Featured Article

FDA Approves New Drugs to Treat Skin, Blood, and Lung Cancers

The FDA recently approved three new cancer drugs, including one to treat non-small cell lung cancer.

In the last 2 weeks of August, the Food and Drug Administration (FDA) approved three new cancer drugs: vemurafenib ²² (Zelboraf), for patients with unresectable ²³ or metastatic melanoma ²⁴ whose tumors harbor a specific genetic mutation in the BRAF gene; brentuximab vedotin ²⁵ (Adcetris), for some patients with Hodgkin lymphoma ²⁶ and anaplastic large cell lymphoma ²⁷; and crizotinib ²⁸ (Xalkori), for patients with locally advanced or metastatic non-small cell lung cancer ²⁹ (NSCLC) whose tumors have a gene fusion ³⁰ caused by a chromosomal translocation ³¹ involving the ALK gene.

The three drugs were approved all well in advance of the time allotted for their review. By comparison, from the beginning of 2010 until the week of August 15, 2011, the agency had approved just six new cancer drugs.

Some common themes ran through the approvals. Crizotinib and brentuximab, for example, were accelerated approvals, meaning that they were approved because clinical trials showed an improvement in a surrogate endpoint, such as tumor response, that is reasonably likely to predict clinical benefit. Both drugs have to be tested in confirmatory trials to verify their safety and clinical effectiveness.

Crizotinib and vemurafenib, meanwhile, were approved with companion diagnostics ³⁸ for each drug. These companion tests are used to identify patients who are most likely to benefit from the drugs, based on the presence of a specific genetic abnormality or other molecular marker ³⁹.

All three drugs also have one more important feature in common, stressed Dr. Richard Pazdur, director of the Office of Oncology Drug Products in FDA's Center for Drug Evaluation and Research: "They all appear to have a more favorable benefit-to-risk analysis than conventional chemotherapy agents." The greater benefit-to-risk ratio "reflects a true understanding of the biology of the disease," Dr. Pazdur continued.

Although targeting specific genetic aberrations is not a new therapeutic approach, greater knowledge of the underlying biology of many cancers means that "there are more targets out there, and researchers better understand how to hit [those targets] and the ramifications of hitting them," said Dr. Jeff Allen, executive director of Friends of Cancer Research, an advocacy organization that promotes research collaborations. "To see that coming to clinical benefit with greater frequency is very encouraging."

In addition, all three drugs have demonstrated or strongly suggested that they are effective for diseases for which effective treatments have been difficult to find. Brentuximab, for instance, is the first new drug approved for Hodgkin lymphoma in 30 years.

A Refined Approach

Dr. Paul Bunn, a lung cancer expert from the University of Colorado Medical Center, said during a press briefing on crizotinib that the drug's approval represents "a new paradigm for drug development, where a small but well-defined fraction of people get a very well-defined drug."

The same holds true for vemurafenib. The FDA based its approval for both drugs on trials that included only patients who harbored the molecular aberrations that the drugs target.

In the case of crizotinib, only 3 to 5 percent of patients with NSCLC harbor the ALK translocation ³¹. (A recent study ⁴⁵ suggests, however, that as many as 8 percent of patients with the adenocarcinoma ⁴⁶ type of NSCLC may have the translocation.) Even though the percentage of patients with the translocation is small, as many as 16,000 NSCLC patients per year may be candidates for the drug, said Dr. Mark Kris of Memorial Sloan-Kettering Cancer Center during the briefing.

The subset of melanoma patients who are candidates for vemurafenib is substantially larger; about half of patients with metastatic melanoma harbor the BRAF mutation.

Such a sizable subset of patients who are candidates for a new molecularly tailored therapy will likely be more of an exception than the rule, said Dr. Razelle Kurzrock, chair of the Department of Investigational Cancer Therapeutics at the University of Texas M. D. Anderson Cancer Center.

For common cancers like lung cancer, Dr. Kurzrock added, it is particularly unlikely that a large percentage of patients will be found to carry a specific genetic alteration. "Common cancers may be more common because there are more pathways for developing the disease than in less common and rarer cancers," she said.

Had crizotinib been developed in the more traditional manner, Dr. Kurzrock said, with the trials including any patient with NSCLC rather than just those with the ALK translocation, the drug could very well have been abandoned because of poor response rates ⁴⁷.

Essential Companions

When there is a marker that clearly identifies a subset of patients who may benefit from a drug, "companies have become very proactive in the development of companion diagnostics," said Dr. Helen Chen of NCI's Division of Cancer Treatment and Diagnosis ⁴⁸.

From fairly early in the development process of vemurafenib and crizotinib, in fact, the drugs' developers, Pfizer and Plexxikon, began working with diagnostics companies to develop tests that would identify appropriate patients to include in clinical trials.

"There has been a great deal of concern in the oncology community about the difficulty of developing diagnostics along with the drug," said Dr. Pazdur. "These are both excellent examples that it can be done…. You can develop your in vitro ⁴⁹ diagnostic [test] with the drug simultaneously and move through the approval process fairly easily."

Go Forth and Multiply

The rapid development and approval of these drugs can have a multiplier effect. Crizotinib's approval is particularly important, Dr. Kurzrock believes. "It provides a real incentive to look for these small subsets of patients, because now it's been shown that it can really pay off."

It can also accelerate the continued development of the approved agents. Brentuximab is already being tested in patients with earlier-stage Hodgkin lymphoma, as well as in CD30-positive non-Hodgkin lymphoma ⁵⁰. And vemurafenib will be tested in trials in combination with a recently approved drug for advanced melanoma, the immunotherapy drug ipilimumab ⁵¹ (Yervoy).

With numerous patients experiencing complete disappearance of their tumors, at least for a time, in the clinical trials that led to these new approvals, all three drugs are already helping patients.

Jeff Wigbels took his first dose of crizotinib last October. His lung cancer had spread to multiple places in his body, including his throat. He had to eat through a tube, he explained during the briefing on the drug's approval.

A week after taking the first dose, some friends came to visit him. They ordered pizza. And he could eat it.

"It was an amazing experience for me that [the drug] could work that quickly," he said.

—Carmen Phillips ⁵²

Cancer Research Highlights

Negative Colonoscopy Associated with Low Colorectal Cancer Risk

People who have no sign of cancer after a colonoscopy ⁵⁷ have a much lower risk of developing colorectal cancer ⁵⁸ over the next two decades than people who never had the procedure, according to a large case-control ⁵⁹ study conducted in Germany. Overall, people who had a negative colonoscopy were nearly 80 percent less likely to develop colorectal cancer than those who had never had a colonoscopy. The study ⁶⁰ was published online August 29 in the Journal of Clinical Oncology.

“Clearly, colonoscopy is not a preventive measure by itself. The strongly reduced risk among patients who underwent colonoscopy without polypectomy ⁶¹ is therefore not a result of colonoscopy, but rather reflects the inherently low risk of patients free of endoscopically ⁶² visible precancerous lesions,” wrote the study’s lead author, Dr. Hermann Brenner of the German Cancer Research Center in Heidelberg and his colleagues.

In the United States, guidelines for colorectal cancer screening with colonoscopy generally recommend screening every 10 years beginning at age 50. More vigilant screening, however, is typically recommended for those with a family history of colorectal polyps ⁶³ or cancer.

The study included nearly 3,600 participants. Case patients were 30 years of age or older who had been diagnosed with colorectal cancer based on symptoms or incidental findings from other procedures but not as a result of screening and were treated at 22 hospitals in southeastern Germany. Control subjects were selected from population registries. Information on previous colorectal endoscopies was obtained from interviews and medical record reviews.

Control subjects were nearly four times more likely than case patients to have had a previous negative colonoscopy. Overall, previous negative colonoscopy was associated with a 79 percent reduction in the risk of colorectal cancer. A single negative colonoscopy was associated with a 72 percent lower risk of developing colorectal cancer over the next 10 to 19 years, and a 60 percent lower risk at 20 years or longer. The lower risks were similar for men and women, former and never smokers, participants who had first-degree relatives ⁶⁴ previously diagnosed with colorectal cancer, and participants with a single previous colonoscopy.

All participants in the study were recruited between 2003 and 2007. Because colonoscopy was not routinely offered as a primary screening test for colorectal cancer in Germany until late 2002, many of the participants in the study who had undergone a colonoscopy did so because of positive findings from a fecal occult blood test ⁶⁵ that was prompted by symptoms or other indications, the authors explained.

This fact could limit the extent to which the results apply to a primary screening setting, the study authors noted. They suggested, however, that the long-term risk of colorectal cancer after a negative colonoscopy might be even lower in an unselected screening population. Indeed, although they note that the results should be confirmed in additional studies, the authors also point out that their study “supports suggestions that the majority of average-risk patients with a negative colonoscopy might not need another screening colonoscopy for at least 20 years, if at all.”

Further reading: “Sigmoidoscopy Markedly Reduces Colorectal Cancer Incidence, Mortality ⁶⁶”

Smokers at Greater Risk of Bladder Cancer than Previously Estimated

Current cigarette smokers have a higher risk of bladder cancer ⁶⁷ than previously reported, and the proportion of bladder cancers due to smoking in women is now comparable to that in men, according to a study by researchers in NCI’s Division of Cancer Epidemiology and Genetics ⁶⁸ (DCEG). Their findings ⁶⁹ were published August 17 in JAMA.

This latest study included data from more than 450,000 participants in the NIH-AARP Diet and Health Study ⁷⁰, a questionnaire-based study that began in 1995, with follow-up through the end of 2006.

“Current smokers in our study had a fourfold excess risk of developing bladder cancer, compared to a threefold excess risk observed in previous studies,” said study co-author Dr. Neal Freedman. “The stronger association between smoking and bladder cancer is possibly due to changes in cigarette composition or smoking habits over the years.”

In the current study, former smokers were twice as likely to develop bladder cancer than those who never smoked. As with many other smoking-related cancers, the risk of bladder cancer fell after people quit smoking.

Previous studies had indicated that only 20 to 30 percent of bladder cancer cases in women were caused by smoking, but the new data indicate that smoking is responsible for about half of bladder cancer cases among women, a proportion similar to that found in men in this and previous studies. 

The increase in the proportion of smoking-induced bladder cancer cases among women may be because smoking rates in men and women are now similar. The majority of the earlier studies were conducted at times or in places where smoking was much less common among women than men. 

“Our findings provide additional evidence of the importance of preventing smoking initiation and promoting cessation for both men and women,” said senior author Dr. Christian Abnet.

Hereditary Mutations in BAP1 Gene Raise Risk of Mesothelioma

In two independent studies, researchers have shown that people with hereditary mutations in a gene known as BAP1 are predisposed to develop malignant mesothelioma ⁷¹ and melanoma ²⁴ of the eye (uveal or intraocular melanoma), and a distinctive type of benign ⁷² skin tumor. The findings, published online August 28 in Nature Genetics, suggest that inherited BAP1 mutations may also be linked to some other forms of cancer, including melanoma of the skin.

In one study ⁷³, supported by NCI, a team of investigators led by Dr. Joseph Testa of the Fox Chase Cancer Center and Dr. Michele Carbone of the University of Hawaii Cancer Center focused on two U.S. families with a high incidence of mesothelioma. The study is the first to show that inherited gene mutations ⁷⁴ can influence a person’s risk of mesothelioma—one of the least curable forms of cancer. Mesothelioma is typically associated with exposure to asbestos ⁷⁵ or to a similar mineral fiber, erionite.

The second study ⁷⁶, led by Drs. Thomas Wiesner of the Medical University of Graz, Austria, and Memorial Sloan-Kettering Cancer Center (MSKCC); Boris Bastian of MSKCC; and Michael Speicher, also of the Medical University of Graz, focused on two families, one from Germany and one from Austria, in which members developed numerous small, benign skin growths starting at an early age. These raised growths occurred in pigment-producing cells called melanocytes ⁷⁷ but were skin-colored, unlike typical moles ⁷⁸.

In both studies, researchers zeroed in on the BAP1 gene after finding genetic changes in or near the region of human chromosome 3 where BAP1 is located. The BAP1 gene encodes a protein known as BRCA1 ⁷⁹-associated protein-1, which is found in the cell nucleus and is thought to suppress tumors. The BAP1 protein has been implicated in a range of cellular processes, including regulation of cell growth and division and response to DNA damage.

In the two families with mesothelioma, Drs. Testa, Carbone, and their colleagues found BAP1 mutations in two individuals with uveal melanoma, one of whom subsequently developed mesothelioma. The research team also found hereditary alterations in BAP1 in 2 of 26 patients with sporadic ⁸⁰ mesothelioma. Both individuals had previously been diagnosed with uveal melanoma, although none of the other 24 patients had. Some of the patients with sporadic mesothelioma were found to have noninherited BAP1 alterations in their tumors.

Their findings, Drs. Testa and Carbone wrote, “suggest that individuals with uveal melanoma who carry [hereditary] BAP1 mutations are at high risk of developing mesothelioma and should be closely monitored."

In each family that Dr. Wiesner and his colleagues studied, one individual with benign skin tumors also had melanoma of the eye. In addition, three members of one family were diagnosed with skin melanoma. These investigators also found noninherited BAP1 mutations in randomly selected tumors from an independent group of patients with melanoma of the eye and skin.

In two other recent studies (here ⁸¹ and here ⁸²), noninherited mutations in BAP1 were found in tumor tissue of sporadic cases of mesothelioma and melanoma of the eye.

Use of Radioactive Iodine to Treat Thyroid Cancer on the Rise

The proportion of patients in the United States with well-differentiated thyroid cancer ⁸³ (papillary ⁸⁴, follicular ⁸⁵, or Hurthle cell ⁸⁶ tumors) who received radioactive iodine ⁸⁷ after surgery rose from 40 percent to 56 percent between 1990 and 2008, and this increase was seen across all tumor sizes. Dr. Megan R. Haymart of the University of Michigan and her colleagues reported ⁸⁸ their findings August 17 in JAMA.  

This increase came despite the low risk of recurrence ⁸⁹ for these patients after total thyroidectomy ⁹⁰ alone and conflicting evidence about whether radioactive iodine treatment benefits patients with low-risk disease. This increase in treatment may be exposing patients to unnecessary harms, as well as incurring greater costs.

The researchers used data from 189,219 thyroid cancer patients in the National Cancer Database, which is maintained by the American College of Surgeons Commission on Cancer and the American Cancer Society and captures about 85 percent of patients with thyroid cancer in the United States.

Patients who were younger and healthier were more likely to receive radioactive iodine, but African American patients and those without health insurance were less likely to receive radioactive iodine. In an analysis of radioactive iodine use by stage ⁹¹ of disease in patients treated in 2004 to 2008, patients with the earliest-stage tumors (stage I) were less likely than those with stage IV disease to receive radioactive iodine. However, there was no difference in use between patients with stage II or stage IV disease or between patients with stage III or stage IV disease.

Hospitals varied widely in their use of radioactive iodine for patients with all stages of the disease, as well. Using data from 2004 to 2008, the authors also found that patients treated at a hospital with a higher case volume of thyroid cancer patients were more likely to be treated with radioactive iodine. “For every 1 additional case a hospital treats, the odds of radioactive iodine use increase by 0.6 percent,” explained the authors.

Characteristics of the patients and their tumors accounted for just 21 percent of the variation among hospitals, and unknown hospital factors accounted for 29 percent of the variation, suggesting that “disease severity is not the sole determinant of radioactive iodine use,” they noted.

The data set used provides “incomplete knowledge about how and why care was delivered in hospitals showing variation,” wrote Dr. Edward H. Livingston of the University of Texas Southwestern Medical Center and Dr. Robert A. McNutt of the Rush University School of Medicine in an accompanying editorial ⁹². “Without an assessment of hard, irrefutable measures of clinical decision making that include individual preferences for treatment, decisions about the appropriateness of clinical treatments and variations of care cannot be made.”

According to data from NCI’s Surveillance, Epidemiology, and End Results ⁹³ program, the incidence of well-differentiated thyroid cancer has more than doubled ⁹⁴ in the last 30 years. “This study highlights the need to develop better prognostic models for these patients, since it’s unlikely that a randomized trial ⁹⁵ [of radioactive iodine] will ever be done,” due to the very large number of patients and long follow-up that would be required, commented Dr. Ann W. Gramza, a physician with NCI’s Thyroid Clinical Research Program at the NIH Clinical Center.

Colorectal Cancer Genomes Sequenced

Researchers have used whole-genome sequencing to survey the genetic alterations in colorectal tumors and matched normal cells from nine patients with the disease. Among other findings, the researchers identified a fused gene ³⁰ in a small percentage of tumors. As with all such genome projects, the results ⁹⁶, published online September 4 in Nature Genetics, are preliminary, and further studies will be needed to capture the extent of genomic changes in this disease.

Dr. Matthew Meyerson of the Dana-Farber Cancer Institute and his colleagues found a range of genomic rearrangements, including exchanges of DNA between and within chromosomes (known as translocations ³¹). The researchers identified 11 translocations that could give rise to fusion genes, which are created when DNA from different parts of the genome merges. Specific fusion genes, and the fusion proteins they encode, have been reported in common cancers ⁹⁷ such as lung and prostate, but little is known about their possible role in colorectal cancer.

One of the 11 gene fusions was of particular interest. The fusion involves the genes VTI1A and TCF7L2, which encodes a transcription ⁹⁸ factor that regulates the activity of genes that are essential for the proliferation and development of intestinal epithelial ⁹⁹ cells. In addition, previous research ¹⁰⁰ has linked the expression of TCF7L2 to survival in colorectal cancer. Among a set of 97 colorectal tumors the researchers tested, three had this fusion. 

To test the functional importance of the VTI1A-TCF7L2 fusion gene, the researchers identified a colorectal cancer cell line that included the fusion. Experiments with these cells suggested that the fusion gene plays a “critical role” in their growth, the researchers found. But they noted that further research is needed to understand what role, if any, the gene may play in the disease.

The discovery of the VTI1A-TCF7L2 fusion gene shows that “functionally important fusion events occur in this disease and suggest that further structural characteri­zation will likely identify additional new recurrent rearrangements,” Dr. Meyerson and his colleagues concluded.  

They noted that large sequencing projects will be needed to catalog the genomic changes in the disease; one such project is The Cancer Genome Atlas ¹⁰¹.

Further reading: "More DNA Rearrangements Found in Prostate Cancers ¹⁰²"

Inside NCI: A Conversation with Dr. Norm Coleman about Radiation Research

The associate director of NCI's Radiation Research Program discusses his research and specialized initiatives of the Radiation Research Program.

Special Report

Clinical Trial Shows Potential Benefit of HPV Vaccine for Anal Cancer

A vaccine used to prevent viral infections that cause cervical cancer also prevents similar infections of the anus that could eventually lead to anal cancer ¹⁰³. The vaccine, Cervarix ¹⁰⁴, protects women against infection with the two types of human papillomavirus ¹⁰⁵ (HPV) that cause the majority of anal  and cervical cancers—types 16 and 18.

Women in the Costa Rica HPV Vaccine Trial who received the Cervarix vaccine had far lower rates of anal HPV infection than women in the control group.

The study was the first to assess the effectiveness of an HPV vaccine against anal HPV infection in women. Dr. Aimée Kreimer of NCI’s Division of Cancer Epidemiology and Genetics ⁶⁸ (DCEG) and her colleagues reported their findings ¹⁰⁶ online August 22 in Lancet Oncology.

“This study gives us our first indication that Cervarix provides excellent protection against two HPV types that cause anal cancer and also possibly other [HPV] types,” said Dr. Diane Harper of the University of Missouri-Kansas City School of Medicine, who was involved in clinical trials for both Cervarix and the second HPV vaccine, Gardasil ¹⁰⁷.

Anal cancer is rare, but the incidence of the disease has been increasing in the United States and several European countries. Women are more likely than men to develop anal cancer; most affected individuals are diagnosed in their 60s.

To assess the vaccine’s ability to prevent anal HPV infections that could lead to cancer, the researchers tested samples from women participating in the Costa Rica HPV Vaccine Trial, a clinical trial of Cervarix to prevent cervical HPV infection, who agreed to have an anal specimen collected 4 years after vaccination. (All participants were screened for cervical HPV infections prior to vaccination and 4 years later.)

Women who did not show evidence of HPV in the cervix at the start of the trial were considered unlikely to have had anal HPV infections as well. Four years after vaccination, the rate of anal HPV infections was nearly 84 percent lower in women who received Cervarix than in women in the control group ¹⁰⁸.

The level of protection against anal cancer was similar to that observed against cervical HPV infections in the trial. HPV vaccination for the prevention of cervical cancer may have the additional benefit of reducing anal HPV infections that could lead to cancer, noted Dr. Kreimer.

“Since our data show a reduction of anal HPV infection rates in vaccinated women, it suggests that, in the future, women who receive the prophylactic HPV vaccines before exposure to the virus will possibly have less anal cancer,” the study authors wrote.

Cervarix is approved by the Food and Drug Administration (FDA) to prevent cervical cancer and precancerous lesions in girls and women ages 10 to 25.

Last December, the FDA approved Gardasil ¹⁰⁹ for the prevention of anal cancer in both males and females ages 9 to 26. (Gardasil protects against infection with HPV types 6 and 11, which cause approximately 90 percent of genital warts ¹¹⁰, as well as HPV types 16 and 18.)

The current study also showed that Cervarix offered protection against infection with several HPV types not targeted in the original formulation—types 31, 33, and 45. This was the first time that an HPV vaccine was shown to have “cross-protection” at a site outside the cervix, noted Dr. Harper, who co-wrote an accompanying editorial ¹¹¹ with her colleague Dr. Stephen Vierthaler.

HPV infection can also cause cancers of the vagina ¹¹², vulva ¹¹³, and penis ¹¹⁴, and some head and neck (oropharyngeal ¹¹⁵) cancers. With the exception of screening for cervical cell changes to prevent cervical cancer, there are currently no effective preventive measures for HPV-induced cancers.

“In developing countries where cervical screening programs do not exist or may not be effective, the burden of cervical cancer is far greater than that of all other HPV-associated cancers combined,” said Dr. Allan Hildesheim of DCEG, who co-led the study. In these countries, the focus of HPV vaccination programs should remain on women and on cervical cancer, he added.

The duration of protection from HPV infection with the vaccines remains a big question, noted Dr. Harper. “If the protection lasts less than 15 years, cancers will likely be delayed rather than prevented,” she said.

Because the Costa Rican trial tested anal HPV infection only once, future studies will need to evaluate the vaccine’s efficacy ¹¹⁶ against persistent anal HPV infections and associated lesions, the researchers said.

In the current trial, researchers offered the HPV vaccine to women in the control group at the end of 4 years. “We are transitioning into long-term follow-up and will eventually have 10 years of data,” Dr. Kreimer said.

—Edward R. Winstead ¹¹⁷

Further reading: NCI Fact Sheet: Human Papillomavirus Vaccines ¹¹⁸  

Spotlight

Targeting the Accomplices: 
Homing in on Immune Cells that Aid Tumors

Cancer researchers are increasingly expanding their focus beyond malignant ¹¹⁹ cancer cells to other players within and near tumors, in what is often called the tumor microenvironment ¹²⁰. Many of these studies have implicated a versatile immune cell, the macrophage ¹²¹, as a common tumor accomplice.

Macrophages in and around tumors, often referred to as tumor-associated macrophages (TAMs), can help combat cancer, killing tumor cells as well as calling in the big guns of the immune system, killer T cells ¹²², to help rid the body of malignant cells.

But accumulating evidence from studies performed in cancer cell lines and animal models of different cancer types indicate that, more often than not, TAMs aid tumors in numerous ways and at numerous points, from the earliest development of a malignant mass to the formation of metastases ¹²³.

Data in humans are far more limited but mostly support the cell line and animal model data, with studies consistently showing a correlation between elevated TAM levels and poor prognosis. Researchers working in this area believe that TAMs may not only be valuable prognostic markers that can help in the selection of therapy, but also an important new target to attack tumors.

The concept, explained Dr. Jeffrey Pollard of the Albert Einstein Cancer Center at Yeshiva University, is straightforward. “We need to do more than just attack the tumor cells,” he said. “We need to target the support system, by pulling the foundation out from under tumor cells as well.”

These images from a mouse lung show a tumor cell escaping from the vasculature (red) and forming a tumor cluster (blue). In the final panel, tumor-associated macrophages (green) gather around the tumor cells as they begin to proliferate beyond the vasculature.

A Diversity of TAMs

More recently, researchers have been studying TAMs more closely to learn what these immune cells are doing in and around the tumor and how they do it. It has become clear, for instance, that specific macrophage subtypes provide the heavy lifting tumors need from their surrounding environment to survive and thrive, such as suppressing the immune response or growing new blood vessels.

“Macrophages are very plastic,” explained Dr. Johanna Joyce of Memorial Sloan-Kettering Cancer Center, whose research focuses on the tumor microenvironment. The functions that TAMs perform, Dr. Joyce continued, can vary tremendously depending on the molecular markers they express, the signals they receive from the tumor itself and other components of the microenvironment, and their location—for instance, in the oxygen-depleted core or at the “invading front” of a tumor.

Even though the field has made great strides, she cautioned that “we’re really just scratching the surface of what we understand about these cells.”

Macrophages are often split into two subtypes—M1, or “classically activated” macrophages, which do things like help heal wounds and present antigens ¹²⁴ to killer T cells, and M2, or “alternatively activated” macrophages, which can promote tumor growth and metastasis.

That classification system is probably too simplistic, explained Dr. Pollard, who has been studying the tumor microenvironment for more than two decades. “I think it can blind people to the complexity of the macrophage response,” he said. In mouse models of breast cancer, for example, Dr. Pollard’s lab has identified at least six different macrophage subtypes.

The M1/M2 classification is primarily based on differentiating macrophages in tissue culture, explained Dr. Lisa Coussens of the University of California San Francisco (UCSF).

“You don’t ever see a whole tumor populated by one species of macrophage,” she said. “You see a broad spectrum of macrophage phenotypes.” The macrophage types that predominate in a given tumor are determined in large part by the immune microenvironment around it, Dr. Coussens continued, such as the presence of different types of helper T cells.

Drs. Pollard and Coussens have a collaborative grant to identify markers or patterns of markers that more clearly identify TAM subtypes that are associated with different pro- and antitumor activities, she noted.

Partners in Crime

In the meantime, studies have begun to unravel the co-dependent relationship between tumor cells and macrophages.

Using a mouse model of metastatic breast cancer, for example, Dr. Pollard and his colleagues recently showed ¹²⁵ how macrophages can promote metastasis. Tumor cells that had traveled to blood vessels around the lungs, they found, secreted a chemokine—a hormone-like protein that attracts and mobilizes immune cells—called CCL2 (also known as MCP-1). This action, in turn, drew macrophages to the tumor cells, at which point the macrophages secreted a growth factor called VEGF ¹²⁶, which made the blood vessels leaky, allowing the tumor cells to move into the lungs.

TAMs have also been implicated in disease recurrence. In a study ¹²⁷ using a mouse model of glioblastoma multiforme ¹²⁸ (GBM), Dr. Martin Brown and his colleagues at the Stanford University School of Medicine found that regrowth of tumors that initially shrank after radiation therapy ¹²⁹ was due in part to macrophages and similar cells that were recruited to the tumor site. This recruitment was abetted in large part by a chemokine called SDF-1 (or CXCL12) produced by tumor cells.

SDF-1 “is not stimulated by the radiation itself, it’s stimulated by the product of the radiation,” Dr. Brown said. As the tumor shrinks, its blood supply is diminished and it becomes oxygen depleted, much like tissue in a wound, setting off a chain of events, beginning with the production of the protein HIF-1. This, in turn, induces the production of SDF-1, which mobilizes the macrophage precursor cells from the bone marrow to come into the tumor. Once in the tumor, explained Dr. Brown, the macrophages are retained or “captured” by another protein, CXCR4.

Dr. Brown said much of his work suggests that “what these macrophages are doing [following radiation treatment] is promoting a blood supply. They’re very pro-angiogenic ¹³⁰, provasculature.”

Transition to the Clinic?

TAM research has yet to have an impact on patient care. But it’s beginning to move in that direction.

Several recent studies, for example, strongly suggest that TAM levels can be used as prognostic markers.An excess of TAMs has been linked with adverse outcomes in a growing list of cancers, including breast ¹³¹, thyroid ⁸³, liver ¹³², melanoma ²⁴, lung ¹³³, glioma ¹³⁴, and several blood ¹³⁵ cancers. 

A study published last year ¹³⁶ in the New England Journal of Medicine (NEJM), for instance, found that in tumor samples from patients with Hodgkin lymphoma ²⁶, higher levels of TAMs—identified by the expression of the CD68 marker—were associated with poorer patient outcomes. In fact, there were no lymphoma-related deaths in patients who had limited-stage disease and low TAM levels. And when the initial treatment failed—which happens in about 20 percent of patients—second-line treatment ¹³⁷ with a stem-cell transplant ¹³⁸ was far more likely to succeed in patients with low TAM levels than in those with high levels.

“I think there is overwhelming evidence now that the number of TAMs in tissue biopsies is a valuable prognostic marker in Hodgkin lymphoma,” said Dr. Christian Steidl of the British Columbia Cancer Agency, the first author of the NEJM study.

More than 10 studies have confirmed that high levels of TAMs are correlated with poor primary treatment outcomes and lower overall survival in Hodgkin lymphoma, he continued. “The next steps would be to incorporate this information into clinical trials, and look in a prospective setting to establish evidence that it improves patient management,” he said.

In another study ¹³⁹ published earlier this year, Dr. Coussens and her colleagues found that patients with breast cancer whose tumor samples had what they called a specific “immune signature”—high levels of TAMs (identified with CD68) and low levels of another immune cell, killer T cells—had poorer outcomes than patients with low TAM levels and elevated killer and helper T cells. The presence of the immune signature was most strongly associated with outcomes in women whose tumors were classified as HER2-positive or basal-type/triple-negative, two aggressive breast cancer subtypes.

In the same study, using a mouse model of breast cancer, the investigators found that an investigational drug called PLX3397 hinders the recruitment of macrophages by the cytokine colony-stimulating factor ¹⁴⁰ 1 (elevated levels of which have also been associated with poor outcomes in breast cancer), reducing the cancer’s spread to the lungs and increasing survival.

Dr. Coussens has a grant from Susan G. Komen for the Cure, in conjunction with Drs. Shelley Hwang and Hope Rugo from UCSF, to build on this work, including conducting a multicenter phase I/II clinical trial ¹⁴¹ of women with triple-negative breast cancers ¹⁴² that have recurred after treatment. With the trial and related studies, she hopes to validate the TAM-based immune signature and evaluate the safety and potential efficacy of PLX3397 in combination with chemotherapy.

Results from Drs. Pollard’s and Brown’s recent studies also used agents that target TAMs, showing that these agents could reduce recurrences and improve survival in mice. Human trials testing these agents may not be far off, they said.

If these therapies prove to be effective in humans, in some cases it may be because they are simply killing macrophages. In other cases, it may be because they are depleting certain macrophage populations that aid the tumor, essentially “reprogramming” the tumor microenvironment in a way that can enhance the effectiveness of other therapies and the body’s own natural defenses, Dr. Coussens said.

“It’s such new biology,” she continued. “I really think that by supporting this kind of research, we’ll be able to improve clinical responses going forward.”

—Carmen Phillips ⁵²

A Closer Look

¹⁴³ This article is part of a series of stories related to cancer communications. You can read more articles in the series here ¹⁴³.

The Art of Medicine: 
Using Narrative and the Humanities in Medical Training

“As physicians, we become involved in the stories of our patients’ lives, sometimes as witnesses telling the story through a medical chart. At other times, we become players in the story.”
—Dr. Abraham Verghese, author, physician, and professor

In a classroom at Children’s National Medical Center in Washington, DC, far from the atrium decorated with brightly colored models of hot-air balloons, a group of third-year medical students meets for an hour-long session with two of their professors. They are meeting not to learn the latest in pediatric medicine but to discuss a short story by William Carlos Williams, a family physician and author of short stories, essays, and poems.

The discussion of the story, “The Use of Force,” about a doctor’s encounter with a particularly recalcitrant young patient, is part of the medical humanities program at George Washington University (GWU). Broadly defined, the field of medical humanities applies the humanities, social science, and the arts to medical education and practice. A related discipline, known as narrative medicine, focuses more specifically on reflective writing in the context of medicine and on the close reading of stories, including poetry, fiction, and memoir.

Listening carefully to a patient's story can provide important clinical information that helps the physician make a correct diagnosis.

Proponents of humanities- and narrative-based approaches to medical training argue that the study of literature and the arts helps develop and nurture skills of observation, analysis, empathy, and self-reflection, which are essential for more humane—and ultimately more effective—medical care.

Over the past 20 or so years, many medical schools in the United States and elsewhere have established programs in medical humanities, although not always by that name. Humanities- and narrative-based instruction has also been introduced in some residency programs and in at least one course for oncology fellows. These efforts offer a counterpoint to an increasingly technology-driven health care system that can be impersonal and dehumanizing.

The Power of Story

From time immemorial, humans have used stories to communicate and make sense of their experiences. In narrative-based medical instruction, stories are used to provide insight into the human condition and human interactions. These stories can include physicians’ and patients’ memoirs, fiction, and film.

“The stories become a text that people have in common, and this enables a conversation that you can’t have otherwise,” because students are unlikely to have seen the same medical cases, explained Dr. Linda Raphael, director of the Narrative Medicine/Medical Humanities program at GWU. Such conversations, she said, “allow a very positive sort of ambiguity, where there are different ways of seeing things, all or several of which have value, and there are no absolute answers.” Such ambiguity often arises in the practice of medicine, though it is rarely addressed in traditional medical education or clinical training, which emphasize factual knowledge and technical expertise.

Stories can also unlock the door to discussing thorny issues such as human pain and suffering, death and dying, and the emotional distress and burnout that physicians may experience in their work.

“Medical culture is so focused on making sure that our interactions with patients, at least in a professional environment, are as technical as possible,” said Dr. Alok Khorana, who co-developed a Narratives in Oncology course for oncology trainees at the University of Rochester. “So it’s hard to sit 10 or 11 fellows around the table and ask them to share their feelings. What the humanities piece allows us to do is to use the narratives as a starting point.”

Literary works like those of author and physician William Carlos Williams are now being used as teaching tools in medical education. (Courtesy of the University of Pennsylvania Archives)

In the classroom at Children’s National, Dr. Raphael and Dr. Terry Kind, director of Pediatric Medical Student Education at GWU, lead a discussion of Williams’ “The Use of Force,” which centers on a child who does not want to let a doctor take a throat culture. The doctor, who narrates the story, is ashamed because he realizes that he wanted to win over the child for the sake of winning, not only to make the diagnosis. Dr. Raphael asks whether the students have encountered similar situations, guiding the students into a discussion of the power relationship between doctors and patients and the significance of reflecting, as the doctor-narrator does, on one’s inner experience.

Dr. Raphael holds periodic narrative-based sessions with all third-year medical students as they do clinical rotations in seven medical specialties. Under her direction, GWU also offers elective courses in medical humanities for first-, second-, and fourth-year medical students. While she includes stories, poems, and films that are not specifically “doctor stories” in the elective courses, for the sessions that take place during the rotations she sticks to fiction and nonfiction narratives of doctors and patients that are specific to each specialty.

Narrative in Oncology

In their course for oncology trainees at the University of Rochester, which was added to the formal curriculum in 2009, Dr. Khorana and his colleagues, Drs. Michelle Shayne and David Korones, focus on narratives written by doctors and patients. The three recently published a paper ¹⁴⁴ in the Journal of Clinical Oncology about their experiences piloting the course. The paper includes an appendix with an annotated course curriculum.

“Oncology is a very difficult clinical science, because you are not dealing just with medical decision making, you are also dealing with a lot of difficult issues, especially end-of-life issues ¹⁴⁵,” said Dr. Khorana. Those issues place a burden not only on patients and their families but also on medical providers, he noted, and “there’s no place for trainees to turn for those types of discussions.”

“The broad theme of the course,” he continued, “is to allow expression and training of providers in physician-patient communication, because that’s what it boils down to: understanding that patients have a story that they’re telling you and that you are privileged to hear.”

Dr. Rohit Sud, an oncologist in private practice and an attending physician in a hospital in Chandler, AZ, was among the first to participate in the 1-hour, monthly narrative-based sessions. “An important thing that I took from the narratives,” he said, “was the ability to relate to what the patients were actually feeling as they were going through the stages of acceptance of their cancer diagnosis.”  Dr. Sud also stressed the importance of maintaining empathy in one’s conversations with patients.

“The medical curriculum is such that it diverts you into really focusing on the disease and the treatment, leaving the patient and the caring by the wayside,” added Dr. Sud. “By the time you become a real doctor you’ve been mechanized to do things, and I think it’s important to feel as you do things, and that’s what is lacking.”

For that reason, he said, “it would be a great thing if humanities-based training could be incorporated into the curriculum at a fellowship level or, even better, during residency,” rather than only being included in medical school at most institutions.

Dr. David Dougherty, an oncologist who recently joined the faculty at Rochester, is also an alumnus of the Narratives in Oncology course. One story that stood out for him was “The Median Isn’t the Message,” in which the author, Stephen Jay Gould, brings his scientific experience to bear on his diagnosis of mesothelioma ⁷¹, writing about how survival data and prognostic data are presented to patients and how they perceive this information.

In narrative-based medical instruction, physicians' and patients' memoirs, literature, and film are used to provide insight into the human condition and human interactions.

“To me that was interesting,” said Dr. Dougherty, “because this is information that patients ask us for all the time, but often we’re not aware of how we’re presenting that information or what it means to the individual patient.”

Dr. Dougherty found solace in being able to talk with others who could relate to what he was going through. He also found the course valuable because it provided a regular forum where fellows could learn from the perspectives of a wide range of people, including fellows in other subspecialties of oncology.

Responding to Skeptics

Both Dr. Sud and Dr. Dougherty said that they and their classmates were initially skeptical about the Narratives in Oncology course. They had preconceived notions about what the course would involve and were concerned that it would take time away from what they saw as more important training experiences—feelings often shared by their counterparts in medical school.

To those who remain skeptical of using a humanities-based approach in oncology training, Dr. Dougherty said, “I really think that it touches a place in all of us which draws us to oncology and cancer care. This type of experience can so improve our day-to-day patient interactions, because although oncology is so scientific, and we are so focused on data, we’re dealing with patients who are in very difficult circumstances and who have a lot of emotions, and we bring a lot of emotions to the situation.”

Yet because the area of medical humanities deals largely with subjective concepts, such as empathy, assessing the value of this approach may not lend itself to the usual objective ways of evaluating medical knowledge.

To Dr. Raphael, “the question wouldn’t be so much should you do this at all, but what readings, films, reflective writing, and discussions really work? Which ones have an impact over a long period of time?”

For her and others, the value of narrative-based medical training is evident. “I think it’s important to be enlightened about the nuances of emotions and moral issues,” she said. “People who haven’t talked about the many human issues that are involved in the practice of medicine, including professional ethics, relationships with patients and colleagues, and personal responses to situations, may resist opening up to these matters. This kind of education invites people to open up to those things, to have a greater understanding of them, and not to close off to them.”

—Elia Ben-Ari ¹⁴⁶

Featured Clinical Trial

Sequencing Treatment with a PARP Inhibitor and Chemotherapy

Name of the Trial
Olaparib in Combination with Carboplatin for Refractory or Recurrent Women’s Cancers (NCI-11-C-0022). See the protocol summary ¹⁴⁷.

Dr. Elise Kohn

Principal Investigators
Dr. Elise Kohn and Dr. Jung-Min Lee (Associate Investigator), NCI Center for Cancer Research 

Why This Trial Is Important  
Cancer cells, like normal cells, have a number of mechanisms to repair damaged DNA ¹⁴⁸, a process that is essential for continued cell growth and survival. One mechanism used by cells to repair damaged DNA is the activation of proteins called PARPs ¹⁴⁹, which help repair single-strand breaks in the DNA double helix. Researchers have been interested in determining whether blocking this repair mechanism would hinder the ability of cancer cells to grow and divide. Therefore, they began to develop a group of drugs known as PARP inhibitors ¹⁵⁰ to target this repair mechanism.

A PARP inhibitor called olaparib ¹⁵¹ (AZD2281) has already shown promising activity ¹⁵² against breast and ovarian cancers in women with BRCA1 ¹⁵³ or BRCA2 ¹⁵⁴ gene mutations. The proteins produced by BRCA1 and BRCA2 are involved in a different DNA repair mechanism than that involving PARPs. It was thought, therefore, that treatment with a PARP inhibitor would further reduce the ability of cancer cells with BRCA gene mutations to repair damaged DNA and increase their rate of cell death, a concept known as synthetic lethality ¹⁵⁵.

Because many chemotherapy drugs work by damaging DNA, another way to use PARP inhibitors would be to combine them with DNA-damaging drugs, such as carboplatin ¹⁵⁶. Increasing the rate of DNA damage while inhibiting DNA repair should also reduce cancer cells’ ability to grow and divide, whether they have BRCA gene mutations or not. In preliminary research ¹⁵⁷, combining carboplatin and the PARP inhibitor olaparib in breast and ovarian cancer patients with BRCA gene mutations was shown to be safe and have anticancer activity.

Since olaparib is expected to increase the effectiveness of chemotherapy, some researchers have suggested that the ideal sequence of administration would be to give olaparib first, followed by carboplatin. However, preclinical research ¹⁵⁸ conducted at NCI suggests that administering carboplatin before olaparib may actually result in more DNA damage and cell death.

In this new clinical trial, patients will be treated in 21-day cycles, with the first cycle being either intravenous ¹⁵⁹ carboplatin on day 1 followed by 7 days of oral olaparib or oral olaparib for 7 days followed by intravenous carboplatin on day 8. In the second cycle, the treatment assignments will be reversed. Subsequent cycles will begin with 7 days of oral olaparib with intravenous carboplatin administered on the first or second day. Carboplatin will be stopped after the eighth cycle and olaparib therapy will continue until disease progression ¹⁶⁰ or unacceptable toxicity ¹⁶¹ occurs. Doctors will test blood samples for differences in drug pharmacokinetics ¹⁶² and pharmacodynamics (effects of the drugs on the body) between the administration schedules and will monitor the patients for any differences in safety.

Patients eligible for the trial include women with gynecologic cancers (ovarian, fallopian tube, primary peritoneal, uterine, or cervical) that have not responded to or recurred after previous treatment, women with any type of breast cancer that is metastatic ³² or unresectable ²³ and for which curative therapies do not exist, and men with BRCA-mutation-positive metastatic breast cancer.

“This trial is testing in patients the hypothesis generated by our preclinical work that administering carboplatin followed by olaparib will cause greater DNA damage than olaparib presensitization of carboplatin,” said Dr. Lee, whose work on this study is supported in part by a 2011 ASCO Jane C. Wright MD Young Investigator Award ¹⁶³. “Currently, most studies of PARP inhibitors and chemotherapy give the PARP inhibitor first or at the same time, but there’s really no published data to support that.”

For More Information
See the lists of entry criteria ¹⁶⁴ and trial contact information ¹⁶⁵ or call the NCI Clinical Trials Referral Office at 1-888-NCI-1937. The call is toll free and confidential.

NIH Update

NIH Seeks Ideas for Common Fund

¹⁶⁶

The National Institutes of Health (NIH) Common Fund ¹⁶⁶ supports exceptionally innovative programs that are inherently high risk but have the potential for high payoff by catalyzing research across the entire agency and in the biomedical research community.

New ideas for Common Fund programs are identified annually by internal and external scientists and stakeholders. The fund is now asking for input on these ideas to help shape new programs for 2013.

Program staff want to know which ideas have the potential to fundamentally change how scientists think about, support, or do research in a specific field or to create a new field altogether.

Provide your input online ¹⁶⁷ by September 14.

For more information about the NIH Common Fund, visit the website ¹⁶⁶.