The small patient pool for medullary thyroid cancer, or MTC, isn't going to move the needle for a large pharmaceutical company. But Exelixis (EXEL -0.48%) hopes the rare cancer will serve as launch point for a cancer-fighting franchise.

A small, important proof of concept
Only 4% of the roughly 60,000 people diagnosed with thyroid cancer this year will be diagnosed with the ultra rare MTC indication. For the most difficult to treat, Exelixis' Cometriq -- approved by the Food and Drug Administration last November -- offers hope.

In phase 3 trials, Cometriq was studied across 330 hard-to-treat MTC patients, including 92% who had already undergone thyroidectomy and 21% who had previously been treated with a tyrosine-kinase inhibitor, such as AstraZeneca's (AZN -0.29%) Caprelsa, which gained FDA approval in 2011. In the study, patients treated with Cometriq had an overall survival rate of 11.2 months versus just four months for placebo. 27% saw tumor size decrease, compared to zero for those receiving  the placebo. 

The results may be strong enough to threaten Caprelsa's position as a first line treatment. Despite Caprelsa having median progression-free survival of 22.6 months versus 16.4 months for placebo in phase 3 trials, the drug carries a high risk of toxicity. As a result, doctors need to gain certification through the FDA's Risk Evaluation and Mitigation Strategy, or REMS, program prior to prescribing the drug as a treatment.

The absence of such a restriction on Cometriq suggests that the drug may win a share of Caprelsa's sales, which totaled $30 million last year. However, despite the advantage, sales of Cometriq for MTC are likely to remain small given that the addressable market is just 500 to 700 patients annually. As a result, Cometriq's sales were just $4 million in the second quarter.

A big opportunity treating prostate cancer
Given the small market for Cometriq in MTC, the company's success or failure is tied more closely to its ability to commercialize Cometriq for other, more lucrative indications, such as metastatic castration-resistant prostate cancer, or mCRPC.  

At ASCO, Exelixis reported that mCRPC patients treated with Cometriq had median overall survival of 10.8 months. That's good news since 73% of those patients were heavily pre-treated with big-money drugs such as docetaxel, Zytiga, Xtandi, or Jevtana. 

Those drugs are racking up hundreds of millions in sales, suggesting a big opportunity for Cometriq.

Sales of Sanofi's Taxotere were $3 billion in 2010 prior to losing patent protection, and Johnson & Johnson's (JNJ 0.05%) Zytiga generated $395 million in sales during the second-quarter of this year. Xtandi -- co-developed by Medivation (MDVN) and Astellas and approved in August 2012 -- had $82 million in second-quarter sales. Sales of Sanofi's Jevtana totaled $54 million in the second quarter, bringing first half total sales to $106 million.

It also suggests that Cometriq will have to battle for share with recently improved therapies, including Medivation and Astellas' Xtandi.  In phase 3 tests, Xtandi patients had a median overall survival of 18.4 months versus just 13.6 months on placebo.  

Johnson & Johnson is unlikely to hand over its share of the prostate market either. As of December, Zytiga can be used prior to chemotherapy in late-stage mCRPC. That change helped drive 70% year-over-year sales growth for Zytiga. J&J also moved to shore up its prostate franchise in June, acquiring Aragon  for $1 billion to obtain Aragon's next generation ARN-509, a long-term potential competitor to Xtandi.

Given the competition, if Cometriq wants a large share of this market it needs to exceed or beat existing efficacy and safety benchmarks when its phase 3 data is released in 2014. 

A possible liver and kidney cancer drug, too
Cometriq is also in phase 3 trials as a treatment for both hepatocellular carcinoma -- or HCC, the most common form of liver cancer -- and renal cell carcinoma -- or RCC, the most common form of kidney cancer.

Patients diagnosed with these cancers would applaud new treatment options. HCC has a high mortality rate, particularly abroad, and cases have tripled in the U.S. from 1975 to 2005.  

The incidence rate of kidney and renal cancer is 15.3 cases per 100,000 people, and five-year survival is 71.8%, according to the National Cancer Institute. 

The size and significant need makes these indications worth billions.  

Currently, HCC patients are treated with Nexavar, co-developed by Bayer and Onyx Pharmaceuticals. Nexavar generated $900 million in sales across all its indications last year, prompting Amgen to spend $10 billion acquiring Onyx in August.  

In kidney cancer, Exelixis would compete against Pfizer's (PFE 0.19%) Inlyta, which was approved as an RCC treatment in 2012. Inlyta generated sales of $71 million in Q2 and $63 million in Q1, giving it 26% share of the second-line market. In phase 3 trails, Inlyta patients saw progression-free survival of 6.7 months versus 4.7 months for patients treated with Nexavar.

Cometriq would also compete against Genentech's blockbuster Avastin and Novartis' Afinitor, which brought in $797 million in revenue during 2012.

Collaborations hint at potential suitors
Cometriq has a lot of irons in the fire, and a larger player may find the drug interesting, particularly if phase 3 trials go as hoped.  Some of these big companies are already collaborating with Exelixis on a variety of compounds.

In January, Genentech announced that it's entering phase 3 trials for a combo drug incorporating Exelixis' GDC-0973 alongside its Zelboraf as a treatment for malignant melanoma. The two companies have been partners since 2006, and if successful, Exelixis will make 50% of eventual sales up to $400 million and 30% of sales above that level.

The company has also worked with Bristol-Myers Squibb for years on various drug development projects and is collaborating with GlaxoSmithKline, Merck, and Sanofi.

Cash burn and valuation are big risks
Exelixis is betting big money can build a franchise around Cometriq.  In a bond and share offering last year, the company raised $416 million to pay for ongoing trials. However, it's going through that cash quickly. The company's stockpile dropped from $634 million at the end of 2012 to $567 million in the first quarter and $524 million in the second quarter.  

Spending has increased everywhere, including $49 million on research and development in the second quarter, up from $33 million a year ago. Selling, general, and administrative costs jumped to $13.2 million from $6.8 million last year. And interest expense popped to $10.9 million last quarter from $3.8 million last year thanks to the bond offering. At current burn rates, Exelixis expects to finish the year with $400 million.

That should be enough money to advance its current phase 3 trials. If prostate data next year is strong, Cometriq could see new indications added to its label in 2015. If so, Cometriq could grow from niche status to a franchise worth hundreds of millions.

How you value that franchise is a big question. Amgen paid 16 times Onyx's annualized second quarter earnings to acquire it last month. For comparison, Exelixis is trading an eye-popping 62 times annualized second-quarter sales. This suggests that Exelixis is best suited for only the most speculative investor, willing to take a big bet Cometriq will succeed in late stage trials.