Hematology/Oncology

Until about 5 years ago, non–small cell lung cancer was considered a homogeneous disease and was treated the same way in all patients, with only factors such as age and performance status to serve as markers of how well a patient might respond to treatment.

Virtually all patients with the disease received platinum-based chemotherapy. But now, advanced NSCLC has been shown to have a variety of different phenotypes, varying in histology and molecular markers that indicate treatment for the disease is not a one-size-fits-all approach.

“Over the past 5 years, we’ve started recognizing differences in effectiveness and toxicity of therapies in squamous cell cancers vs. non-squamous cell cancers,” Edward Kim, MD, associate professor and chief of Head and Neck Medical Oncology at The University of Texas MD Anderson Cancer Center, told HemOnc Today. “More recently, we have had the integration of biomarker assessment in a very early setting to individualize therapy and not treat patients just based on histology or performance status.”

There have been several practice-changing studies regarding the treatment of NSCLC, ranging from the inclusion of bevacizumab (Avastin, Genentech) in treatment for non-squamous NSCLC, to the addition of therapies that target mutations in the epidermal growth factor receptor or translocations in the anaplastic lymphoma kinase (ALK), to maintenance therapy after completing induction chemotherapy.

Together, EGFR mutations and ALK translocations are present in approximately 20% of patients with NSCLC. These discoveries, among others, have led to an era of truly personalized medicine for advanced non-squamous NSCLC.

“Five years ago, when a diagnosis of NSCLC was made, everyone received the same chemotherapy. The fact that we have changed this represents a real paradigm shift in how we approach these patients,” Mark Kris, MD, chief of the thoracic oncology service and the William and Joy Ruane Chair in Thoracic Oncology at Memorial Sloan-Kettering Cancer Center, said in an interview. “There are now early initiatives to do the same thing in squamous cell NSCLC. Within the next 5 years, every lung cancer is going to be analyzed at diagnosis, and treatment will likely be recommended because of information obtained from the initial biopsy.”

Mark Kris, MD, chief of the thoracic oncology service at Memorial Sloan-Kettering Cancer Center, said recent discoveries have led to an era of personalized medicine for advanced non-squamous NSCLC. Photo courtesy of Mark Kris, MD

HemOnc Today spoke with lung cancer experts to discuss advances in treatment for advanced NSCLC, the most important changes to the treatment paradigm within the past 5 years, and future directions in research for what is now known as a heterogeneous disease.

Histological distinction

The first step after diagnosing a patient with NSCLC is to determine the histology, according to Corey Langer, MD, professor of medicine and director of thoracic oncology at the University of Pennsylvania.

“In the past, we were somewhat cavalier about this,” Langer said in an interview. “If the pathologist said it was NSCLC, they didn’t really test the biopsy for other histologies. Now, we need to know whether the NSCLC is squamous cell carcinoma or non-squamous cell carcinoma. The pathologist has become a major partner in our interdisciplinary care of these patients.”

This is important because the results of two trials in the past 5 years have shown a clear difference in treatment effectiveness based on the different histologies.

For example, the ECOG 4599 trial — published in The New England Journal of Medicine in 2006 — was a phase 3 trial of 898 patients with recurrent or advanced NSCLC who were randomly assigned to paclitaxel (Taxol, Bristol-Myers Squibb) plus carboplatin alone or with bevacizumab. Overall, patients who received bevacizumab plus chemotherapy had a PFS of 12.3 months vs. 10.3 months for patients who received chemotherapy alone. But the treatment benefit was confined to patients with non-squamous cancers, as patients with squamous cell cancers had a higher incidence of bleeding.

“This trial irrevocably changed the therapeutic landscape for advanced NSCLC,” Langer said. “It is the first trial I know of in advanced NSCLC that makes a histologic distinction between patients who would benefit or not benefit from treatment. The trial is also significant because bevacizumab is the first targeted therapy to be approved in combination with chemotherapy for advanced NSCLC.”

In another trial, published in the Journal of Clinical Oncology in 2008, researchers compared cisplatin plus gemcitabine (Gemzar, Lilly) with cisplatin plus pemetrexed (Alimta, Lilly) in chemotherapy-naive patients with advanced NSCLC. They found the OS was similar for both regimens. However, for patients who had adenocarcinoma instead of squamous cell cancer, the OS was superior for those who received the cisplatin plus pemetrexed: 12.6 months, compared with 10.9 months for the cisplatin plus gemcitabine combination.

“This trial is very important because of the inclusion of histology as a pre-specified comparator,” Langer said. “Molecular models had shown that pemetrexed might work better in adenocarcinoma than in squamous cell cancer, and it panned out clinically.”

Importance of EGFR

One of the most significant discoveries in NSCLC is that approximately 10% to 12% of patients with NSCLC harbor an EGFR mutation. This is significant because these patients are eligible to receive an EGFR inhibitor as their single treatment, compared with the chemotherapy regimen that is much more toxic.

In the IPASS study — a phase 3, open-label study conducted in Asia — more than 1,200 previously untreated patients with advanced adenocarcinoma were randomly assigned to receive gefitinib (Iressa, AstraZeneca) or the standard treatment of carboplatin plus paclitaxel.

Edward Kim

The 12-month rate of PFS was 24.9% for patients who received gefitinib and 6.7% for patients who received carboplatin and paclitaxel, demonstrating the noninferiority of gefitinib, and also the superiority of gefitinib in the intent-to-treat population. Perhaps the most significant finding of this study is the fact that, in a subgroup of 261 patients with EGFR mutations, PFS was significantly longer in patients who received gefitinib compared with those who received carboplatin and paclitaxel. In the subgroup of patients without the EGFR mutations, PFS was significantly longer in those who received carboplatin and paclitaxel compared with gefitinib.

This practice-changing study, published in The New England Journal of Medicine in 2009, included patients who fit the bill for EGFR mutations.

“What we know from these years of experience is that patients with EGFR mutations define a distinct subset of patients,” Alice Shaw, MD, PhD, assistant professor of medicine at Harvard Medical School and attending physician in the Thoracic Cancer Program at Massachusetts General Hospital, told HemOnc Today. “They tend to be never-smokers, have adenocarcinoma and are more often women than men. Almost all oncologists are aware of the importance of EGFR mutations and nowadays will screen for these mutations upfront, especially in patients with those particular features.”

Although gefitinib was the EGFR inhibitor used in the IPASS study, these data were extrapolated and applied to erlotinib (Tarceva, Genentech), the EGFR-inhibitor used in the United States. Erlotinib is not indicated for this use, but its use for this setting is recommended in both ASCO and National Comprehensive Cancer Network (NCCN) guidelines, and most insurance companies cover its use for this purpose, according to Kim.

Timing of erlotinib

There is some question as to whether patients should receive front-line erlotinib. To determine whether a patient would benefit from it, tumors must undergo gene sequencing to identify the presence of EGFR mutations. This is not always done before the patient begins therapy.

“Clearly, anyone with an EGFR mutation should receive erlotinib as soon as possible,” Roy S. Herbst, MD, professor of medicine and chief of medical oncology at the Yale Cancer Center, said in an interview. “Some may say that the order in which you give the drugs doesn’t matter. But chemotherapy is more toxic than an EGFR inhibitor, and some patients may not make it through their chemotherapy to receive the erlotinib. Earlier truly is better when it comes to erlotinib.”

Roy S. Herbst

However, in a second-line or third-line setting, erlotinib’s benefit is not necessarily restricted to patients with EGFR mutations. In the INTEREST study published in The Lancet in 2008, researchers compared gefitinib with docetaxel (Taxotere, Sanofi-Aventis) in 1,466 patients with advanced or metastatic NSCLC who previously were treated with platinum-based chemotherapy.

The study found gefitinib was not inferior to docetaxel for OS. More importantly, gefitinib was not superior to docetaxel in patients with EGFR mutations, indicating gefitinib also worked as a second-line or third-line treatment in patients with EGFR wild-type.

“This was an important paradigm because, prior to this data, most people thought that erlotinib should be saved as a last-ditch effort for patients who were really sick,” Kim said. “We’re finding that erlotinib, aside from being better for the population of patients with EGFR mutations, is also effective in the non-mutated population.”

It now is necessary to use genetic analysis to determine which patients without the EGFR mutation also will benefit from erlotinib, Herbst said. In the BATTLE trial, researchers found a gene signature that might identify which patients who are EGFR wild-type might respond to erlotinib. This is being validated now in large data sets.

ALK and crizotinib

The biggest newcomer to affect the treatment of NSCLC is the ALK translocation, Shaw said. ALK, discovered as a new target for NSCLC about 4 years ago, is present in about 6% to 8% of all non-squamous lung cancers.

“This is very exciting because ALK, like EGFR, is a kinase that can be turned off with kinase inhibitors,” Shaw said. “When this discovery was made, there was already an ALK inhibitor in the clinic that hadn’t originally been developed for this purpose. When the first report came out, the inhibitor was already being tested in a phase 1 trial. In the beginning, nobody knew that this drug was going to be very relevant to a subset of lung cancer patients.”

Patients with the ALK translocation often have characteristics similar to those patients with EGFR mutations. They tend to be never-smokers and have adenocarcinoma histology.

In a study published in The New England Journal of Medicine in 2010, researchers analyzed the effect of ALK inhibition in 82 patients who had advanced ALK-positive disease. The patients received 250 mg of the ALK-inhibitor crizotinib (Xalkori, Pfizer) twice daily in 28-day cycles. After a mean treatment duration of 6.4 months, the overall response rate was 57%.

The FDA granted the drug accelerated approval based on the results of two single-arm studies. The first study included 136 patients with locally advanced or metastatic ALK-positive NSCLC, according to the FDA’s summary review document. After a median duration of treatment of 22 weeks, the overall response rate was 50%. The second study included 119 patients with locally advanced or metastatic ALK-positive NSCLC. The median duration of treatment was 32 weeks, and the overall response rate was 61%.

The ALK translocation is present in approximately 6,000 to 11,000 patients with lung cancer per year, according to Kris. That number is high compared with other cancers.

“That’s more than the number of people with testes cancer, and about 11,000 people per year have chronic myelogenous leukemia, which has had staggering success with imatinib,” Kris said. “Is it the number of people with prostate cancer? No. But it’s quite a substantial number, and for those people, it makes a huge difference in their lives.”

There are two active phase 3 trials that are testing crizotinib in patients with ALK-positive NSCLC. In the PROFILE 1007 trial, patients are being randomly assigned to crizotinib or standard-of-care chemotherapy with pemetrexed or docetaxel. The trial will include an estimated 318 patients, and the primary objective is PFS. In the PROFILE 1014 trial, patients are being randomly assigned to crizotinib or chemotherapy with pemetrexed and cisplatin or pemetrexed and carboplatin. This trial will include an estimated 334 patients, and the primary objective also is PFS.

Option for maintenance therapy

Exempt from mutation or translocation status, maintenance therapy is a potential option for all patients with NSCLC, a paradigm that also has shifted within the last 5 years.

“When patients finished their chemotherapy, the traditional approach was to watch and wait,” Langer said. “This is still an option, but there is certainly justification for using a maintenance therapy that could extend survival.”

Among the options for maintenance therapy is erlotinib, which has shown a benefit as a second- or third-line treatment and as a maintenance therapy even among patients without an EGFR mutation. But the big player in maintenance therapy is pemetrexed.

Corey Langer

In a 2009 study published in The Lancet, researchers randomly assigned 663 patients with stage IIIB or stage IV NSCLC to maintenance pemetrexed plus best supportive care or placebo plus best supportive care. All of the patients had received four cycles of platinum-based chemotherapy and had not progressed. Patients who received pemetrexed had a significantly improved PFS of 4.3 months vs. 2.6 months for those who received placebo. OS also was significantly improved for those taking pemetrexed: 13.4 months vs. 10.6 months.

“Maintenance therapy for lung cancer is a terminology that we never really used or taught to our fellows,” Kim said. “Now it is part of the common verbiage. Is it required that patients take maintenance? No. But is it a standard option? Yes.”

Future directions

As exciting as the research findings have been within the past 5 years, there still is a long way to go.

“Patients with EGFR mutations or ALK translocations comprise only 20% of the patient population with NSCLC,” Kim said. “We’ve got another 80% of the population to go.”

According to Kim, one protein that is being explored is Met. Some lung cancers have Met overexpression, which leads to resistance to erlotinib. In a phase 2 study, the drug MetMAb (Roche) was shown to double the PFS when combined with erlotinib, compared with erlotinib alone, in patients with Met overexpression. The addition of MetMAb to erlotinib also tripled the OS in patients with Met overexpression: The OS was 12.6 months for patients who received the combination vs. 3.8 months for patients who received erlotinib alone. Those patients without Met overexpression had worse outcomes when given the combination treatment compared with those who received only erlotinib.

KRAS proteins also are a potential target in NSCLC, according to Langer. Patients who have the KRAS mutation are not likely to have an EGFR mutation or an ALK translocation, making this another potential new group of patients. However, a specific drug to target this mutation has not yet been identified. There is a striking need to help patients with this mutation, as they tend to have tumors that are resistant to most therapies, Kim said.

“Understanding how to target mutant KRAS is going to be key to future progress in this disease,” Herbst said. “As we start to sequence tumors more fully, this is going to result in better understanding of all the different mutations that can occur in these tumors, and how to combine drugs better. It will also help identify early on resistance pathways so that we can combine drugs to prevent resistance. Clearly, it’s important to understand exactly the kind of lung cancer we are dealing with to treat it in the most effective way.” – by Emily Shafer

For more information:

• Ciuleanu T. Lancet. 2009;374:1432-1440.
• FDA. Summary review. 2011. www.accessdata.fda.gov/drugsatfda_docs/nda/2011/202570Orig1s000SumR.pdf. Accessed Oct. 31, 2011.
• Kim ES. Lancet. 2008;372:1809-1818.
• Kwak EL. N Engl J Med. 2010;363:1693-1703.
• Mok TS. N Engl J Med. 2009;361:947-957.
• Sandler A. N Engl J Med. 2006;355:2542-2550.
• Scagliotti GV. J Clin Oncol. 2008;26:3543-3551.

Disclosure: Dr. Herbst reports no relevant financial disclosures. Dr. Kim received research support/honorarium and is a consultant for Lilly, Genentech and ImClone. Dr. Kris is a consultant to Pfizer, Boehringer Ingelheim and Roche China. Dr. Langer serves on the advisory board for Lilly, Genentech, Bristol-Myers Squibb, Pfizer, Bayer and Celgene, and receives research support from GlaxoSmithKline and Genentech. Dr. Shaw is a consultant for Pfizer.

Should genetic testing be mandated for all patients before undergoing treatment for NSCLC?

Thomas Lynch

It should be the standard for all patients.

Molecular testing should be standard for patients with lung cancer. The testing, at minimum, should include testing for EGFR mutation status, ALK translocation status and KRAS mutation status. Treatments are based on this information. There really is no one group of patients with NSCLC that you can avoid testing, because you will find these abnormalities that will guide therapy.

In patients with EGFR mutations, starting with a tyrosine kinase inhibitor (TKI) can improve outcome. If you have an ALK translocation, starting with an ALK inhibitor can improve outcome. It makes such a difference because how you treat patients depends on their molecular profile. A study in The New England Journal of Medicine conducted in Asia showed a clear-cut benefit to giving an EGFR inhibitor first line compared with chemotherapy. There also is a big study in The New England Journal of Medicine that showed giving an ALK inhibitor upfront improves outcomes.

Against activated oncogenic cancers, you want to give a TKI, when it is effective, as early as you can in the course of treatment. Right now, we use them as single agents. Down the road, it may make sense to use them in combination with chemotherapy, an approach that is being studied.

Thomas Lynch, MD, is the Richard Sackler and Jonathan Sackler Professor of Medicine and director of the Yale Cancer Center. References: Mok TS. 2009;361:947-957. Kwak EL. 2010;363:1693-1703. Disclosure: Dr. Lynch reported no relevant financial disclosures.

Howard (Jack) West

Testing should be done on an enriched population with a meaningful probability of a positive test, based on smoking history and tumor histology.

There is a striking disparity between the NCCN guidelines and ASCO guidelines because they’re totally different. The NCCN guidelines recommend EGFR mutation testing upfront in every patient with advanced nonsquamous NSCLC. ASCO guidelines say all mutation testing remains investigational, is of unproven benefit, and is not routinely recommended. Any time there are two disparate recommendations on a complex topic, I think the optimal answer is in between.

There is a clear incentive to identifying an EGFR mutation or ALK translocation. However, there are certain populations who are going to be more or less likely to have a mutation or translocation. The patients who have a minimal or no history of smoking have a significant chance of having an EGFR mutation, and I would recommend screening in these patients. Patients with an adenocarcinoma also have a strikingly higher probability than other histologies.

In the other extreme, if patients have a very significant smoking history, and if they have a squamous or a large-cell carcinoma histology, the probability is not zero but is exceptionally low. In that setting, you cannot make a strong argument for testing for an EGFR mutation to determine timing of erlotinib, which you could give first-, second- or third-line when there is no clear evidence of a survival benefit if the drug is given immediately.

The ALK story is a little more complex. The only way you can receive the therapy is if you test positive for the translocation, and you won’t test positive if you don’t test at all. But if this typically is a $1,500 test, it’s difficult to decide that society should assume the cost of spending $60,000 or more to even identify one patient with a positive result — if the probability is in the 1% to 2% range for a longtime smoking patient with a squamous NSCLC — just to enable the possibility of receiving a non-curative therapy that then will cost nearly $10,000 per month.

At the same time, tissue is of limited availability. Many patients would require an extra biopsy, which is not without risk. It also adds another level of expense — in the range of thousands of dollars — to obtain more tissue.

Society simply doesn’t have infinite resources to do every test, regardless of cost or probability that the result will impact patient care. We can no longer practice in a way in which cost of our management is completely insensitive to magnitude of benefit.

Molecular oncology is the future, and I definitely want to get more tissue as opposed to less any time the patient is diagnosed, but I would not want to be so heavy-handed to say it is mandatory to test everybody.

Howard (Jack) West, MD, is a medical oncologist at the Swedish Cancer Institute in Seattle. Disclosure: Dr. West reports no relevant financial disclosures.