The Yosemite oncology fund is moving faster than its founder expected. Reed Jobs, who launched the San Francisco-based venture firm in 2023, sat down with TechCrunch to take stock of a sector that has shifted dramatically beneath his feet: a historic pharmaceutical patent cliff, an acquisitive big-pharma market flush with post-pandemic cash, and AI accelerating drug discovery at a pace that has surprised even insiders.
The backdrop matters. Yosemite has grown to a team of 17 and roughly 25 companies across two funds. Its second fund held a $200 million first close, targeting $350 million, according to Venture Capital Journal. New limited partner Amgen joined that close alongside returning backers including Emerson Collective, John Doerr, MIT, and Memorial Sloan Kettering Cancer Center.
About a third of the fund goes into companies Yosemite builds itself, often alongside academics at Yale, Berkeley, and Stanford. The remaining two-thirds backs externally founded companies the firm wants to join. A further 2.5% of assets under management flows into a donor-advised fund: no-strings-attached grant money, plus $1 million per year drawn from management fees. Two of the 20 portfolio companies in Fund I originated directly from grants.
The Patent Cliff Pharma Can’t Outrun
Jobs points to a collision of forces that has changed the fundraising environment entirely. Interest rates have eased. More consequentially, large pharmaceutical companies are sitting on record cash reserves from the pandemic years, even as a cluster of blockbuster drugs approach the end of patent protection at roughly the same time.
The result: an acquisitive spree. Eli Lilly’s purchase of Kelonia Therapeutics illustrates both the scale and the direction of travel. Under the terms of that deal, Lilly will pay Kelonia shareholders $3.25 billion upfront, with up to a further $3.75 billion in clinical, regulatory, and commercial milestone payments, for total potential consideration of $7 billion. Kelonia’s lead candidate, KLN-1010, is a potentially first-in-class lentiviral in vivo anti-BCMA CAR-T therapy for multiple myeloma, currently in Phase 1 trial.
As CNBC reported, the acquisition is part of Lilly’s plan to grow beyond its GLP-1 franchise, with a strategy that blends early-stage deal-making with later-stage purchases of drugs that already carry clinical data. PharmExec notes that the deal signals a potential shift in the cell therapy landscape: in vivo approaches aim to sidestep the cost, complexity, and access problems that have constrained traditional CAR-T treatments.
Meanwhile, Revolution Medicines’ drug daraxonrasib has delivered one of oncology’s more striking recent results. In its pivotal Phase 3 study, the drug helped pancreatic cancer patients live nearly six months longer than those on standard-of-care chemotherapy, with results described as statistically significant and clinically meaningful, according to MedCity News. The FDA has also selected daraxonrasib for a pilot programme that compresses review time to one or two months for products addressing national health priorities. Revolution Medicines plans to seek regulatory approval based on the preliminary results.
Jobs had cited Revolution Medicines’ work on KRAS, one of the most commonly mutated cancer-driving genes, as an example of the shift now under way. Historically, KRAS was effectively undruggable: its surface lacked the natural pockets a drug molecule needs to latch onto. Scientists at Amgen found a cryptic binding site roughly a decade ago, leading to the first approved KRAS inhibitor, Lumakras. AI has since mapped the remaining variants and opened new blocking strategies that were previously out of reach.
From p53 to Histotripsy: What Yosemite Is Actually Betting On
The Yosemite oncology fund is pursuing what Jobs calls the hardest targets in cancer. The largest single bet is on p53, the tumour suppressor gene mutated or silenced in the vast majority of human cancers. Yosemite has three separate companies working on it across several strategies. Jobs frames it plainly: ‘almost every cancer has to knock it out to exist in the first place.’
Another portfolio company, Tune Therapeutics, is using epigenetic editing to silence hepatitis B, which affects over 250 million people and drives the bulk of liver cancer cases globally. Rather than altering the underlying DNA sequence, the approach adds or removes methyl groups at specific sites, turning gene activity up or down like a dimmer switch.
Histosonics sits outside Yosemite’s usual remit. It is a device company, which Jobs acknowledges is unusual for the firm. It uses histotripsy to destroy tumours non-invasively: creating and collapsing tiny air pockets to obliterate tissue with precision. Its lead programmes target pancreatic and liver tumours, a pairing Jobs considers natural given that most pancreatic cancer eventually spreads to the liver.
On AI, Jobs is bullish but precise. He sees it accelerating grunt work with reproducible outcomes and opening previously inaccessible parts of the genome. Historically, roughly 15% of the genome was druggable; AI is widening that window.
Two portfolio companies across both funds have failed, on scientific grounds rather than commercial ones. Jobs expects some failure at this stage and says Yosemite structures its investments in tranches against scientific milestones to contain the damage when it comes.
The pipeline, the LP base, and the deal environment have all surprised him to the upside. The question now is whether the science can keep pace with the ambition. If p53 falls, that answer becomes considerably clearer.
