Just In... Meeting News
Key news updates from recent oncology and hematology meetings.
Sunday, March 29, 2015
BY ROBERT H. CARLSON
MIAMI BEACH--The Internet has something for everyone, but when it comes to medical management who's got the time to look at all the potentially helpful websites?
To the rescue comes Hyman B. Muss, MD, Director of Oncology at the University of North Carolina and Director of Geriatric Oncology at the Lineberger Comprehensive Cancer Center, Chapel Hill. In a presentation here at the Miami Breast Cancer Conference, Muss described the “Five Most Useful Websites for Breast Cancer Specialists.”
The online live tour got underway with only one or two of the inevitable technical glitches (“This page can’t be displayed”), but ended with some comic relief.
BC Cancer Agency—Information for Health Care Professionals, Chemotherapy Protocols: www.bccancer.bc.ca/HPI/ChemotherapyProtocols
This Canadian site is the most useful overall, said Muss, who learned about it from an oncology fellow from British Columbia.
The British Columbia Cancer Agency site details information on commonly used regimens for all cancers, along with protocols, dosages, schedules, patient information, printouts on toxicity for specific chemotherapy regimens, and much more information, Muss noted.
Users can print out protocols as documents—important, he said, because it has all the current dose modifications.
Physicians can sign the appropriate orders, and it is especially useful if the office is moving into electronic medical records (EMR), he said. “If you are building EpicCare or another of the EMRs and you need a plan to do your own bills, this site is great.”
The site also has excellent information for patients--www.bccancer.bc.ca/PPI--with practical information such as lists of side effects of chemotherapies for patients, he noted.
“It shows percentages (of side effect rates), which is important because if you just print out side effects of something like cyclophosphamide and doxorubicin and show it to a patient, they will panic. Probably 90 percent of physicians giving adjuvant therapy are using three to four regimens, so you can print these forms out and give them to patients and it will be of terrific help.”
Estimating Survival of Older Patients:
The value of ePrognosis is in estimating the survival of older patients, Muss said. It gives information on life expectancy that can be factored into treatment decisions for older patients with cancer.
The site helps the user choose from a variety of calculators, depending on whether the patient is living at home, in a nursing home, or in a hospital. The time frame and quality of the calculators is shown--“This is excellent for cancer screening, and for determining who is hospice eligible,” Muss said.
Also included are links to many other useful sites. “This site asks you some questions not usually asked about the patient: do you have a memory problem, difficulties with activities of daily living, problems pushing or pulling large objects?”
Muss said these questions have been validated in large studies to predict survival of older patients. “It asks you for your best guess, and then predicts five- and 10-year survivals, which is very helpful for deciding treatment.”
PREDICT, he said, is similar to the well-known Adjuvant! Online (adjuvantonline.com).
“Adjuvant! Online is an outstanding program, but it has certain issues, and one is that it can't compute the added
value of chemotherapy and trastuzumab in HER2-positive patients, and that value is a big deal now,” Muss said.
PREDICT also allows for use of Ki67, he noted, adding, though, that his institution does not use Ki67 because of possible laboratory variation.)
Also, he said, you don't have to be a health care professional to use PREDICT--it's simple and fast, and no password or user name is needed. It's validated on a large number of UK patients.”
Some downsides of PREDICT, though, he said, are that the user cannot enter any comorbidity, and the site does not predict mortality from other causes.
Muss said that he considers this excellent site to be underused, possibly because there is a nominal cost. But it is valuable for anyone involved in maintenance of certification (MOC), self-evaluation programs (SEP), and other teaching modules.
“There is great genetic information on this ASCO site, a lot of tumor boards for virtually all disease sites, and information on curricula in fellowship training.”
The site can also access CancerLinQ, the health information technology platform that ASCO is using to build a massive database on cancer patients, he added.
Take a Break with Quotes from Comedian Steven Wright:
For those who are having a very bad day, Muss said, he concluded with a site (one of several) featuring the humor of Steven Wright, the great one-liner comedian.
· “To steal from one person is plagiarism; to steal from many is research.”
· “If everything seems to be going well, you have obviously overlooked something.”
· “The early bird gets the worm, but the second mouse gets the cheese.”
Sunday, March 29, 2015
By Robert H. Carlson
Miami BEACH--Heterogeneity in cancer genes is ubiquitous--between patients, within the primary tumor, between primary and metastasis, and even between metastases.
“Heterogeneity is ubiquitous, it’s important, and I would argue that we routinely underestimate it,” said George W. Sledge, Jr., MD, Professor of Medicine and Chief of Medical Oncology at Stanford University School of Medicine, who outlined the implications of tumor heterogeneity in a presentation here at the Miami Breast Cancer Conference.
“First, certainly at this point in the history of our field, we need to re-test biomarkers that are of treatment relevance. Most of us already do this routinely when a patient develops metastatic disease, but perhaps we’re not doing it enough.”
The ubiquity of heterogeneity also calls for a definition of what forms are clinically relevant. Sledge said some first steps have been taken in this direction, such as the American Society of Clinical Oncology/College of American Pathologists (ASCO/CAP) Clinical Practice Guideline regarding heterogeneity (OT 11/10/2013 issue).
Another implication of tumor heterogeneity is that when a whole series of mutations is measurable, perhaps the ones that need to be attacked first are the so-called “truncal” mutations, the mutations that are found in all of the tumors. This idea springs from preclinical work, he said.
“And we need to develop treatments that sidestep the genomic instability that troubles us so much.” He noted that for the very first time there are reports of successful immunotherapeutic approaches that may indeed sidestep genomic instability—for example, a Phase Ib study of pembrolizumab reported by Rita Nanda at the most recent San Antonio Breast Cancer Symposium (OT 2/10/15 issue).
Whenever intratumoral heterogeneity is seen it is associated with resistance, Sledge said. “To my mind this is perhaps the leading cause of cancer death, not just in breast cancer, but in all cancers.”
Differences in Orders of Magnitude
Sledge described the magnitude of the genomic complexity and variation, starting with a report from the Cancer Genome Atlas Research Project (Lawrence et al: Nature 2013;499:214-218). This showed in a broad array of different human cancers an extraordinary variation in mutation frequency and spectrum within cancer types.
“There is a three orders of magnitude difference between the least mutated and the most mutated cancers. This is true across all cancers, and also true within cancers to a significant degree.”
One can only imagine the significance of a 1,000-fold difference in the number of mutations a cancer carries, he said.
Within human breast cancers there is again a fairly broad display of heterogeneity. When a triple-negative breast cancer is compared with an estrogen receptor-positive breast cancer, the ER-positive breast cancer has on average about half as many mutations per mega base pair as a triple-negative breast cancer--“So it's not surprising that those are the ones that give us trouble in the metastatic setting.”
Heterogeneity is seen in driver mutations, Sledge said. An analysis of 100 breast cancer genomes showed that the number of somatic mutations varied markedly between individual tumors, with strong correlations between mutation number, the patient’s age when cancer was diagnosed, and the histological grade (Stephens et al: Nature 2012;486:400-404).
The study found driver mutations in at least 40 different cancer genes, and 73 different combinations of driver mutated cancer genes. Furthermore, while 28 of those cancers had a single driver mutation, some had as many as six driver mutations--“We have never targeted six drivers!” Sledge said.
He added: “We have never intentionally targeted six mutations, but we’ve done it unintentionally due to the wonderful promiscuity of many of our receptor tyrosine kinase inhibitors.”
When Is Heterogeneity Clinically Relevant?
Heterogeneity within a patient’s individual tumor plays out in the clinic in terms of outcomes of patients, Sledge said. “Analyses by Edith Perez, MD, and others from large, randomized trials would suggest that having intra-tumor heterogeneity with HER2-positive tumors is what dominates one’s response to therapy. This is also reflected in the ASCO/CAP guidelines that we now use on a regular basis for heterogeneity.”
A paper published in Nature investigated genomic diversity within tumors by using single nucleus genome sequencing to study two tumors--a triple-negative ductal carcinoma and an ER-positive breast cancer (Wang et al: Nature 2014;512:155-160). (“It’s amazing to think that the world’s greatest scientific journal could still get contributions with just two cases,” Sledge added parenthetically.)
The researchers took 50 individual cancer cells from each tumor and applied a deep sequencing approach on each cell.
A heat map derived from the sequencing showed a striking amount of genetic heterogeneity within each cancer, Sledge said. In the triple-negative breast cancer, for example, there were three different cancers within a single cancer.
“This intra-patient heterogeneity is profound, but it gets even worse. Looking again at that triple-negative breast cancer, there were 374 clonal mutations among all the cells, in addition to 154 sub-clonal mutations with three major populations, each of which in turn has its own sub-clonal mutations.”
And as many as a quarter of the mutations identified in this study damage protein function. “The frightening conclusion to the paper was that no two cancer cells within a cancer have the same genome. Let me repeat that: no two cancer cells in the same cancer have the same cancer genome. Think about what that means when you treat a patient with a chemotherapeutic or hormonal therapy or HER2-targeted therapy--we are in very deep waters here when we start looking at tumor heterogeneity.”
Primary-Met, Met-Met Heterogeneity
The discordance between the primary tumor and a metastasis is evident whether the estrogen receptor, progesterone receptor, or HER2 are analyzed, Sledge said. “And it certainly applies if one is looking at a patient with newly diagnosed metastatic disease,” he said. “That disease may be functionally and biologically different than the primary tumor, different in very specific ways.”
If one looks at a patient who has received adjuvant hormonal therapy for estrogen receptor-positive breast cancer, one of the changes that can be seen is the development of mutations in the estrogen cell.
“These estrogen receptor mutations in essence inactivate our ability to treat the patient with many of the hormonal therapies that we use routinely,” he said. “This occurs across a number of studies, somewhere in the range of 10 to 20 percent of the time.”
Estrogen receptor mutations are very rare in the primary tumor, but common in the metastasis. There are less data about heterogeneity between metastases because biopsies for many metastases are not usually done in the clinic, Sledge said.
But there is an exception, a study from the rapid autopsy program at Johns Hopkins School of Medicine (Wu et al: Clin Cancer Res 2008;14:1938-1946).
Rapid autopsies (done one to four postmortem) were performed on 10 consenting patients who died of metastatic breast cancer. Tissues were taken from multiple metastatic sites and compared with the primary tumor. “What was seen, to no one’s astonishment, was that between the metastases there were real differences,” Sledge said.
In one case, HER2 testing with fluorescence in situ hybridization (FISH) showed that some of the metastatic tumors were FISH-negative and therefore HER2-negative, and others were FISH-positive and therefore HER2-positive.
“Think about that the next time you’re in the clinic with a patient with newly diagnosed metastatic breast cancer and you send the patient off for a CT-guided biopsy. I don’t know about your radiologist, but mine would always just find the biggest, juiciest looking [tumor] that’s the easiest one to stick a needle into.
“But we know, as a result of studies like this, that there is real heterogeneity here, and we are making treatment decisions based upon inadequate information.”
Saturday, March 28, 2015
BY ROBERT H. CARLSON
MIAMI BEACH--Obesity’s association with the risk of breast cancer is not as appreciated or understood as well as the connection with other clinical and public health problems such as heart disease and diabetes.
But Clifford A. Hudis, MD, Chief of the Breast Cancer Medicine Service at Memorial Sloan Kettering Cancer Center (MSKCC), clearly explained the implications in a presentation here at the Miami Breast Cancer Conference.
Hudis also described tantalizing preclinical data that could explain this phenomenon: obesity causes a chronic microscopic inflammatory state in breast adipose tissue, which increases aromatase expression, estrogen levels, and cancer risk.
“In a nutshell, this may explain the paradox of ER-positive breast cancer in postmenopausal women, even though estrogen production from their ovaries has precipitously declined,” he said.
Obesity in America
Obesity is a well-known public health problem, with the average American consuming about 20 percent more calories per day than they did in the 1950s: “If you take in more calories and fail to burn them, you have to store them--it’s as simple as that.”
The fundamental shift in national phenotype started in the deep South in the 1950s “and spread out like a virus,” he said. The Robert Wood Johnson Foundation has projected that by the year 2030, some 60 percent of people living in those southern states will be morbidly obese, and no state in the Union will have an incidence of obesity less than 40 percent.
The problem is global, he said, and even in Africa, which is associated with starvation, obesity is now a major public health problem. “While there is no doubt that what we eat matters, we have to examine the calorie balance.”
Obesity Activates Aromatase
There is a two-fold increase in worsened outcomes in breast cancer with obesity, Hudis said, but to be clear, he added, obesity is not associated with elevated risk of premenopausal breast cancer.
“It is however associated on a decile-by-decile linear basis with the incidence of postmenopausal problems.”
Multiple pathways are involved that could directly link obesity with breast cancer--“and it is not as simple as we were taught in medical school.”
Medical students learn that adipocytes are “factories for estrogen,” that aromatase is active in adipocytes, and therefore, as the more obese one is, the more estrogen is made.
But Hudis says it’s more complex than that. He posits that postmenopausal breast cancer risk elevated with obesity can be a consequence of the inflammation that accompanies obesity.
The specific hypothesis is that elevations in estrogen synthesis are related to an increase in pro-inflammatory mediators leading to chronic, low-grade inflammation. Key suspects are prostaglandin E2, TNF-α, and IL-1β.
These combine to cooperatively activate CYP19, the aromatase gene, which leads to increased aromatase activity leading in turn to increased local production of estradiol from the conversion of androgen precursors via aromatization.
In mice fed a high calorie diet, the white adipose tissue inflammation was increased in both the mammary gland and visceral fat, Hudis noted. And around the lesions were elevated levels of the pro-inflammatory mediators, seen in the stromal-vascular fraction and characterized as crown-like structures.
Hudis said the crown-like structure had already been identified by researchers in diabetes. “This is the connection that matters with ER-positive breast cancer. And it explains why the advantage of anastrozole over tamoxifen is widest in women who are lean, and why the aromatase inhibitor’s efficacy decreases in women who are obese.”
Now Hudis and colleagues are pursuing that connection in women with breast cancer.
In a series of 30 women undergoing routine surgery at MSKCC for breast reduction or contralateral risk reduction, the researchers found an association between crown-like structures in breast adipose tissue and increased body mass index (Morris, Hudis, et al: Cancer Prev Res 2011;4:1021-1029).
“This was consistent with what the animal model had predicted,” Hudis said, adding that an ongoing series with similar women continues to show the association between obesity and the crown-like structures.
Additional Risk Factors
There are additional risk factors for inflammation that explain why some lean patients have elevated pro-inflammatory mediators and crown-like structures in the breast. On the other hand, some obese and overweight patients do not have inflamed adipose tissue.
“The inflammation is a systemic phenomenon, and the white adipose tissue of the breast can be a sentinel for generalized inflammation. Moreover, the presence of crown-like cells may be a biomarker of increased breast cancer risk or poor prognosis.”
Hudis said this research suggests that interventions to address the obesity-inflammation-aromatase axis should potentially focus across all domains, including diet, weight control, and anti-inflammatory agents (Howe, Subbaramaiah, Hudis, Dannenberg: Clin Cancer Res 2013; 19: 6074-6083). Other potential interventions are being tested to serve as treatment and prevention.
Hudis concluded by noting that last year was the 50th anniversary of the government report linking tobacco use with lung cancer, and that obesity is now predicted to replace tobacco as the leading modifiable risk factor for cancer in the U.S.
“We have an important obligation to tackle this, and not see obesity undo the advances of the past 50 years.”
Saturday, March 28, 2015
By Robert H. Carlson
MIAMI BEACH--It has been just over a decade since Larry Norton, MD, was honored with the David A. Karnofsky award by the American Society of Clinical Oncology, given “in recognition of innovative clinical research and developments that have changed the way oncologists think about the general practice of oncology.” At the time it was called “sweet vindication” for Norton, now Deputy Physician-in-Chief for Breast Cancer Programs and Medical Director of the Evelyn H. Lauder Breast Center at Memorial Sloan Kettering Cancer Center (MSKCC), for his then radical concept of applying mathematics to cancer biology.
Dose-dense administration of cytotoxic chemotherapy to optimize adjuvant chemotherapy is a practical application of the concept championed by Norton.
Clifford A. Hudis, MD, Chief of the Breast Medicine Service at MSKCC, a speaker here at the Miami Breast Cancer Conference, said the practice of dose-dense chemotherapy is alive and well. In fact, he emphasized that it appears to provide a better baseline platform for the addition of targeted therapies than standard treatment scheduling.
Dose-dense scheduling of chemotherapy refers to the use of optimal standard doses administered more frequently, often with the aid of growth-factor support. The simplest idea about dose-dense therapy is that more frequent or dense dosing will decrease the time for tumor growth; it will allow for ever more treatment of ever smaller volumes of residual tumor; and it results in greater overall cell-kill.
Delving into the history of dose-dense scheduling, Hudis said the practice-changing trial was CALGB 9741 (JCO 2003;10:1431-1439). Hudis, who was an investigator on this trial, said this was key in the acceptance of dose-dense chemotherapy because it was positive, and well controlled.
The trial compared concurrent AC (doxorubicin-cyclophosphamide) versus sequential A then C, and in both cases paclitaxel was given as a single agent. Using a two-by-two factorial design, the other comparison was every-three-week therapy versus every-two-week, thus requiring colony stimulating factor support.
“What made this a test of dose density was that every single patient got the exact same total dose, the exact same number of doses of each drug, and the exact same dose sizes. So the variable was schedule or sequential versus concurrent dosing.”
This study was positive when first reported, he said, and it remains that way through a 10-year follow-up, showing that every-two-week dosing is superior than every-three-week, and that was true for overall survival as well.
“This was a trial that fundamentally made dose-dense therapy a standard option,” Hudis said. Six years later, a clinical trial of dose-dense AC plus targeted therapy with trastuzumab (JCO 2009;36:6117-6123) brought a pleasant surprise concerning cardiac toxicity with the dose-dense schedule--a rate of approximately two percent—in line, he said, with what would be expected with doxorubicin.
“Notwithstanding everybody’s a priori bias that it must be more toxic, these data suggest that it is not more dangerous in terms of cardiac toxicity.”
Subsequent trials at MSKCC have highlighted a secondary benefit of the dose-dense platform: that alternative or additional targeted therapies such as trastuzumab or lapatinib can be added.
Six Cycles Not Superior to Four
Hudis described a more recent study, CALGB 40101 (JCO 2012;33:4071-4076), also a two-by-two factorial design, that asked two questions:
· “First, in low-risk breast cancer, using every two-week scheduling and filgrastim support, do we need the anthracycline-cyclophosphamide combination, or would single-agent paclitaxel do the job in these low-risk patients?” and
· “Secondly, would six cycles of therapy be better than four?”
The answer was that there was no difference whatsoever in progression-free survival or overall survival between six cycles and four.
Similarly, single-agent paclitaxel was not non-inferior.
“I’m sorry for the double-negative, but from a statistical design point of view, that’s the way this study was set up,” Hudis said. “The practical take-home message is, don’t use single-agent paclitaxel in the adjuvant setting, it really doesn’t do the job. You need AC here.”
Hudis said he was not arguing the need for combination therapy, though: “What I’m arguing is that for low-risk breast cancer in the adjuvant setting, you do need more than one chemotherapy drug for chemo-sensitive tumors.”
Even shorter intervals have been tried in pilot studies. The MSKCC 03-092 trial (Clin Cancer Res 2007;13:223-227) looked at 10- or11-day intervals, using epirubicin and cyclophosphamide followed by paclitaxel, with filgrastim support (not pegylated). This approach was feasible, Hudis said.
Then, clinicians in the Memorial network tested the CMF regimen (cyclophosphamide-methotrexate-fluorouracil) in a dose-dense schedule. That trial, MSKCC pilot study 07-0133, “made IV-CMF into a more dose-dense regimen for convenience, if nothing else.
“Indeed, we were able to deliver this at 14 days, and then we could even deliver it at 10- or 11-day intervals, all of which was possible” (Clin Breast Cancer 2010;6:440-444).
‘Remains a Standard’
Hudis said that fundamentally, the test of time for dose-dense therapy is that it remains a standard in the administration of chemotherapy: “It’s clearly important at the extremes--if you don’t give your chemotherapy at a frequent enough interval, you lose efficacy.
And it is also less or equally toxic--not more toxic--than standard scheduling. That’s one of the long-standing myths about dose-dense scheduling that is hard to overcome.
“You clearly need to use supportive care and growth factors to do [dose-dense scheduling],” Hudis concluded, “But then this allows you to deliver more effective chemo.”
Saturday, March 28, 2015
BY ROBERT H. CARLSON
MIAMI BEACH--The prevailing evidence shows that routine use of laboratory studies and imaging to monitor for recurrence in asymptomatic patients with early-stage breast cancer is of no value, a recommendation that is part of the guidelines of both the American Society of Clinical Oncology and the National Comprehensive Cancer Network. The topic was discussed here at the Miami Breast Cancer Conference by Hyman B. Muss, MD, Director of Oncology at the University of North Carolina and Director of Geriatric Oncology at the Lineberger Comprehensive Cancer Center.
Muss noted that both ASCO and the NCCN suggest that routine follow-up care be limited to clinical history and physical examinations, yearly mammography, gynecologic examinations (especially for women taking tamoxifen), and assessment for genetic counseling.
Because metastatic breast cancer remains incurable regardless of the volume of disease when first detected, routine use of costly, high-technology studies should be discouraged, he said. “The most difficult issue about applying these guidelines is convincing patients that some of the amazing technologies we have--circulating tumor cells, tumor markers, exquisite imaging procedures--don't save lives.”
The current guidelines are prudent, Muss said, and are also part of ASCO’s Choosing Wisely list of tests and procedures to question (www.choosingwisely.org/doctor-patient-lists/american-society-of-clinical-oncology)--“I recommend that we all adhere to these follow-up guidelines,” Muss said.
Perception off on Tumor Markers
Assaying tumor markers in the follow-up is controversial in the setting of early-stage disease, he noted. Even though there is no evidence that these assays are useful in this setting, there is a strong perception among patients and some physicians that the tests are helpful.
“I tell women [who want tumor marker tests] that about 30 percent of women who die of breast have tumor markers that are normal,” he said.
It's a psychological burden for the patient and for the oncologist when the tumor marker finding goes up slightly but nothing is found on CT scans and other tests--“The patient is sitting by the telephone for months after [the test], and it is miserable.”
That said, recurrences can come on very quickly, Muss said, and some 30 to 40 percent of occurrences are found between routine visits.
“Instruct your patients on the signs and symptoms of metastasis,” he said:
· Bone: pain and tenderness;
· Skin: characteristics of lesions;
· Lung: dyspnea, pleurisy;
· CNS: focal findings, loss of function; and
· Gastrointestinal: pain, fatigue.
“And give everybody the ASCO follow-up guidelines. They're easy to get and print out from cancer.net/survivorship--and document it!”
Who Should Do the Follow Up
“Are oncologists the best resource for serving these patients?” Muss asked. “We [oncologists] are running out of manpower, and there is evidence that family physicians do as good a job in monitoring for recurrence as oncologists.”
He noted a Canadian study several years ago of 968 patients with early-stage breast cancer assessed nine to 15 months after diagnosis, randomly selected between 1997 and 2001 for follow-up with either an oncologist or a family physician (Grunfeld et al: JCO 2006;6:848-855). The family physicians were provided with appropriate guidelines. The recurrence rates were 11.2 percent for family doctors and 13.2 percent for oncologists; mortality rates were 6.0 versus 6.2 percent, respectively; and the rates of serious events were 3.5 versus 3.7 percent.
“In other words, in terms of the patients’ physical and mental health, there was no difference between the family doctors or oncologists doing the follow-up,” Muss said.
He added, however, that the situation is a bit different today, with the importance of compliance with endocrine therapy--the standard of care for follow-up, besides to detect clinical recurrence, is to monitor compliance with endocrine therapy. “The majority of breast cancer survivors now have ER-positive disease, but there is tremendous drop-off in compliance with endocrine therapy. And if they are on aromatase inhibitors, you've got to pay attention to their bone health.”
What's the Rush?
Finally, Muss addressed that underlying question: “What's the rush to find metastatic disease?”
He listed nine clinical trials of first-line chemotherapy for metastatic breast cancer, in which median survival was 15 to 20 months.
“And some trials now are up to 25 months,” he said. “That improvement may be due to better medical treatments or to lead time bias with better imaging or to both, but in any case there is no hurry to find incurable disease in patients who are asymptomatic.”
It is even questionable whether such studies detect metastasis earlier than waiting for symptoms of metastasis to appear, he said.
“We are not doing patients any service, not with the current state of technology.”