We present highlights on value in oncology, immunotherapies, and genomics from the 2014 ASCO Annual Meeting.
This year marked the 50th anniversary of the first Annual Meeting of the American Society of Clinical Oncology (ASCO). Presenting a “wrap-up” of such a momentous and mammoth event is a formidable challenge. We haven’t the space to be anywhere near comprehensive. In addition, we know that by this point, most oncologists have doubtless read or heard about the important abstracts presented at the meeting. Thus, we thought what would be of greatest value to our readers would be perspective and insight. To this end, we have solicited reflections from the journal’s esteemed Editorial Board members-asking them not just the usual questions about what studies they thought were most important, but also what they thought were especially interesting “sleeper” studies, which trial results were likely to have the greatest long-term impact on practice, and which sessions they hadn’t intended to go to but wound up being really glad they did attend. Keenly aware that oncology is one of the fastest-changing medical specialties, we also thought it important to get the perspective of the field’s younger members. We thus enlisted the help of what we’ve dubbed our “ASCO Street Team”-a group of more than a dozen talented and passionate young oncologists who contributed on-site reporting and blogging from Chicago.
As we pored over the thoughts and submissions of this large and varied group of contributors, a few themes began to emerge. It seemed that many of the most significant aspects of ASCO 2014 fell into the following categories:
• Value in oncology
• Immunotherapy
• Genomics
In the pages that follow, we present highlights in these three areas.
-The Editors
If there was any consensus as to the overarching theme at ASCO this year, it was the issue of what constitutes value in oncology, and how we create that. This was a point made by a number of ONCOLOGY’s Editorial Board members, as well as by several of the young oncologists who blogged for the journal’s ASCO Street Team. ASCO’s outgoing president, Dr. Clifford Hudis, made the issue a central theme in his presidential address. Questions of value were raised in the discussions of two of the three plenary session abstracts. And “cost” and “value” were central issues in a number of heavily attended sessions.
"We cannot avoid today the issue of the cost of care. . . . If we intend to achieve social justice in cancer care, we must define value in cancer care. . . . The assessment of value is a far more complex issue than simply adding up the price of drugs. We are clearly best positioned to serve as honest brokers in this discussion."
-Clifford A. Hudis, MD, in his Presidential Address
Results from the phase III E3805 clinical trial, which were presented at the plenary session, showed that adding docetaxel to standard hormone therapy extends survival for men with metastatic hormone-sensitive prostate cancer by 13.6 months. The extension of survival was greatest among men with extensive disease-a difference of 17 months. The trial was led by the Eastern Cooperative Oncology Group and funded by the National Cancer Institute.
“I am not aware of any prostate cancer study that has offered this magnitude of improvement in survival,” said current ASCO president Clifford Hudis, MD, chief of the Breast Cancer Medicine Service at Memorial Sloan Kettering Cancer Center in New York City, at a press briefing. “Across all the solid tumors, this is an almost unprecedented improvement in median survival.”
A significant reason for the excitement this trial generated was that the stunning results were achieved with an agent-docetaxel-that is relatively inexpensive compared with newer targeted therapies that are used to treat advanced prostate cancer.
The results of the Tamoxifen and Exemestane Trial (TEXT) and the Suppression of Ovarian Function Trial (SOFT) demonstrated that combining exemestane with ovarian function suppression (OFS) in women with premenopausal hormone-sensitive breast cancer results in a 34% reduction in the risk of breast cancer recurrence, compared with treatment with tamoxifen in combination with OFS. (Presented at 2014 ASCO Annual Meeting; Abstract LBA1; reproduced with permission.)
In her discussion of the TEXT/SOFT results, ONCOLOGY Editor-in-Chief Nancy Davidson, MD, of the University of Pittsburgh, did not shy away from the question of what role costs are likely to play in future decisions about the optimal management of younger women with hormone-sensitive breast cancer. While exemestane plus OFS have now been shown to confer an advantage, this came at a considerably increased cost compared with tamoxifen-and an overall survival benefit has yet to be proven.
The above prices, which were included in Dr. Davidson's presentation, were obtained from the University of Pittsburgh Medical Center pharmacy. Note that drug costs vary greatly across medical settings. Slide reproduced with permission.
The subject of cost, not surprisingly, was discussed at several Health Services Research sessions that I attended at ASCO. In the oral abstract sessions for Health Services Research, cost-effectiveness was discussed on a more tangible level. One abstract was titled “Bevacizumab in Addition to 5-Fluorouracil–Based Chemotherapy for Metastatic Colorectal Cancer: A US-Based Cost-Effectiveness Analysis for 1st and 2nd Line Therapy” and was presented by Daniel Goldstein, MD. Using Markov models, the authors calculated the cost-effectiveness of using FOLFOX with or without bevacizumab. They found that the incremental cost-effectiveness ratio (ICER) for the addition of bevacizumab to chemotherapy was $240,195 per quality-adjusted life year (QALY), and using bevacizumab in the second line yielded an ICER of $219,724/QALY.
The threshold level for an ICER in the United States is still a matter of great debate. The authors of the study mentioned that $50,000–$100,000/QALY is acceptable. Sandra Wong, MD, subsequently discussed the study, and she thought that the threshold value for an ICER should be $150,000/QALY. Still, the numbers found in the bevacizumab study far exceeded any of these thresholds. The study was very well conducted and described at ASCO; yet, I wonder if and how it will actually affect practice.
Will we actually stop adding bevacizumab to chemotherapy as a standard option for metastatic colorectal cancer because it is not cost-effective to do so? Certainly the benefit from bevacizumab is modest and the cost is quite high. If we are to curb the unsustainable growth of medical spending, things like using bevacizumab in this setting should be reconsidered.
-Amy Soni, MD
On Sunday at ASCO, ONCOLOGY Street Team blogger Aaron Falchook, MD, attended a Health Services Research Education Session entitled “Overdiagnosis and Overtreatment in Cancer: Point/Counterpoint.” The room was packed, he reported. As anyone who has attended a scientific meeting knows, presentations often take place in large rooms and sometimes many seats are empty. This session was extremely full, with just a few open seats available. Clearly, this year's ASCO attendees voted with their feet, showing that they think the topic of overdiagnosis and overtreatment is extremely relevant to oncologists today.
One solution proposed and discussed during the exchanges was the development of decision-aid tools that patients and providers could use to interpret the results of screening tests and to more rationally weigh treatment options. A study by Greenup et al showed that using an algorithm to help patients decide among treatment options for early-stage breast cancer and ductal carcinoma in situ resulted in a 43% reduction in treatment costs.
During the after-presentation Q&A session, one of the questions received enthusiastic applause from the audience. The question related to the fact that providers do not have incentives to reduce potentially unnecessary diagnostic tests and/or treatments. Potential strategies to mitigate this critically important issue were discussed. For example, state legislatures can pass “safe harbor” legislation, which protects providers from liability if they forgo screening tests in accordance with published national guidelines. Also, the current fee-for-service reimbursement model incentivizes more care, though not necessarily better care. The point was made that potential changes to payment models (fee-for-quality and payment bundling) may, in the future, help create a financial incentive to deliver high-value and high-quality care.
However, an eloquent rationale for opposing the use of physician-directed financial incentives to reduce healthcare costs was provided at another debate. At this session, entitled “The Value of Cancer Care and the Professional and Ethical Obligations of the Practicing Oncologist: A Debate,” Daniel Sulmasy, MD, PhD, of the University of Chicago, argued that bedside medicine has always been-and always will be-about individual patients. He made the point that financial incentives for bedside rationing require physicians to put aside their professional ethos and undermine trust by making the doctor an agent of the state or the market. He added that not only will any type of bedside rationing automatically result in injustice, but it is also unlikely to achieve the goals that its advocates claim it will achieve.
Dr. Sulmasy readily acknowledged that something needs to be done to rein in healthcare costs. Moreover, he believes that oncologists have a definite role to play in this “reining-in.” However, he proposed a strategy that would constitute a clear alternative to the growing tendency to treat medicine as a business and to reward physicians financially for using less costly treatments. This alternate strategy would consist of two components. The first of these would be to strive to practice the art of medicine better and more rationally, aiming for what he termed “therapeutic parsimony” (using only enough therapy to make the patient well) and “diagnostic elegance” (ordering only those tests necessary to help the patient, resisting the impulse to satisfy physicians’ and patients’ technophilia). Both of these goals would be pursued purely in the interest of the patient; cost savings would inevitably result-but these would be just a happy side effect.
The second component of the strategy that Dr. Sulmasy proposed was for oncologists to involve themselves in what he termed “engaged citizenship.” Everyone in the United States, he said, needs to engage in the difficult political conversation about how much healthcare the country can afford and how this should be distributed fairly. Physicians, as much as other citizens, should be involved in this conversation.
A bit surprisingly, Dr. Sulmasy’s “opponent” in this debate, Reshma Jagsi, MD, of the University of Michigan Health System, was largely in agreement with Dr. Sulmasy. Not only should physicians involve themselves in the national conversation about healthcare costs, they should play a leading role in it, she said. She noted that physicians’ specialized knowledge-which is almost always acquired at least in part through public funding-gives them a professional duty to participate in a prominent way in the national effort to rein in healthcare spending. “We are in such a privileged role,” she said, “we should be leading the charge.”
Implementation of a national lung cancer screening program using low-dose computed tomography (LDCT) will identify almost 55,000 additional lung cancer cases over 5 years, but would add more than $9 billion to Medicare expenditures, according to results of a new study. Medicare is expected to release a draft decision on screening coverage in November of this year.
In the new study, Joshua A. Roth, PhD, of the Fred Hutchinson Cancer Research Center in Seattle, and his colleagues used a model to forecast the 5-year results of implementation of a screening program in comparison with no screening program. They used National Lung Screening Trial data along with other information from the SEER database, and peer-reviewed literature. The group modeled various scenarios involving faster and slower diffusion of the program. In a scenario assuming gradual diffusion of LDCT screening (which Roth likened to the way in which mammography was introduced), the model forecast 54,900 additional lung cancer diagnoses, most of which would be stage I. Currently, 15% of lung cancers are detected at the localized stage, but the LDCT program would raise that number to 32%.
The model also showed that an additional 11.2 million LDCT scans would be performed, with 2 million false-positive results. Overall, this would result in $9.3 billion in excess expenditure; the bulk of that amount, $5.6 billion, would be the cost of the CT scans themselves. Another $1.1 billion would be for diagnostic work-up, and an excess $2.6 billion would be spent on cancer care.
Presented at the 2014 ASCO Annual Meeting; Abstract 6501; reproduced with permission.
“This is truly a brave new world of immunotherapy,” remarked Steven O’Day, MD, clinical associate professor of medicine at the University of Southern California, Keck School of Medicine, who chaired the press conference on immunotherapy at ASCO this year. “The message is, the revolution is here, [and] it’s ongoing.”
Immunotherapy for the treatment of melanoma has made headlines at the past several ASCO annual meetings. This year, the results of three especially noteworthy trials of immunotherapies in melanoma patients were presented: a stage III study showed that ipilimumab, when used adjuvantly, can be effective in earlier-stage melanoma; more mature data from the phase I trial of ipilimumab plus nivolumab, first presented at ASCO 2013, are even more impressive than the early data; and in one of the largest phase I studies ever conducted, a new anti–programmed death (PD)-1 agent, pembrolizumab (formerly MK-3475), has demonstrated a high response rate in patients with advanced disease who had stopped responding to ipilimumab.
The results of the randomized phase III study of ipilimumab in patients with stage III melanoma previously treated with surgery and at high risk for relapse (EORTC 18071) were presented by Alexander Eggermont, MD, PhD, director general of the Gustave Roussy Cancer Center in Paris. “This trial with ipilimumab is the first to show we may be able to give these new drugs earlier in the course of disease, where they can do more good and potentially cure more patients,” Dr. Eggermont said. At a median follow-up of 2.7 years, the recurrence rate was substantially lower in the ipilimumab group (34.8%) than in the placebo group (46.5%).
The long-term follow-up data from the phase Ib trial of ipilimumab plus nivolumab were presented by Mario Sznol, MD, professor of medical oncology at the Yale School of Medicine in New Haven, Connecticut. In this study of 94 patients with inoperable stage III or IV melanoma who had undergone up to 3 prior systemic therapies, 41% of the patients who received concurrent therapy (n = 53) responded to the treatment, and 17% (n = 9) had a complete remission-up from the 10% complete response rate reported in 2013. Moreover, the responses were durable, with 82% ongoing at the time of analysis. Dr. Sznol underscored the remarkable nature of these data: “Just a few years ago,” he said, “median survival for patients diagnosed with advanced melanoma was as little as a year or less, and . . . we’re seeing a median survival over 3 years in this trial.”
Some of the most exciting new drug data to be presented at ASCO this year came from the large phase I study of pembrolizumab. As a single agent, pembrolizumab improved on the results achieved with single-agent ipilimumab, and with significantly less toxicity. Of the 411 patients in the study, 34% achieved a response, and these responses were durable, with 88% ongoing at the time of analysis.
A proof-of-principle, nine-patient study demonstrated that women with advanced metastatic cervical cancer can respond to immunotherapy. An adoptive T-cell technique was used in which a patient’s own immune cells were harvested, specific T-cells that could respond to two viral antigens that target human papillomavirus (HPV) were selected and expanded, and then billions of these cells were infused back into the patient. Two of the women are currently in complete remission (for 15 and 22 months, respectively) after a single T-cell infusion. One patient had a partial response. The results are exciting for a disease that has a typical median survival of only 3.5 months. Many more studies are needed to better test the methods of this approach. Reproduced with permission from Hinrichs CS, Stevanovic S, Draper L, et al. HPV-targeted tumor-infiltrating lymphocytes for cervical cancer. 2014 American Society of Clinical Oncology Annual Meeting; Abstract LBA3008.
"I think the PD-L1 story in bladder cancer was the most dramatic one in genitourinary oncology. Finally there is a credible therapy that is likely to shift survival in an important and necessary way in this disease in which there has been little progress for decades; this further emphasizes the range of progress that immunotherapy has made in reshaping oncology."
-Paul Mathew, MD (Tufts Medical Center)
One of the things I found most exciting at ASCO was the series of presentations focusing on unleashing the potentials of the immune system in genitourinary cancers-an area with a great need for effective, novel therapeutic approaches. Ground-breaking results were presented by Thomas Powles, MBBS, MRCP, MD, clinical professor of genitourinary oncology at Barts Cancer Institute of Queen Mary University of London. He discussed his abstract on the phase I study of inhibition of programmed death ligand 1 (PD-L1) by MPDL3280A in patients with metastatic bladder cancer. As the response rate data were displayed, one could almost hear a collective sigh of relief in the overcrowded Arie Crown Theater, one of the biggest presentation halls at McCormick Place. The 68 patients had advanced disease that had progressed through 2 or more prior lines of chemotherapy, with 75% having visceral metastasis. In such a setting, any single-agent cytotoxic approach is expected to provide a response rate on the order of 5% to 10%, with a dismal prognosis. There are no approved agents for second- or third-line treatment of metastatic urothelial cancer.
The authors were proud to present a tantalizing overall response rate of 52%, with 2 patients achieving a complete radiologic response. Patients who responded included those with visceral metastasis, and all of them continued to show a response at the time of clinical cutoff. Making this approach even more promising, the exceptional safety profile of the drug was unmatchable. There was only a 4% rate of grade 3/4 toxicities, with no immune-related adverse events or any nephrotoxicity.
-Ali Amjad, MD
Several of ONCOLOGY’s Editorial Board members and a number of our ASCO Street Team bloggers expressed a keen interest in the sessions at this year’s annual meeting that were devoted to advances in genomic profiling-or to questions regarding this new and fast-growing field. Drs. Vered Stearns and John Sweetenham both cited studies involving genomic tests as among the most interesting they heard presented (one of these, interestingly, was also in part an “immunotherapy study”-still more testimony to the importance of that theme at the conference). Blogger Ali Amjad noted that in his opinion, genomics and immune therapies-in that order-were the two predominant themes at the meeting. And perhaps not surprisingly, given the complexities of these assays, a session entitled “How to Talk to Your Pathologist” proved quite popular. A cautionary note was expressed by Board member Paul Mathew, however, on account of the costs involved (going back to the first of our three themes). “I am less certain that the whole genome profiling [field] will yield personalized solutions to oncology beyond a few storied examples,” he said. “So there is a tension between the need to personalize medicine and the cost-yield benefit of screens for personalized and actionable targets using such strategies.”
What, in your opinion, was the biggest “sleeper” story at ASCO this year in your field?
Vered Stearns, MD (Johns Hopkins): “smallish studies with new biomarkers of immune response in breast cancer.”
What study presented at ASCO this year do you think will have the greatest long-term implications for your specialty?
John Sweetenham, MD (Huntsman Cancer Institute): “a preliminary study from Stanford assessing the use of tumor-specific DNA in the plasma of lymphoma patients as a method of disease surveillance-could replace surveillance imaging. Very preliminary data but very intriguing.”
Edith Perez and colleagues used whole-genome DASL technology to identify genes associated with relapse-free survival (RFS) in patients enrolled in the NCCTG N9831 adjuvant trastuzumab trial. Ten of the 13 enriched biological processes associated with increased RFS were found to be linked to immune functions; these 10 processes defined a cohort of 87 immune function genes, 14 of which were found to be independently correlated with RFS in patients who received trastuzumab but not in patients who did not receive trastuzumab. Reproduced with permission from Perez EA, et al. Association of genomic analysis of immune function genes and clinical outcome in the NCCTG (Alliance) N9831 adjuvant trastuzumab trial. J Clin Oncol. 2014;32(suppl 5s):abstr 509.
In this study of 50 patients with diffuse large B-cell lymphoma (DLBCL), high-throughput sequencing (HTS) to detect tumor-specific immunoglobulin (IG) genes was used to monitor minimal residual disease (MRD). The performance of this parameter in both the cellular and acellular compartments of blood were compared, and both were compared with other measures of disease. Compared with radiographic assessment, IG-HTS in the acellular compartment (ie, in plasma) was found to have a sensitivity of 67% and a specificity of 100%. Reproduced with permission from Kurtz DM, Green MR, Bratman SV, et al. Noninvasive monitoring of cellular versus acellular tumor DNA from immunoglobulin genes for DLBCL. J Clin Oncol. 2014;32(suppl 5s):abstr 8504.
With the advent of tumor genomic profiling and the molecular stratification of tumors, we have entered into an age of personalized medicine. There has been a transition from single-gene testing to multi-gene testing, and now to a much broader approach that can be offered by whole-exome and whole-genome sequencing. The amount of information that’s at our hands can be overwhelming. While the promise of tumor genomic profiling clearly offers the hope of an individualized approach for each patient’s particular cancer, there are both foreseeable and unforeseeable ethical dilemmas that loom. What are the risks and benefits for both cancer patients and providers in the clinical implementation of tumor profiling? This was the topic of a thought-provoking ASCO Education Session entitled “Tumor Genomic Profiling in Cancer Care: Field of Dreams or Ethical Minefield?”
As the discussion focused on the application of ethical principles to our clinical use of tumor profiling, I found many of my own questions and concerns articulated throughout the talk. The principle of “justice and fairness” refers to the ability to offer “equitable access and outcomes for diverse patient populations.” The principle of “respect for persons” raises the questions of how patients should be educated about molecular testing and its implications, how we need to counsel our patients, and whether or not informed consent is necessary.
As I envision having these types of sophisticated and important conversations with my patients about tumor genetic profiling, I anticipate multiple potentially difficult scenarios. First of all, while these tests are theoretically available, it is not always clear when to use them or if they are accessible at any given site. Furthermore, educating a patient about the possibility of incidental findings, such as identifying mutations that we have no treatment for and no understanding of, may be challenging for both the patient and provider. This becomes even more difficult in a setting in which a patient has limited medical knowledge or other barriers-language or cultural.
Furthermore, the talk raised the question of whether these genetic profiling technologies have been, or have the potential to be, “over-adopted.” While position papers exist, there are currently no strict oversight or regulatory committees that govern the use of tumor profiling in the clinical setting. Should this testing be only under the auspices of a research study with informed consent, or can these tests be used alone in the clinical care setting? Is there an obligation to report all the details of the tests to each patient, even when we ourselves as physicians may be unsure how to interpret the results? Where does our professional responsibility lie in reporting findings, and do we need to follow up on potential secondary findings? And ultimately, will the results of these tests change our management of our patients? At the present time, I think the answer is unclear. However, for the future, I believe the answer will be yes.
-Sarah Weiss, MD