How Jnana picked PKU
How Jnana picked PKU
Unlocking platform value meant finding its first killer app
Image courtesy of Jnana Therapeutics https://www.jnanatx.com/rapid-platform/
In 2017, Joanne Kotz (Jnana CEO), Stuart Schreiber (Broad/Harvard), Ramnik Xavier (Broad/MGH) and one of us (Joel Barrish, CSO) co-founded Jnana Therapeutics to pursue modulators of solute carrier (SLC) transporters, proteins which ensure that hundreds of critical small molecule metabolites in the body are in the right place at the right time.
On August 1, Jnana was acquired by Otsuka Pharmaceutical Co. for $800 million up-front plus milestones. Otsuka’s big prize is JNT-517, a potential first-in-class small molecule pill to treat phenylketonuria (PKU), a rare, inherited metabolic disorder (you may only know of it from the heel-stick blood test for many genetic diseases that newborns receive when hospitals screen for the condition).
August 16, 2024
Written by:
Joel Barrish
Joel Barrish is an Advisor at Raven, RA Capital’s healthcare incubator.
Jaclyn Henderson
Jaclyn Henderson is an Entrepreneur in Residence at Raven, RA Capital’s healthcare incubator.
As a co-founder of the company (Joel) and the scientist who conceived of the PKU program at Jnana (Jaclyn), we know that the story of how PKU became the first killer app for Jnana’s SLC discovery platform — a key pivot point in demonstrating the platform’s value — didn’t quite go in a straight line. The story of ‘517 is emblematic of the calculated risk taking that we embrace in biotech. For starters, there were no systematic drug discovery technologies existing for SLCs, and relatively few SLCs clinically validated as targets. We needed new approaches to build a pipeline. Starting with a blank slate is exciting – the ability to create a platform, a pipeline, a team from scratch is a big reason why one comes to biotech – but it requires a fearless and innovative team and bold investors and partners.
Building a foundation
Photo by Iain Kennedy on Unsplash
SLCs are important in human health and disease – 20 of these proteins have been targeted by approved small molecule drugs including compounds such as Prozac, a selective serotonin reuptake inhibitor for depression, and Ritalin, a dopamine transport blocker for ADHD. Both were discovered before it was even known that their protein targets were SLCs. Now we know much more about this group of molecular chaperones and with more than 450 members, we know this target class has enormous untapped potential.
When we were helping to build Jnana, we knew that our drug discovery platform would be the central foundation for the company, enabling everything we did. But SLCs are a dauntingly diverse target class, with 65 subclasses featuring variations in structural motifs, metabolite substrates, and cellular locale. They are also difficult to express and purify and, because they reside in cell membranes, were not readily amenable to biophysical methods when we began.
The early Jnana team, especially Justin Rettenmaier and Matt Labenski, realized that the optimal approach was one that was more general in nature, one that was less SLC-specific but could in theory be applied to other proteins. With significant experience in chemical biology, they focused on identifying SLC small molecule binders that could lead to all kinds of pharmacological modalities – not only inhibitors but compounds that could activate the transporter or affect its cellular location or half-life. Being able to identify binders to fully interrogate the SLC protein and discover any binding site – orthosteric or allosteric – was paramount.
Leveraging recent advances in chemical biology and chemoproteomics combined with additional proprietary methodologies, the result of this home-grown effort was RAPID (Reactive Affinity Probe Interaction Discovery), an approach to quickly identify progressible small molecules, within weeks. The team recognized early on that the initial covalent binders (Reactive Affinity Probes or RAPs) to any SLC could be used in a competitive displacement binding assay to screen large libraries of small molecules, a defining moment for the new platform. With a general solution to hit generation in hand, we were then able to turn to applying it to targets and diseases where we could have significant impact on patients.
Where to start?
As with the platform blank slate, choosing where to begin Jnana’s journey to identify SLC modulators that could impact disease was a formidable task. Initial targets were guided by strong human genetic associations, however these presented significant biological and technical challenges, such as identifying small molecules which could increase the activity of these transporters.
Searching for a target with a more tractable path we continued to be guided by human data but with a switch in focus from human genetics to metabolomics. In many diseases where systemic or organ-specific metabolite levels are critical drivers of the disease, there is an SLC that is clearly associated with controlling those metabolite concentrations. The attraction of these metabolite-dependent diseases is that the translational path to the clinic and beyond is clearly defined by those metabolite levels. Phenylketonuria (PKU) is one such disease.
PKU is a genetic disease caused not by a defect in an SLC, but instead by the loss of the enzyme which metabolizes the amino acid phenylalanine (Phe). Build-up of Phe, which is a component of all dietary protein, causes severe neurological deficits. To avoid these symptoms, PKU patients (who are identified through that heel-prick blood test at birth) must significantly reduce their protein intake. Meat, fish, and dairy are all off the menu, and even many fruits and vegetables contain significant Phe and should be reduced or avoided.
Because PKU is caused by a build-up of the metabolite Phe in the blood, we looked for an SLC transporter that is involved in its uptake, either from food in the stomach/intestines or reabsorption from urine in the kidney. SLC6A19 is a key transporter of neutral amino acids and is expressed almost exclusively in the kidney and small intestine. Inhibiting SLC6A19, we wagered, should reduce the amount of Phe absorbed from the diet and increase excretion of Phe in urine.
But SLC6A19 transports several amino acids, not just Phe. Would targeting this transporter lead to side effects from deficits of the other amino acids? Human data helped answer this key question – there are people with partial to total loss of this transporter, a condition called Hartnup disorder. Luckily these people are generally healthy and have normal levels of amino acids in their blood. And so while reduced levels of SLC6A19 transport appear to have no impact in healthy individuals, blocking the same transporter in PKU patients would increase the urinary excretion of Phe, thereby reducing plasma Phe levels. Our confidence in this approach was also bolstered by inhibitors of the related SGLT transporters SLC5A1 and/or SLC5A2 such as Farxiga, which increase urinary excretion of glucose and reduce blood glucose levels.
A crowded space
An excerpt from an Orphan Disease: Loss of Function/Extracellular Factors disease map (2022) from TechAtlas, RA Capital’s knowledge engine.
Before starting on the project there was one other big question: PKU had become the poster child for almost every new modality. Messenger RNA, microbiome, cell and gene therapy companies have all taken a swing. Not only did this mean significant competition, but most of these companies were already approaching the clinic. If we were successful with our PKU program would there still be a market? Would investors back a small molecule approach over these more in-vogue and but less well-tested modalities?
Market research gave the theory a boost. In particular, the Boston-based PKU expert Harvey Levy at Children’s Hospital, who had been seeing patients for more than 30 years, suggested that PKU patients might prefer a daily pill rather than the unknown risks of an mRNA or gene therapy, and patients also had concerns the efficacy with the latter may eventually wear off. Importantly, Sanofi published data showing that a genetic knock-out of SLC6A19 in a mouse model of PKU had a significant effect on blood Phe levels, giving additional confidence in the target and mechanism. The SLC6A19 project was a go!
Although the project initially seemed straightforward, it was not without its twists and turns. The RAPID platform lived up to its name, yielding a highly potent and selective hit within weeks. The team hit a speed bump when despite a promising in vitro profile our early analogues flopped on initial in vivo testing. After some scrutiny we found that our compounds were selective for the human homolog of SLC6A19 and did not inhibit the mouse SLC, making optimization and preclinical profiling more challenging. Ultimately a cryo-EM structure revealed the allosteric site where the compounds bind, and the reasons for the significant difference between species. The hit was ultimately optimized under the leadership of Discovery Sciences Head Dean Brown to create JNT-517, Jnana’s first-in-class program.
Investors like RA Capital who came on board after the PKU program was in progress saw it as critical validation for the RAPID platform. As we think about new platform builds, we aim to match the platform to the right killer apps as early as possible. And a shameless plug for anyone still figuring out your own killer app: talk to RA’s TechAtlas team, who can boil the ocean of possibilities, navigate crowded spaces, recognize when incumbents have weaknesses that can be improved upon, and size markets.
‘Pee the Phe’!
Given its excellent preclinical profile through IND-enabling studies, JNT-517 entered a randomized, double-blind, placebo-controlled Phase Ia trial in healthy volunteers in Australia in late October 2022. Biomarker data were exactly as expected based on the preclinical SLC6A19 knockout data as well as the Hartnup disorder profile: a dose-dependent aminoaciduria (urinary increase in levels of amino acids transported by SLC6A19) without clinically significant changes in plasma amino acid levels.
Based on these compelling proof-of-mechanism data and its excellent safety and tolerability profile, in August 2023, JNT-517 progressed to a Phase 1b study. In January this year, the company announced positive interim results and said it will meet with regulators and seek to advance JNT-517 directly into a pivotal Phase 3 study in the first half of 2025.
An oral pill that is simply administered with an easy-to-understand mechanism (‘Pee the Phe’) has generated excitement with both PKU patients and their physicians. Like most endeavors in biotech, the strides taken by Jnana in building out its platform and this lead program in PKU took the incredible efforts of a strong and determined team, who collectively navigated over, around, and through the inevitable scientific obstacles. We are pleased to see Jnana R&D continue as a subsidiary of Otsuka and are looking forward to additional PKU data as well as advancement of Jnana’s immune portfolio – both SLCs and now challenging non-SLC targets – enabled by their unique small molecule RAPID platform. Congratulations to the entire Jnana team!