Basket Cases: Changing the Way Cancer Is Treated

CURELung Special Issue
Volume 1
Issue 1

When determining treatment plans, some doctors are turning to genetic mutations that caused the cancer instead of the cancer type.

Cancer patients have traditionally been treated based on where their tumors originate — lung tumors with treatments for lung cancer, breast tumors with therapies developed specifically for breast cancer. But what if doctors started caring less about where a tumor first emerged and more about the genetic mutations that caused it to appear in the first place? That principle is at the essence of precision medicine, a science focused on understanding each particular tumor and the molecular messages it sends, and treating it accordingly.

Ultimately, this branch of investigation could mean that a breast and a prostate cancer are treated with the same medicine, or that two people with melanoma of the same stage and location are given entirely different therapies. It's something that's already happening more and more in the world of cancer treatment, and it's given rise to a host of clinical trials with a new type of design.

Known as "basket trials" or "umbrella trials," these studies enroll patients with a variety of cancer types, find the specific mutations or abnormalities that drive their tumors, and then match the patients with medications that target those glitches. It doesn't matter where a cancer first emerged; tumor genetics are what determine trial eligibility. In basket trials, patients are eligible if their tumor harbors a specific genetic mutation that is a target for the therapy in question. A basket trial could enroll patients with several different types of tumors (skin, lung or colon cancer), but all of the tumors would share the same genetic abnormality. Alternatively, an umbrella trial might enroll patients with the same underlying origin of cancer (such as breast or pancreatic), but allocate patients to treatment with one of several different targeted therapies.

An example of a large umbrella trial is ALCHEMIST, or the Adjuvant Lung Cancer Enrichment Marker Identification and Sequencing Trial, a government-sponsored study that is enrolling patients with early-stage, nonsquamous non-small cell lung cancer (NSCLC) after surgery. The trial will screen patients for EGFR mutations and ALK rearrangements, and then place them in two separate trials based on those biomarkers, evaluating treatment with either EGFR-targetingTarceva (erlotinib) or ALK-targeting Xalkori (crizotinib) versus placebo. Other large trials, such as NCI-MATCH and TAPUR, are enrolling patients with numerous cancer types and looking for a variety of genetic abnormalities to treat with targeted drugs, and so combine both approaches.

Many clinical trials of cancer drugs include some degree of personalization these days, but basket and umbrella trials remain a minority among studies because the level of collaboration they involve can make them difficult to organize. The costs of biopsies and genomic sequencing are also challenges. Still, experts say, we can expect to see more and more of these trials emerging over time.


In February of 2001, The Human Genome Project announced that a sequencing of the human genome had been completed. That project identified the locations of, and provided information about, the structure and organization of specific genes. It was the breakthrough that ushered in a new era of oncology research that has allowed investigators to study the genetic makeup of cancer tumors and reclassify them according to their composition.

A decade-and-a-half later, in his January 2015 State of the Union address, President Barack Obama recognized the significance of this line of study, and its evolution into an actionable body of work, by announcing the launch of the Precision Medicine Initiative, calling for $215 million dollars in government spending for fiscal year 2016. Seventy million dollars of that initial allocation went to the National Cancer Institute (NCI) for research in targeted, precision oncology.

Indeed, within the last decade, a significant focus within cancer research has been the push to understand how tumor genetics drive different cancers, and how to develop more focused drugs to target these molecular abnormalities. "The last decade has seen the development of several targeted therapies that can be very effective for a subset of patients with certain types of cancers, including lung cancer or melanoma. The challenges now are to try and extend these developments to other patients and to develop strategies to treat resistant tumors," says Amanda Redig, a medical oncologist at Dana-Farber Cancer Institute.

The gene mutations that occur in cancer are numerous. Primo Lara, a clinician-scientist at UC Davis Comprehensive Cancer Center, uses the commonly mutated gene KRAS as an example. It appears in 25 percent of lung cancers, 90 percent of pancreatic cancers and about half of all colon cancers. Further, approximately one-half of all cancers have a mutation in the tumor suppressor gene p53; this mutation releases the brakes on tumor growth, allowing unfettered cancer cell proliferation. Efforts are underway to identify and develop drugs that specifically target KRAS and p53, and to identify and clinically validate many more cancer-related mutations that could be clinically actionable.

Among approved drugs are those that target EGFR, the epidermal growth factor receptor. Mutations to EGFR are found in 10 to 15 percent of lung cancers in the U.S. and in more than 50 percent of lung cancers in Asian populations, more prevalently in females than in males, according to Shakun Malik, head of thoracic and head and neck cancer therapeutics at NCI’s Cancer Therapy Evaluation Program.

Mutations to a different gene, BRAF, were at the heart of the first published results of a basket study — the Zelboraf (vemurafenib) trial — reported just last year by researchers at Memorial Sloan Kettering Cancer Center (MSKCC) in New York. BRAF inhibitor Zelboraf was approved by the U.S. Food and Drug Administration (FDA) in August 2011 for the treatment of metastatic melanoma, but since BRAF mutations occur across a variety of cancers, investigators wanted to test the drug's effect on other cancer types, including lung, colorectal and ovarian. They enrolled 122 patients from centers around the world, all of whose cancers carried the BRAF mutation.

In general, results showed that Zelboraf was able to target this mutation across a wide variety of cancers, proving that cancer type does not always matter.

Medical oncologist David Hyman, acting director of developmental therapeutics at MSKCC and the study’s first author, says an advantage of basket studies is that they can democratize access to targeted therapy for those who have rare cancers, who are often underrepresented in clinical trials. "The discovery has been that some patients with very rare cancers have similar mutations (as more common cancers) at even higher rates, and some have no ongoing trials or therapies available (specifically for their disease types)," he says.

A number of large precision medicine trials are underway now, with the eyes of the scientific and patient communities watching closely to see where this kind of study will take us.


Throughout the country, lung cancer is the No. 1 cancer killer, causing over 27 percent of cancer deaths, or 150,000, each year. But because squamous cell lung cancers that are often caused by smoking include a wide variety of cell types and can be deadly, progress in finding treatments has been slow.

Mary Redman, the architect of the statistical design of the Lung-MAP trial and a biostatistician at Fred Hutchinson Cancer Center in Seattle, says the hope is that by using targeted therapy — both single-agent and combination regimens — treatments will be found that are effective and can prolong lives. Lung-MAP is a public-private collaboration composed of the NCI, SWOG Cancer Research (a worldwide network of researchers that designs and conducts clinical trials), Friends of Cancer Research, the Foundation for the National Institutes of Health, five pharmaceutical companies, Foundation Medicine and several lung cancer advocacy organizations. It's a large trial for individuals with advanced squamous NSCLCs who have undergone no more than one round of chemotherapy treatment and are not responding to that treatment.

Lung-MAP's co-principal investigator and co-chair of its trial oversight and drug selection committees, Roy S. Herbst, says patients can join the trial at any of more than 700 medical centers, where their cancer cells will be tested for over 200 possible gene alterations. Based on the results of this screening, eligible patients are assigned to one of four sub-studies testing investigational or FDA-approved treatments, based on which best suits their cancer's genomic profile. Three arms are testing targeted drugs, and the fourth is testing immunotherapeutic treatments that do not target a specific mutation, possibly benefiting patients whose cancers lack genetic alterations that match any of the targeted investigational therapies. Specifically, the immunotherapy arm is testing the combination of Opdivo (nivolumab) and Yervoy (ipilimumab) compared with Opdivo alone.

Unlike most trials that require the development of a protocol for each drug to be tested, Lung-MAP uses a "master protocol" that can be changed and adapted quickly as new drugs become available or drugs that are proven not to be useful are removed from the study, says Herbst, professor of medical oncology at Yale Cancer Center-Smilow Cancer Hospital. Furthermore, the study design is unusual in that, if it demonstrates that a drug is safe and effective, that agent will automatically be eligible for approval by the FDA.

Jerry Valentino was diagnosed two years ago with squamous NSCLC at the Veterans Administration in West Haven, Connecticut, a teaching affiliate of Yale Medical School. The 69-year-old underwent 35 radiation treatments in addition to chemotherapy, resulting in the disappearance of the tumor in his lung. But in September 2015, he thought he was having a stroke. An MRI showed that his lung cancer had traveled to his brain.

After receiving brain radiation, with few standard treatment options available, Valentino instead entered the Lung-MAP trial at the VA and submitted his tissue for genome profiling. Based on the results, he was assigned to a sub-study and received an oral CDK4/CDK6 inhibitor called Ibrance (palbociblib), which was approved by the FDA in 2015 for the treatment of breast cancer. During each of three cycles, he took one pill a day for 21 days, followed by a one-week break. Valentino has undergone restaging scans as part of Lung-MAP's surveillance regimen, which confirmed stability in two lung lesions and partial response in another.

"I have experienced no side effects except fatigue, my brain tumor has shrunk and there is no lung nodule visible now," he says, adding that he is glad to be getting treatment that doesn't make him feel sick all the time.


The NCI’s Molecular Analysis for Therapy Choice (NCI-MATCH) trial, which opened for enrollment in August 2015, explores more genetic abnormalities and drugs than any other cancer trial — and in the largest number of cancer types ever, according to lead investigator Keith Flaherty.

Funded entirely by the NCI and with investigational drugs provided by pharmaceutical corporations, the trial was co-developed by the NCI and the ECOG-ACRIN Cancer Research Group. Its aim is to match tumor signals with targeted drugs and drug combinations, leading to further study of the strategies that prove effective.

Flaherty explains that the study arms, which will grow from 10 to 20 as the trial progresses, are for adults with advanced solid tumors and lymphomas that no longer respond to, or have never responded to, standard therapy, and for those with rare cancers for which there is no standard treatment. While patients with all kinds of cancers can enroll for screening, the goal is that at least 25 percent of enrollees will have cancers that are considered rare.

The trial is looking at 143 genes associated with cancer using cutting-edge molecular analysis. Tumor biopsies will be taken from about 3,000 patients. Those biopsies are sent to any of four labs, where the DNA is sequenced and analyzed in other ways to identify genetic abnormalities believed to be driving the growth of cancer. Drugs and drug combinations being tested have been approved by the FDA for specific cancers or have been used in other trials and have shown some effectiveness against certain genetic alterations.

When and if genetic abnormalities are found in participants, such as mutations in BRAF or EGFR, or HER2 amplification, these patients will be further evaluated for eligibility to enroll in the corresponding treatment arm; if they qualify, they will be treated for as long as their tumors shrink or remain stable.

So far, of 645 biopsies successfully evaluated, 9 percent (56 patients) had a gene mutation that matched a drug or drug combination being studied, 5.1 percent (33 patients) were assigned to a treatment arm and 16 patients actually entered into seven of the 10 available arms. Ultimately, investigators want to treat 1,000 patients across all arms of the trial. For an approach to be considered promising within any arm of the study, at least 16 percent of patients have to show tumor shrinkage. If a first drug treatment is unsuccessful, a patient may be eligible to receive therapy in another arm of the trial if there is a remaining drug that matches one of the mutations found in his or her tumor. For each arm of the trial, the major endpoint will be the overall response to treatment, with a second endpoint being six months of progression-free survival, time to disease progression, toxicity and biomarker status.

The trial was designed, says Flaherty, so that when 500 patients were accepted as eligible to submit tissue for mutation screening, the study would be paused for analysis to determine how it should proceed, with patients already being treated continuing to receive their medications. During the current pause that began in November 2015, the diversity of cancer types is being studied, data for those with mutations that correlate with treatments in the trial are being analyzed, and an assessment of changes needed to move forward is being worked out. Facilitating a steady flow of tissue screening through the participating labs is one task being addressed during the pause, but a more vital one has involved adding more study arms, which will lead to the treatment of more patients, leaders say.

The interim analysis will be completed soon, and enrollment in the trial is expected to resume in late May. "It was assumed years ago that cancers that harbored the same genetic mutation, for which we have targeted therapies in development, would be equally likely to respond," says Flaherty. "We have learned recently that there is a lot of variability based on the presence of co-occurring mutations or biology that stems from the tissue of origin for a given cancer. The MATCH trial is the largest endeavor undertaken to explore the consistency or variability of response to targeted therapy across the entire cancer population."


The first trial ever to be run by the American Society of Clinical Oncology (ASCO), the Targeted Agent and Profiling Utilization Registry (TAPUR) is being launched at the sites of the Michigan Cancer Research Consortium, the Cancer Research Consortium of West Michigan and the Levine Cancer Institute at Carolinas HealthCare System. Its aim is to simplify access to cancer treatments for patients with various tumor types that carry genetic variations known to be the targets of any existing approved drugs — including lung and other kinds of advanced solid tumors, multiple myeloma or non-Hodgkin lymphoma. Patients who have at least one genomic variant that can potentially be targeted by one or more available drugs and are not responding to standard therapies, or for whom no treatment is available, are eligible for enrollment in TAPUR.

In addition to providing drugs to which patients might not otherwise have had access, the objectives of TAPUR are to assess safety and efficacy and to target multiple tumor types. To begin with, the study will enroll patients at three sites nationally to evaluate 10 to 15 drugs, with up to 28 patients per drug treatment.

Unlike Lung-MAP, which is studying both investigational and approved drugs, TAPUR will look strictly at drugs that have been OK'd by the FDA. The study will include more genomic aberrations and more drugs than Lung-MAP, which exclusively targets squamous NSCLC. The trial has more in common with NCI-MATCH, because, like MATCH, it targets rare genetic aberrations with drugs designed to quiet them. However, it is simpler than MATCH, as there is not a different protocol for each drug being tested.


As those trials across a spectrum of cancers and mutations continue, others are popping up that consider more specific questions.

Protocol RXDX-101-02 (also known as the STARTRK-2 study) is a phase 2 basket clinical trial that is open to enrollment in the U.S. for patients with solid tumors. Eligible patients will be treated with entrectinib, an investigational drug that specifically targets certain cellular pathways that are involved in the development and progression of cancer. Recently launched by the drug’s developer, Ignyta, Inc., the clinical trial will seek to enroll patients with locally advanced or metastatic solid tumors that have a specific mutation (called a “gene rearrangement” or “fusion”) that is found in the NTRK1, NTRK2, NTRK3, ROS1 or ALK genes. It is believed that these types of mutations have the potential to be oncogenic drivers. Ignyta and other labs offer a test to help determine if one of these mutations is detected within a sample of tumor tissue. The trial will include a variety of cancer types (i.e., different baskets), among them non-small cell lung cancer, colorectal cancer, salivary gland cancer, papillary thyroid cancer, melanoma and sarcoma. Depending on cancer type and gene rearrangement/fusion, eligible patients will be assigned to a specific basket for treatment with entrectinib.

"What we are seeing is that, regardless of the originating tumor type, these oncogene-driven tumors behave similarly in their response to targeted drugs," says Robert C. Doebele, an investigator for the entrectinib trial at the University of Colorado School of Medicine, where he is an associate professor in the Division of Medical Oncology. "This seems to be a new paradigm in study design."