwal said that she expects the test to be fully covered by Medicare or private insurance. While Nephrosant has not spoken to private payors Sarwal noted that her team has begun discussions with Medicare
Sarwal said that Nephrosant also aims to work with strategic partners to incorporate the assay in prospective drug trials. The firm is currently working with undisclosed pharmaceutical groups to use the QiSant assay to measure the efficacy of immunosuppressant drugs in randomized clinical trials.
"We anticipate that the QiSant's … inclusion for serial testing after transplantation in the context of clinical trials could provide an important quantitative surrogate to evaluate the clinical efficacy of new immunosuppressive drugs in ongoing trials," Sarwal explained. "The QiSant test can be reflexive for transplant patient management, for early detection, treatment, and follow-up of rejection."
While Nephrosant is collaborating with SRI International and DCN Diagnostics to develop a dipstick version of the assay for the point-of-care space, Sarwal noted that the team's primary focus is to launch the LDT as the firm's first commercial test for the transplant space. Afterward, the group will return its focus on existing research collaborations.
Sarwal also highlighted that Nephrosant expects to launch an assay for patients at high risk for chronic kidney disease by 2021, but she did not disclose more information at this time.