Roche (OTCQX:RHHBY) took a hit in December 2010 when the FDA proposed revoking the breast cancer indication from Avastin, its top selling oncology drug for lack of benefit. Businessweek had estimated this could result in an annual loss of about a billion dollars in sales to the firm. While Avastin-- a targeted therapy that shuts down new blood vessel formation, or angiogenesis-- is broadly applicable across many tumor types, individuals are not tested for any particular tumor markers or mutations. That may change with this decision as company officials negotiate with the FDA. Roche is now considering studies to identify populations of breast cancer patients who will benefit from Avastin, continuing its march toward personalized medicine.
Berkeley based Plexxikon’s strategy with PLX4032 has always been the delivery of personalized medicine. This small molecule compound is a potent and selective inhibitor of mutated BRAF V600E, an oncogenic protein found in about half of melanomas as well as other solid tumors, particularly thyroid and colorectal cancer.
Plexxikon first began working on this compound in 2002, with the discovery of BRAF V600E in melanomas during the Cancer Genome Project. BRAF is a member of the MAPK pathway, transmitting signals from the cell surface to the nucleus to regulate survival and proliferation. Its mutated form is overactive, and under certain conditions, turns the cell cancerous. PLX4032 preferentially inhibits the mutated BRAF. In vitro studies showed cells harboring the V600E mutation were susceptible to the compound while wild-type cells were not.
In September 2006, Plexxikon filed an IND for the compound to begin clinical testing; a month later, Roche signed an agreement with Plexxikon for worldwide rights to PLX4032 in return for $40 million upfront and up to $660 million in milestones as well as royalties.
Roche's timing could not have been better. Even before the Phase I trial was complete, data was so compelling that the company began Phase II and Phase III studies. In a small Phase I study, tumors shrank by at least 30 percent in 24 of the 32 patients, with tumors disappearing completely in two patients. This was unheard of for metastatic melanoma.
Earlier this January, Plexxikon and Roche announced positive interim results of their 675 patient, randomized, open-label Phase III trial, BRIM3, indicating patients treated with PLX4032 had an improved overall survival (OS) and longer progression-free survival (PFS) compared to patients treated with dacarbazine, the current standard of care. The pre-specified endpoints for both of these criteria had been met.
Specific data has not been released and will be presented at a major scientific conference later this year. A regulatory filing for the drug, along with its co-developed companion diagnostic to detect mutated BRAF is expected in 2011. Sales projections range from $500 million to over $1 billion.
This isn't the end of the road for the BRAF project. Researchers have noted that although patients respond very well to PLX4032 initially, resistance is seen over time. New multi-targeted molecules from Plexxikon aim to solve this problem by simultaneously hitting BRAF and the PI3K pathway.
An early investment in Immunogen (IMGN
) is also paying off in a big way for Roche. For a mere $3 million upfront, $40 million in milestones, and mid-single digit royalties, Genentech (now Roche) gained worldwide rights to use Immunogen's antibody drug conjugation (ADC) technology. The technology, known as Tumor-Activated Prodrug (TAP) consisted of Immunogen's proprietary maytansinoid toxin and antibody-drug linker.
The first antibody Genentech fitted out with this technology was its blockbuster HER2 targeted breast cancer drug, Herceptin. The new drug, Trastuzumab-DM1, or T-DM1, if successful, had the potential to maintain Genentech's breast cancer therapy franchise.
Although effective, some patients continue to progress after Herceptin treatment. Side effects may also be harsh due to the need for concurrent treatment with chemotherapy drugs. An armed Herceptin carrying a highly potent payload could potentially negate the need for concurrent chemotherapy, reducing side-effects. Herceptin is also associated with cardiotoxicity due to previous treatment with doxorubicin; this is also expected to be eliminated with the use of the T-DM1 conjugate.
By 2002, promising preclinical data in mice showed tumors resistanct to Herceptin were exquisitely sensitive to T-DM1. In 2010, Genentech filed an accelerated BLA with the FDA based on strong data from a single armed Phase II trial of T-DM1 in highly refractory HER2 positive breast cancer patients. Results of the trial showed T-DM1 treatment shrank tumors in one-third of these patients. But to the company's surprise, the FDA rejected its BLA, saying all lines of treatment have not been exhausted for patients entering the trial. As usual, the FDA was unpredictable as ever.
This setback knocked a couple points off Roche's stock, but cut Immunogen's in half. Yet it was far from a death knell. The rationale behind antibody drug conjugates is now stronger than ever. Roche has a series of wide-ranging trials for T-DM1, from first line to combination to refractory, and even has plans for its use in adjuvant, or preventative therapy.
Recent results from a small Phase II of 137 women presented at the European Society of Medical Oncology (ESMO) in October 2010 showed that after six months, patients on T-DM1 had an overall response rate of 48% compared to 41% for Herceptin plus a taxane. For this patient size, the results were considered comparable. Perhaps more impressive, the rates of clinically relevant adverse events were much lower in the T-DM1 group at 37 percent, compared to 75 percent for women on Herceptin plus Taxotere.
An ongoing Phase III trial, EMILIA, is evaluating T-DM1 for second line use in preparation for marketing approval in the US and Europe in 2012. Roche estimates peak sales for T-DM1 could reach between $2-5 billion.
Pundits have speculated personalized medicines would require companies to sacrifice sales due to the smaller markets available for each targeted drug. It would appear this is not necessarily the case, as seen by potential blockbusters PLX4032 and T-DM1.
Disclosure: I am long RHHBY.PK.