For a drug that became the first of a new generation of lung cancer treatments, Iressa® started out as one of the most puzzling medicines in history.
When given to patients with non-small cell lung cancer – or NSCLC, by far the most common form of the disease – the drug generally didn't provide much benefit, except in a small minority of cases where the results were spectacular. Doctors couldn't predict which patients would be among the lucky few. Some experts argued the drug shouldn't even be approved for lung cancer, since it worked so sporadically, but the successes were too compelling to ignore.
Researchers at Dana-Farber and their colleagues began to piece together evidence earlier this decade that Iressa's rare but dramatic effects were tied to the genetic makeup of NSCLC tumors.
One hint came from the experience of treating patients: Doctors had noticed that Iressa (chemical name: gefitinib) has a higher success rate in Japanese patients than Americans. When Dana-Farber's Matthew Meyerson, MD, PhD, and William Sellers, MD, scanned a collection of NSCLC tumors for malfunctioning, or mutated, genes, they found that the gene EGFR (which stands for "epidermal growth factor receptor") was faulty in 15 Japanese patients, but in only one from the United States. Research had shown that Iressa acts directly against the protein made from EGFR. (Meyerson also holds an appointment at the Broad Institute of MIT and Harvard; Sellers is now based at Novartis, a pharmaceutical company.)
A second hint came from work by Dana-Farber's Bruce Johnson, MD, and Pasi Jänne, MD, PhD. They found that tumor tissue from a woman with cancer that had spread to the lining around her lungs – a condition called adenocarcinoma – was very responsive to Iressa when tested in a laboratory dish. When the adenocarcinoma's DNA was analyzed, it was found to have the same EGFR mutation that Meyerson and Sellers' group had found.
"We knew that EGFR mutations are more frequent in women, in Japanese, and in people with adenocarcinoma," says Johnson, director of Dana-Farber's Lowe Center for Thoracic Oncology. "These are the same groups that are most likely to experience tumor shrinkage when treated with Iressa."
To test whether Iressa indeed works best for patients with EGFR mutations, the researchers analyzed tumor samples from five patients whose tumors had responded to the drug and four whose tumors hadn't. All of the responders had EGFR mutations; the other four did not.
The publication of this work by the Dana-Farber team in 2004 – and the simultaneous publication of a related study by other investigators – propelled lung cancer treatment into the genomic age. "It was the first demonstration that a ‘targeted therapy' [aimed at a particular misbehaving gene or set of genes] could be effective against a form of lung cancer," says Johnson. And it became the model by which other gene-focused treatments for NSCLC could be identified, tested, and approved for patient use.
This summer, Lowe Center researchers and physicians opened a genotyping service where NSCLC patients at Dana-Farber/Brigham and Women's Cancer Center can have their tumors scanned for eight gene abnormalities associated with the disease. The information will help investigators gauge the prevalence of such mutations and the effectiveness of various targeted therapies. The service, established with financial support from the Friends of Dana-Farber Cancer Institute, provides a foundation for the kind of personalized medicine that Bruce Johnson and others envision: a place where lung cancer patients can have their tumors "typed" – much as bacterial infections are categorized by the kind of bacteria involved – and treated accordingly.
The 2004 breakthrough could not have come against a more implacable opponent. Despite declines in cigarette smoking and other forms of tobacco use, lung cancer remains the leading cause of cancer death in men and women in the United States. This year, the disease will claim the lives of 161,840 Americans, according to the American Cancer Society – more than breast, colorectal, and prostate cancers combined. Not only are cure rates for lung cancer lower than for most other major forms of cancer, they have barely budged in recent decades. Lung cancer patients have only a 15 percent chance, on average, of being alive five years after their initial diagnosis.
Against this backdrop, it's important to appreciate the significance, and the limitations, of the advance involving Iressa. (Iressa itself is not currently approved for new patients with lung cancer in the United States; doctors generally prescribe the alternative drug Tarceva®.) Lung cancer is classified as either small cell or non-small cell, accounting for 13 percent and 87 percent of all cases, respectively. Of the non-small cell group, only about 12 percent have an EGFR mutation that makes them candidates for treatment with Tarceva. (Among Japanese and other Asian peoples, the prevalence of EGFR mutations in patients with NSCLC is considerably higher – as much as 40 or 50 percent.) Twelve percent may sound relatively small, but with 187,000 people diagnosed annually in the U.S. with NSCLC, a 12 percent portion works out to 22,440 patients who potentially can benefit from Iressa or Tarceva each year.
Lung cancer remains the leading cause of cancer death in men and women in the United States. This year, the disease will claim the lives of 161,840 Americans.
Also to be considered are the gains in length and quality of life associated with Iressa or Tarceva. For most patients with NSCLC, conventional chemotherapy provides an additional four to six months of survival over what they would have without treatment. Patients who respond to Iressa or Tarceva, by contrast, can expect another year or more. (Like most cancer drugs, Iressa and Tarceva lose their power over time as tumor cells become resistant to them.) And, because they concentrate their effects on cancer cells, rather than a broad range of cells, Iressa and Tarceva generally produce fewer side effects than standard treatments do.
To an outsider, such gains might seem modest, but, as Johnson points out, it represents a doubling of survival time – no mean feat in a field where progress was frustratingly slow for many years.
In some respects, lung cancer is a relative latecomer to targeted therapy. Breast cancer treatment is increasingly personalized, with treatment regimens geared to specific gene mutations in patients' tumors. Lung cancer research, by contrast, has long suffered from the difficulty of obtaining tissue for study and from a reputation as an area with few good scientific leads.
The discovery of NSCLC's molecular vulnerability has given the field fresh energy, by essentially mapping a route toward the development of new gene-oriented therapies. At Dana-Farber, Johnson has assembled a team of scientists to carry the process from start to finish. Meyerson is leading the effort to scan tumor tissue for potential cancer genes; Kwok-kin Wong, MD, PhD, is conducting tests in laboratory cell cultures and mice to determine if abnormalities in those genes are indeed linked to cancer; and Pasi Jänne is studying the effectiveness of therapies for such target genes in patients.
The goal, as Johnson puts it, is to "chip away at the problem of non-small cell lung cancer" by finding and defeating additional genes linked to the disease. Chances are, each of these genes will be responsible for only a small percentage of NSCLC cases – perhaps just 2 percent or 3 percent – but cumulatively they may affect a quarter of all patients with the disease.
"We want to make it routine to test patients' tumors for known mutations, so their treatment can be tailored to their specific abnormalities."– Bruce Johnson, MD
"We want to make it routine to test patients' tumors for known mutations, so their treatment can be tailored to their specific abnormalities," Johnson says. Some of those therapies may already be "on the shelf " – used for other diseases and therefore readily available.
Dana-Farber researchers are currently on the trail of several NSCLC-linked genetic abnormalities – instances of misspellings of a gene's chemical code, or cases where a gene is overcopied or stuck in the wrong portion of a chromosome. In some instances, these same abnormalities are found in other cancers for which drugs are available, raising the possibility that they'll work in certain non-small cell lung tumors as well.
Meyerson is co-leader of a group that has sequenced more than 600 genes in almost 200 lung cancers known as adenocarcinomas, and has discovered more than 15 significant new mutated genes. "In this one survey, we have more than doubled the number of known genes whose mutation can lead to lung cancer," Meyerson says.
Dana-Farber scientists also have been the first to show how NSCLC cells gain the ability to shrug off Iressa or Tarceva and resume their rampant growth. In one study, Jänne and colleagues found that after a year of Iressa or Tarceva use, 50 percent of patients acquire an additional mutation, and 20 percent have an overcopying, or "amplification," of genes, which loosens the drugs' grip on their runaway growth. "We're at work on a second generation of EGFR inhibitors that circumvent the problem of resistance," Jänne says.
"Today, we can identify the genes driving tumor growth in a substantial portion of non-small cell lung cancer patients," observes Wong. "The first challenge is to devise successful treatments keyed to those genes. The other is to find additional genetic culprits." Wong admits that these mutations already linked to the disease may be the easiest to find – "low hanging fruit," as they're called – and that additional ones may be more elusive. But, he adds, "There is a lot of low-hanging fruit."
Fall/Winter 2008 Table of Contents
Dana-Farber Cancer Institute, 450 Brookline Avenue, Boston, MA 02215 | Call us toll-free: