Biomarker-guided targeted therapy has become a standard approach to oncology R&D, but few new targeted drugs have reached the market for lung cancer. A new wave of research could change that, and improve the outlook for lung cancer treatment.

Virtually every major pharma and dozens of biotechs are running lung cancer trials. Some 130 drugs are in development for the disease, according to the market research firm Datamonitor.

Research on mutation analysis in lung cancer – as in other cancers – has been under way for years, dating back to the early part of the last decade at least. AstraZeneca PLC's fiasco with Iressa (gefitinib), an epidermal growth factor inhibitor, ironically, helped forge the way. The drug got FDA approval in 2003 for a broad refractory lung cancer population, but its use was later restricted in the U.S. due to lack of efficacy, and, after much scientific effort, it is available elsewhere in the world because the company was able to show efficacy in certain patient subtypes.

The multi-drug BATTLE study (Biomarker-integrated Approaches of Targeted Therapy for Lung Cancer Elimination) has also been pivotal in identifying biomarkers for lung cancer. The Phase II trial, released at the American Association for Cancer Research in April 2010, had a number of interesting findings, including that lung cancer patients with KRAS mutations did better with Bayer/Onyx’s Nexavar (sorafenib), while others with EGFR mutations fared better with Roche Holdings/Astellas Pharma's Tarceva (erlotinib) (Also see "For Nexavar In Lung Cancer, Focus Moves From First To Second Line" - Pink Sheet, 14 June, 2010.). BATTLE was too small by itself to initiate broad change in clinical practice, but follow-up trials are planned, as is another BATTLE trial.

Overall, however, work on mutations hasn't translated into drugs that significantly alter clinical practice. That could be about to change as new drugs reach the market and move ahead in clinical research.

A Growing Field Of Mutations To Target

"There is now a lot of excitement about lung cancer, excitement about the clinical trials for lung cancer, excitement about how well some patients are doing," said Pasi Jänne, MD/PhD, an associate professor of medicine at Harvard Medical School and a lung cancer researcher who has worked extensively in the area of EGFR mutations.

EGFR mutations have been at the forefront, but the EML4-anaplastic lymphoma kinase fusion mutation also appears to be a viable target for drugs that could impact the disease. Tyrosine kinase genes, including KRAS, NRAS, BRAF, P13KCA and c-MET, also contain mutations that could potentially be important in lung cancer. Drugs targeting these mutations are in earlier stages of development.

"For about 25% to 30% of patients, we have some genetic alteration which is potentially treatable with a drug," Jänne said.

Turning the promise of a biomarker approach into a commercial reality has not been easy.

Turning the promise of a biomarker approach into a commercial reality has not been easy, however. For starters, most mutations are found in small subgroups, each representing less than 5% of lung cancer patients. Just as importantly, efforts to develop clinically validated, accurate diagnostic tests to identify the patients who will respond to particular drugs are still in their infancy.

Many Drugs, But Bleak Outcomes

Complex hurdles challenge drug development in lung cancer. Patients are generally diagnosed late, when their disease is advanced and therefore harder to treat. Because the patient population is large and heterogeneous, non-targeted drugs haven't shown significant clinical benefit in trials.

Running an expensive Phase III trial has largely been a death-knell. Treatments for non-small cell lung cancer had the worst clinical success rate from Phase I to FDA approval within oncology, just 2%, in a recent study by the Biotechnology Industry Organization and the institutional research service firm BioMedTracker, which evaluated clinical-stage drugs from year-end 2003 to year-end 2010 (Also see "Cancer Drugs Have The Worst Phase III Track Record, BIO Study Shows" - Pink Sheet, 15 February, 2011.).

But the unmet need is great. More people die from lung cancer each year than the next three most common cancers combined (colon, breast and prostate), according to the American Cancer Society. An estimated 157,300 Americans are expected to die from lung cancer in 2010, accounting for 28% of all cancer deaths. Roughly 220,000 new cases of lung cancer were diagnosed in 2010.

Treatment decisions for advanced lung cancer are currently driven by histology, health status, smoking status, ethnicity, age and, increasingly, gene mutations. A chemotherapy regimen of drugs like gemcitabine, docetaxel or vinorelbine is often given as first-line therapy, as are platinum-based chemotherapies. Targeted agents like Roche /Genentech's VEGF inhibitor Avastin (bevacizumab) are also frequently used as first-line agents, and Lilly's tailored chemotherapy Alimta (pemetrexed) has become a preferred choice for the non-squamous form of NSCLC, both first-line and as a maintenance therapy. Tarceva is approved as a second-line treatment for NSCLC and is paving the way for a biomarker-driven approach to the disease.

Despite the options, median survival for patients on treatment remains just a little over one year.

Clinicians have long segmented patients based on histology, which includes small cell lung cancer, the more aggressive lung disease, and non-small cell lung cancer, which includes various sub-types: adenocarcinoma, squamous cell and large cell. Patients with NSCLC make up 80% of cases, and thus drug makers are most focused on that disease segment. Within the lung cancer pipeline, 84% of the drugs in development are for NSCLC and just 16% for SCLC, according to Datamonitor.

Scientists say the disease will be sub-typed even further into a collection of genetically distinct diseases. "We are going to move to a situation where we are not only going to talk about the type of lung cancers – what they look like under the microscope – but go one level deeper and say it’s a lung cancer with this particular genetic alternation," said Jänne.

Tarceva And EGFR Inhibitors Are Pioneers

The near-term focus is on Tarceva, which demonstrated a more substantial PFS benefit in patients with EGFR mutations in an interim analysis of a trial called EURTAC.

Tarceva is currently approved as maintenance and second-line therapy in the U.S. and Europe for patients with advanced NSCLC, regardless of EGFR mutation status, and its marketer, Roche, is seeking to broaden the label to first-line use. Because of the positive signals in mutation-positive patients, the sponsor halted the study in January (Also see "Tarceva Could Lock Down Market For First-Line Non-Small Cell Lung Cancer With EGFR Mutation" - Pink Sheet, 28 January, 2011.).

The full data will be presented at the American Society of Clinical Oncology annual meeting June 3, but the top-line results are already bumping up off-label use in the first line, in some instances. Sales of Tarceva are benefiting from its uptake in NSCLC, with sales up 8% to CHF 317 million ($358.9 million) in the first quarter. It is also approved for pancreatic cancer.

Roche is developing an EGFR diagnostic for use with Tarceva, under a sub-license from Genzyme.

Roche has already submitted the data to European regulators and is planning to submit the results to FDA by the end of the year. The company is also developing a polymerase chain reaction (PCR)-based companion diagnostic to test for EGFR mutations, under a sub-license from Genzyme Corporation.

EGFR mutations, which fuel tumor growth, are found in about 10% of NSCLC patients, making it one of the larger sub-types thus far identified. And physicians are most comfortable with the EGFR mutation as increasing evidence has linked it to NSCLC for some time.

Iressa put a spotlight on the class of drugs for years and, despite its early controversy, helped move the field forward. FDA granted accelerated approval as a third-line treatment based on Phase II results. But access was ultimately restricted in the U.S. after Iressa later failed to show a survival benefit in a larger trial. AstraZeneca eventually found Iressa's niche in EGFR-mutation positive patients after demonstrating a benefit in such patients in the iPASS trial in 2009. Iressa is available in Europe and other countries around the world.

In NSCLC, Testing Is Not Routine

Nonetheless, most patients are not routinely tested for the mutation. There is no FDA-approved test on the market, though laboratory tests are available. The practicalities of testing in clinical practice pose challenges to Tarceva – and to other biomarker-targeted drugs in development. The issues largely revolve around the size and quality of the tissue sample that is required and the accuracy of existing tests. About 30% of NSCLC patients are tested for EGFR in the U.S., according to Roche, and the figure is substantially lower in Europe.

"The biggest hurdle to get [patients] is you need to get them tested," said Ashwin Dandekar, regional manager of the New York office for the consulting firm Campbell Alliance. "That is a huge problem because currently testing is not that common."

Kenneth Algazy, MD, clinical professor of medicine at the University of Pennsylvania and a lung cancer specialist, agreed that testing remains a serious challenge. Most lung cancer patients don't receive full biopsies. Instead, a small amount of tumor tissue is collected through a needle procedure known as an aspiration cytology, which means the size and quality may not be sufficient for identifying genetic mutations.

"The markers have yet to prove so important that you will go into the chest to find out," Algazy said. "The biggest thing that could happen now is the ability to do these tests with a minimum of tissue, and that is what will drive survival statistics to be better then they are currently."

Some physicians rely on surrogate markers like smoking history and ethnicity instead to prescribe Tarceva. For example, EGFR mutations tend to be more frequent in non-smokers and in people of Asian heritage.

But increasingly, industry groups like the National Comprehensive Cancer Network and American Society of Clinical Oncology are recommending EGFR testing for patients prior to initiating first-line treatment with an EGFR inhibitor. Both have issued opinions on the topic this year.

In its Provisional Clinical Opinion on EGFR targeted drugs released in April, ASCO recommended in favor of testing but limited the recommendation to a test known as mutation analysis, which it says has proven the most reliable. Fluorescent in situ hybridization (FISH) and immunohistochemistry testing for EGFR "at the present time are not recommended for the purpose of treatment decisions because they do not reproducibly predict outcomes," ASCO said.

Launch Of Crizotinib Will Muddy The Waters

The issues will be further complicated as new drugs enter the market targeting other, even smaller subsets of patients, like those with ALK or BRAF mutations. Then, questions will arise about how to determine a priority for molecular screening.

Pfizer Inc. is likely to be the next drug maker to test the waters with the anticipated approval and launch of the ALK inhibitor crizotinib. The company announced FDA had accepted its NDA submission for crizotinib and granted a six-month priority review May 17. The first survival data, released in an abstract by ASCO May 18, showed a survival benefit in patients treated with the drug compared to historical controls (Also see "Survival Data For Pfizer's Crizotinib Appears Promising, Positioning It For Accelerated Approval" - Pink Sheet, 19 May, 2011.).

In 82 ALK-positive patients treated with crizotinib, one-year overall survival was 77% and two-year overall survival was 64%; median overall survival had not yet been reached. Among 37 historical ALK-positive controls, one- and two-year overall survival rates were 73% and 33%, respectively, and median overall survival was 20 months. Nonetheless, the data from the single-arm Phase I trial are early and Pfizer is still recruiting patients in an ongoing Phase III study.

Because lung cancer is such a widespread disease, drugs targeting even small mutations like ALK are commercially viable – as long as doctors have means to identify the patients.

The ALK mutation is found in less than 5% of NSCLC patients, and thus Pfizer will need to track the right patients and do so early in the course of treatment to realize crizotinib's commercial potential. Because lung cancer is such a widespread disease, drugs targeting even small mutations like ALK are commercially viable – as long as doctors have means to identify the patients.

Strong efficacy data will help to convince payers, patients and physicians that testing is worthwhile, but a test will need to be made widely available and be sensitive enough to base accurate treatment decisions on. False negatives in such a refined patient population will result in critical missed opportunities for Pfizer and could eventually lead community oncologists to dismiss testing as fruitless. Pfizer is partnered with Abbott Laboratories Inc. on the development of a companion diagnostic, based on the FISH platform, which has also been submitted to FDA.

But Pfizer has largely downplayed the challenges ahead with getting physicians and patients on board with routine biomarker testing. "If you step back and think about the way we approach breast cancer, it is routine for a woman who has a breast biopsy or resection to have the surgeon remove that and send it to the pathologist for molecular testing," Mace Rothenberg, senior VP of clinical development and medical affairs for the Pfizer Oncology Business Unit, said at a press briefing at the company's headquarters May 17. "It is not a huge leap to think about us doing that now for other cancers."

Further challenges will arise when it comes to using targeted agents in combination, which may ultimately be the best way to impact patient survival.

Pfizer is conducting Phase I trials of crizotinib in combination with Tarceva and separately in combination with PF299804, a pan-HER inhibitor.

Genentech is studying Tarceva in combination with its MET inhibitor, MetMAb. Top-line results from a randomized Phase II study released in an ASCO abstract May 18 showed that patients whose tumors had high levels of Met who were treated with MetMAb plus Tarceva had significant improvements in PFS (2.9 months) and OS (12.6 months) compared to Tarceva alone (1.5 months and 3.8 months, respectively).

Despite the positive outcome for patients with high-Met tumors, those categorized as having low-Met tumors had worse outcomes when given MetMAb plus Tarceva in combination (PFS of 1.4 months versus 2.7 months for Tarceva alone and OS of 8.1 months compared to 15.3 months for Tarceva alone).

The data highlight that targeting treatments to the right patients is as much about avoiding negative outcomes as getting positive ones. Having an accurate Met test approved by FDA will thus be critical to the development program. Roche is developing a test to analyze Met protein levels using an IHC platform.

Genentech's Chris Bowden, VP-product development, oncology, signaling franchise head, acknowledged there remain obstacles to getting proper tissue samples in lung cancer. "That is always something that is there and is a potential barrier," he said in an interview. The aim is that the efficacy benefit will be enough to persuade patients and physicians to get tested. "Hopefully, people will see that and make the choice with that in mind," Bowden said.

The hurdles will be significant, but many similar drugs – developed with companion diagnostics – are moving through the pipeline and will hopefully advance the field.

By Jessica Merrill

NSCLC Market Snapshot: Promising Biomarkers, Testing Challenges

More from Archive

More from Pink Sheet