GBM HAS 4 MOLECULAR SUBTYPES

January 21, 2009 — Glioblastoma multiforme has 4 distinct molecular subtypes, according to new data emerging from The Cancer Genome Atlas (TCGA) Research Network. The different subtypes show varying responses to aggressive chemotherapy and radiotherapy, with a difference of around 50% between the subtypes.

The new findings, published in the January 19 issue of Cancer Cell, could eventually lead to more personalized approaches to therapy.

The hope is that this will improve the current poor prognosis for glioblastoma multiforme, which has a median survival rate of about 1 year. This is "one of the most feared of all human diseases, both for its near uniformly fatal prognosis and its associated loss of cognitive function," the researchers write.

"We discovered a bundle of events that unequivocally occur almost exclusively with a subtype," said lead author D. Neil Hayes, MD, from the University of North Carolina at Chapel Hill. "These are critical events in the history of the tumor's development and spread, and evidence is increasing that they may relate to the initial formation of the tumors."

The nature of these events indicates that the pathology of each subtype might begin from different types of cells, and these differences might explain the variation in response to therapy, the researchers explain.

"The ability to differentiate glioblastoma multiforme tumors based on their altered genetic code lays the groundwork for more effective treatment strategies to combat this deadly cancer," Eric Green, MD, PhD, director of the Human Genome Research Institute (NHGRI), said in a statement issued by the National Institutes of Health.

The NHGRI, together with the National Cancer Institute (NCI), provides funding for TCGA Research Network, where the discovery was made.

Four Distinct Subgroups

The work carried out by Dr. Hayes and colleagues was based 206 patient samples and sequence data from 91 patients and 601 genes.

Genomic profiling resulted in a gene-expression-based molecular classification of glioblastoma multiforme into 4 distinct subtypes: proneural, neural, classical, and mesenchymal.

The reproducibility of this classification was demonstrated in an independent validation set, so it is highly unlikely that this is a spurious finding, they write.

It is also unlikely that patients will transition from one subtype to another during the course of their disease, they add.

Dr. Hayes used the new data in conjunction with clinical data from a previous study (J Clin Oncol. 2008;26:3015-3024) to examine the effect of treatment on the different subtypes. In this investigation, intensive treatment was defined as concurrent chemo- and radiotherapy, or more than 3 subsequent cycles of chemotherapy.

They found that response to such intensive therapy varies by subtype. The greatest benefit was seen in the classical and mesenchymal subtypes, where intensive therapy significantly reduced mortality. There was a suggestion of efficacy in the neural subtype, but no benefit from intensive therapy was seen in the proneural subtype, they report.

This might be the most clinically relevant finding; it suggests that gene-profiling-based classification could indicate different treatment strategies, the researchers note.

"These new findings offer critical insights into the stratification of patients based on the unique molecular characteristics of their disease," John Niederhuber, MD, director of the NCI, said in the press release. "As we learn more and more about the genetic underpinnings of cancer, we hope to achieve a similar level of molecular understanding for all cancers and, eventually, to generate recipes for highly targeted therapies uniquely suited to the individual patients."

Cancer Cell 2010:17:98-110.

OXALIPLATIN USEFUL FOR OLDER PATIENTS WITH STAGE III COLORECTAL CANCER

January 22 (Orlando, Florida) — In contrast to some previous studies that found no benefit with newer chemotherapy regimens in elderly colorectal cancer patients, a randomized phase 3 trial showed better outcomes with adjuvant capecitabine and oxaliplatin (XELOX) than with standard 5-fluorouracil and leucovorin (5-FU/LV). The results came from a subanalysis of the NO16968 study, which compared adjuvant XELOX with standard chemotherapy in stage III patients.

"Half of all newly diagnosed colon cancer patients are over 70 years old and they are frequently not offered the same care as younger patients," said Daniel G. Haller, MD, the Deenie Greitzer Professor of Gastrointestinal Oncology at the University of Pennsylvania's Abramson Cancer Center in Philadelphia.

He presented his findings here at a press briefing as part of the 2010 Gastrointestinal Cancers Symposium. The annual meeting is cosponsored by the American Gastroenterological Association, the American Society of Clinical Oncology (ASCO), the American Society for Radiation Oncology, and the Society of Surgical Oncology.

"The use of newer chemotherapy drugs in the adjuvant setting for older patients with stage III colon cancer has been a topic of controversy, and these findings contradict conclusions from 2 recent studies," he noted. An analysis of the ACCENT database presented at ASCO 2009 (McCleary NAJ et al: Abstract 4010) found no additional benefit with newer oxaliplatin- or capecitabine-based regimens over 5-FU/LV in patients 70 years and older. Similarly, in the recently reported MOSAIC trial (J Clin Oncol 2009;27:3109-3116), adjuvant treatment with oxaliplatin and 5FU/LV did not offer additional benefits in elderly patients, he said.

"These data have prompted some clinicians to decide against using these drugs for their older patients," he said. In addition, on the basis of these studies, German clinical guidelines were revised, and at least 1 European trial was amended to exclude patients older than 70 years, he said. "We felt the need to mine our XELOX database to see if the elderly subgroup benefited in our study," he told journalists.

The study consisted of 1886 patients with stage III colon cancer who were randomly assigned to receive XELOX or 5-FU/LV after surgery. After a median follow-up of 57 months, analysis of 3-year disease-free survival found the effectiveness of XELOX to be similar for both the younger and older age groups.

Overall, 3-year disease-free survival was 71% for the XELOX group and 67% for the 5-FU/LV group, a difference that was statistically significant, Dr. Haller said. Patients younger than 70 years of age (n = 1477) receiving XELOX had a 3-year disease-free survival of 72%, compared with 69% for those receiving the standard regimen (hazard ratio [HR] 0.79; 95% confidence interval [CI], 0.66 - 0.94). For those 70 years or older (n = 409), disease-free survival was 66% and 60%, respectively (HR, 0.87; 95% CI, 0.63 - 1.18), Dr. Haller reported.

In response to a journalist's question, Dr. Haller acknowledged that the confidence intervals crossed 1.0 for the elderly patients and therefore lacked statistical significance. "What we were looking at were trends," he explained. "In the subgroup over 70, we were down to 400 patients, only 200 per arm, so this affects the hazard ratio."

Nevertheless, Dr. Haller believes the data are strong enough to "challenge" the MOSAIC findings. Although NO16968 is one of largest trials in the adjuvant setting, he pointed out that MOSAIC looked at only about 1200 patients and had a larger proportion of low-risk stage II patients, who derive questionable benefit from any chemotherapy.

He added that he anticipates a "vigorous debate" over the findings. "My concern is that, based on one 10-minute presentation at ASCO last year, physicians went back to their practices and stopped treating patients in their early 70s [with XELOX]."

"We conclude that, in addition to 2 other 5-FU/oxaliplatin regimens, XELOX is a new standard of care, regardless of age," he said. "With the oral capecitabine, treatment is more convenient, requires the fewest physician visits of any combination, and spares the patient the need for a surgically implanted access device and infusion pump."

Press briefing moderator Robert Sticca, MD, professor of surgery at North Dakota School of Medicine and Health Sciences in Grand Forks, called the findings "an important addition to our current knowledge" about adjuvant regimens in elderly patients." He added: "This is convincing, as the numbers in the study are very impressive."

Dr. Haller reports relationships with Sanofi-Aventis and Hoffmann-La Roche. Dr. Sticca has disclosed no relevant financial relationships.

2010 Gastrointestinal Cancers Symposium (GICS): Abstract 284. Presented January 22-25, 2010.

HPV TESTING FOR WOMEN OVER 35

January 21, 2010 — For women 35 years of age or older, human papillomavirus (HPV)-based screening is more effective in detecting high-grade lesions and preventing invasive cervical cancer than cytology. However, this screening modality is less effective in younger women because it can lead to overdiagnosis of cervical intraepithelial neoplasia (CIN), which is likely to regress on its own.

In a study published online January 19 in the Lancet Oncology, the researchers conclude that for women older than 35 years, HPV testing should replace cytology in routine screening.

An accompanying editorial notes that HPV testing "shows a great deal of promise to revolutionize cervical cancer screening, especially in developing countries."

In the first part of the trial, a similar number of invasive cancers were detected in the group screened with cytology (n = 9) and that screened with HPV testing (n = 7). But in the second round of screening, there were no invasive cancers detected in the HPV group and 9 in the cytology group, suggesting that HPV-based screening is more effective in preventing progression to cancer because of earlier detection of clinically relevant lesions and treatment of precancers.

"I think that there is sufficient evidence to shift to stand-alone HPV testing," said lead author Guglielmo Ronco, MD, from the Centro per la Prevenzione Oncologica in Turin, Italy. "Cytology should be used only to triage HPV-positive women."

In countries that have population-based screening programs, plausible pilot programs are advisable for making the transition, he told Medscape Oncology. This will help to assess costs and quality assurance and to determine the best method of communicating with women.

"In addition," explained Dr. Ronco, "there is a need for training professionals; the application of appropriate assessment protocols of HPV-positive women is crucial."

Extended Screening Intervals and No Changes for Younger Women

Dr. Ronco does not recommend any changes in the screening paradigm for younger women. "For the moment, they should continue with Pap [tests]," he said. "In the future, research on biomarkers could allow us to identify which CIN2 cases have the potential to progress, so as to avoid overtreatment."

"However," he added, "that is not possible at the moment."

The researchers also conclude that intervals between screenings can be extended with HPV testing, but more follow-up is needed to define how long screening intervals can be safely extended. "In my view, there is already evidence to say that 5-year intervals are safe," said Dr. Ronco.

HPV testing is more sensitive but less specific than cytology for detecting high-grade CIN. As previously reported by Medscape Oncology, when HPV DNA testing is used as a primary screening strategy, followed by cytological triage and repeat screening for persistent HPV type-specific infection, the accuracy of cervical cancer screening could be increased.

HPV testing might also be the most effective method of cervical cancer screening in low-resource settings, where it has been shown to significantly reduce the incidence of invasive cervical cancer (N Engl J Med. 2009;360:1385-1394).

Improved Detection With HPV Test

In the current study, Dr. Ronco and colleagues assessed the efficacy of different HPV-based screening policies and the most appropriate age of application. Their large phase 3 randomized trial consisted of 2 rounds of screening for each of 2 separate recruitment phases. The primary end point was the detection of CIN2/3 and invasive cervical cancers during the first and second screening rounds.

For both trial phases, 47,001 women were randomly assigned to cytology and 47,369 to HPV testing. Of this group, 33,851 women from the cytology group and 32,998 from the HPV-testing group participated in a second round of screening.

During the first phase, HPV-positive women 35 to 60 years old were referred to colposcopy, whereas those 25 to 34 years old were referred to colposcopy only if cytology was abnormal or HPV testing was persistently positive. In the second phase, participants in the HPV group were referred for colposcopy if the HPV test was positive. Each phase consisted of 2 rounds of screening, and all women underwent cytology testing only in the second round.

In the 2 rounds of screening, 18 invasive cancers were detected in the cytology group and 7 were detected in the HPV group (P = .028). All participants who were subsequently diagnosed with invasive carcinoma in the second round had normal cytology in the first round.

For women 35 to 60 years of age, the detection of CIN2 or CIN3 was significantly higher in the HPV group than in the cytology group, with a detection ratio of 2.03 (95% confidence interval [CI], 1.60 - 2.57), during the first screening round. Conversely, in the second round, the relative detection ratio for CIN2/3 was significantly lower in the HPV group than in the cytology group (detection ratio, 0.51; 95% CI, 0.28 - 0.93).

Overall, the ratio between the HPV group and the cytology group was 1.66 (95% CI, 1.34 - 2.06) for the total detection of disease during the first 2 screening rounds, and results were similar for the detection of CIN2 and CIN3.

Among women 25 to 34 years of age, the researchers observed a significant difference between the 2 recruitment phases in the relative detection of CIN3 with HPV testing. In phase 1, the detection ratio between groups was close to 1 during both the first and the second rounds. In the second phase, the detection of CIN3 was much higher in the HPV group than in the cytology group during the first screening round, and lower in the second round.

In phase 2, the total detection ratio of CIN3 for both screening rounds was 2.14 (95% CI, 1.28 - 3.59) for HPV vs cytology. For CIN2, when both phases were pooled, the detection ratio was 4.09 (95% CI, 2.24 - 7.48) for HPV vs cytology in round 1, and 0.64 (95% CI, 0.23 - 1.27) in round 2.

Shows Promise, Research Needed to Optimize Strategy

In an accompanying editorial, Philip E. Castle, PhD, MPH, and Hormuzd A. Katki, PhD, both from the National Cancer Institute in Bethesda, Maryland, write that HPV testing has the potential to "revolutionize" cervical cancer screening. However, there is still room for improvement, they add, by reducing the number of prophylactic treatment procedures and improving the management of equivocal disease of low invasive potential.

"We advocate that clinical management be based on estimating a woman's individual risk of cervical precancer, rather than complex algorithms," they write. "Data from the current study could be used to develop risk estimates to make the promise of more effective and cost-effective cervical cancer prevention a reality."

Dr. Ronco agrees. "We are conducting a full cost-effectiveness analysis and, clearly, prolonged screening intervals will contribute to reducing the overall costs," he said. "Another point is the economic and human cost of diagnostic work-up and treatment."

"Appropriate management of HPV-positive women is crucial and, clearly, these women should not be directly referred to colposcopy," Dr. Ronco added. "Nevertheless, there is a need for research in order to optimize screening, and I expect substantial improvements in this field in the near future."

The study was funded by the European Union; the Italian Ministry of Health; the Regional Health Administrations of Piemonte, Tuscany, Veneto, and Emilia-Romagna; and the Public Health Agency of Lazio. Dr. Ronco reports being an occasional adviser to Gen-Probe. One of his coauthors, Jack Cuzick, PhD, from Queen Mary University in London, United Kingdom, reports being an occasional adviser to Qiagen, Roche, Gen-Probe, and Abbot, another coauthor, Francesca Carozzi, PhD, from the Institute for Cancer Study and Prevention in Florence, Italy, reports being an occasional adviser to Gen-Probe, Abbott, Sanofi, and GlaxoSmithKline. The editorialists have disclosed no relevant financial relationships.

Lancet Oncol. Published online January 19, 2010.

INCREASED ADVERSE EVENTS WITH CEDIRANIB

NEW YORK (Reuters Health) Jan 19 - Most cancer patients who take cediranib develop hypertension, often within days, and many develop proteinuria within a few weeks as well, a phase II trial has shown.

Cediranib is an oral inhibitor of the vascular endothelial growth factor (VEGF) receptor signaling pathway. Because cediranib is more potent than older VEGF inhibitors, lead author Dr. Emily S. Robinson, from Brigham and Women's Hospital, Boston, and co-investigators theorized that it might produce more side effects.

In fact, as they reported online January 7th in the Clinical Journal of the American Society of Nephrology, prevalence rates for hypertension and proteinuria "are much higher" with cediranib than with older VEGF receptor signaling pathway inhibitors.

The open-labeled, single-agent trial took place between 2005 and 2008 and involved 46 women, ages 41 to 77, with recurrent epithelial ovarian carcinoma.

Women stayed in the study for a median of 84 days (range 6 to 544 days). When the study began, cediranib was started at 45 mg, but the initial dose was dropped to 30 mg after the first 11 patients had excessive side effects, mainly fatigue and diarrhea.

Hypertension developed in 67% of the women within 3 days, in 73% within 7 days, and in 87% by the end of the study. Forty-three percent developed Grade 3 hypertension (or worse).

Average systolic blood pressure rose from 120 mm Hg at baseline to 139 mm Hg by day 3, while diastolic blood pressure increased from 73 to 90 mm Hg (p <>

The only predictor of hypertension by day 3 was an age of 56 years or older. In addition, by day 3, women aged 57 or older had a mean systolic pressure increase of 15.9 mmHg, whereas younger women had a mean increase of 7.0 mmHg.

Antihypertensive treatment prevented blood pressure from rising further.

Thirty percent of the patients developed grade 1 or 2 proteinuria, half within 2 weeks and the remainder by 6 weeks. The only significant risk factor for proteinuria was the starting dose of cediranib, the research team reports.

The authors point out that up to 65% of patients with grade 3 or worse hypertension did not develop proteinuria, and in at least one patient, proteinuria developed before the hypertension - suggesting that VEGF receptor signaling pathway inhibitors produce the two side effects via different mechanisms.

The early onset of hypertension might be the result of acute inhibition of VEGF-mediated vasodilation, they add.

Neither hypertension, proteinuria nor a combination of the two was correlated with treatment response, "but the confidence intervals were wide for these analyses," Dr. Robinson and colleagues maintain.

"Clinicians must be vigilant in early detection and management of toxicities of this expanding drug class, especially in older patients," they conclude.

Clin J Am Soc Nephrol 2010.

CD24 FOR COLORECTAL CANCER AND PAM4 FOR PANCREATIC CANCER DETECTION

January 21, 2010 (Orlando, Florida) — Investigators reported promising results from 2 blood tests that could allow for the early detection of colorectal and pancreatic cancers here at the 2010 Gastrointestinal Cancers Symposium.

The annual meeting is cosponsored by the American Gastroenterological Association, the American Society of Clinical Oncology, the American Society for Radiation Oncology, and the Society of Surgical Oncology.

A team from Israel reported that a simple test for blood levels of the CD24 protein is more than 90% sensitive and specific for detecting colorectal cancer, and more than 80% accurate at identifying adenomas.

The CD24 protein is a cell-surface protein and P-selectin ligand that is involved in cell adhesion and metastasis, and has previously been associated with colorectal cancer (CRC) (Gastroenterology. 2006;131:630-639). The aim of the current study was to evaluate CD24 protein expression in peripheral blood lymphocytes (PBLs) from normal subjects and from subjects with CRC or adenomas.

"There is currently no serum-based test that is of sufficient sensitivity or specificity to be useful," said Sarah Kraus, PhD, who presented the findings. Dr. Kraus heads the research laboratory at Tel Aviv Souraski Medical Center in Israel.

"We found that serum CD24 is elevated in the majority of patients with CRC or adenomas. We believe a simple blood test that measures the level of CD24 in PBLs can successfully distinguish healthy subjects from those with neoplasia, and may serve as a biomarker for early detection and surveillance of CRC," she said at a press briefing.

In the study, 150 consecutive subjects underwent colonoscopy. PBLs were isolated from serum samples, and protein extracts were analyzed for these subjects. The same procedure was followed in a second validation trial of 73 consecutive subjects, for whom results were presented here.

CD24 protein levels proved to be sensitive and specific for distinguishing patients with no abnormal findings on colonoscopy from those with CRC and adenoma, Dr. Kraus reported. The protein levels were approximately 2000 OD/mm² in normal subjects and 12,000 OD/mm² in patients with either adenomas or cancer.

The sensitivity for the detection of CRC was 92.3% and for adenomas was 75.0%. The specificity for each, respectively, was 83.8% and 89.2%. In the analysis of both groups combined, sensitivity was 78.4% and specificity was 86.8%.

Pancreatic Cancer Also Detected Early

In another study, an immunoassay showed promise for detecting early-stage pancreatic cancer with a high degree of accuracy. The assay identifies and quantifies serum PAM4 protein, an antigen that is present in nearly 90% of pancreatic cancers but is not typically observed in association with benign lesions or other malignancies.

"We found that the PAM4 protein is quite accurate at identifying patients with pancreatic cancer and, if validated in larger studies, would be a promising tool for detecting the disease when our potentially curative procedures have a better chance for a good outcome," said David V. Gold, PhD, from the Garden State Cancer Center in Belleville, New Jersey. The study was done in collaboration with Johns Hopkins Medical Institutes in Baltimore, Maryland.

The PAM4 monoclonal antibody, also known as clivatuzumab, identifies a unique biomarker with a high specificity for a mucin glycoprotein expressed by pancreatic adenocarcinoma. The protein is not detectable in normal pancreatic cells or in cells of patients with chronic pancreatitis, a condition that can be otherwise hard to distinguish from cancer.

"The PAM4 test has the ability to detect PAM4 antigen in the blood, a protein produced by the tumor that may act as a marker, not only for diagnosis but for targeted therapies," Dr. Gold told media representatives.

The PAM4 enzyme immunoassay was performed on 68 patients who had undergone pancreatic surgery and 19 healthy controls. Its overall sensitivity for detecting pancreatic cancer was 81%; it has 62% sensitivity for detecting stage I pancreatic cancer, 86% sensitivity for stage II disease, and 91% sensitivity for stage III/IV disease, Dr. Gold reported.

"The PAM4 assay detected the overwhelming majority of patients with early-stage pancreatic cancer and 91% with late-stage disease," he said. "PAM4 is very specific for pancreatic cancer. It indicates a high possibility that a patient will have pancreatic cancer."

Dr. Gold predicted that the test will not be used for universal screening, but would best be applied to individuals suspected of having cancer or deemed to be at high risk for pancreatic cancer. This includes people with long-term chronic pancreatitis and diabetes, heavy smokers and drinkers, and those with a family history of hereditary forms of pancreatitis or pancreatic cancer, he said.

Both investigators indicated their assays are being further developed and might be clinically applicable within 2 to 3 years.

Robert Sticca, MD, professor of surgery at North Dakota School of Medicine and Health Sciences in Grand Forks, who moderated the press briefing, explained that these investigations are part of a larger research effort to develop tests that are highly accurate for detecting gastrointestinal cancers at an early stage.

"For pancreatic cancer, the medical profession has long struggled with the treatment of this disease, and these are very exciting data. If we had a test to detect it earlier, we would certainly do much better at managing it and preventing deaths," he said.

"The colorectal data are also preliminary, but if this test can be validated, it should be very beneficial in preventing deaths from this very common form of cancer."

The researchers have disclosed no relevant financial relationships.

2010 Gastrointestinal Cancers Symposium: Abstracts OP162 and 135. Presented January 22-24, 2010.

DO NOT BE TOO AGGRESIVE WITH DCIS

January 21, 2010 — The term carcinoma in the phrase ductal carcinoma in situ (DCIS) is misleading and troubling and ought to be dropped, or at least its dropping should be considered, suggest some recent editorials in major journals.

Both editorials also suggest that DCIS is a possible candidate for management by active surveillance, a treatment strategy of growing importance in prostate cancer in which low-risk patients do not receive radiotherapy or surgery unless they progress to higher risk.

However, unlike in prostate cancer, where active surveillance and a revised sense of what is acceptable management in low-risk patients have been gaining strength for a number of years at multiple North American centers, these potential changes for DCIS are still at an early stage. And, at this point, much of the push for DCIS changes — proposed and actual — appears to be emanating from 1 center in particular, the University of California in San Francisco (UCSF).

The prospect of changing terminology and treatment options in DCIS is complicated in the United States by what 2 different experts described as "hysteria" surrounding breast cancer.

Nevertheless, investigators at UCSF have gone ahead and are investigating what has been called "an important first step in the direction" of active surveillance for DCIS.

The UCSF pilot study involves 40 women with estrogen-receptor-positiveDCIS who received hormonal therapy for 3 months before surgery.The outcomes include change in tumor volume during thisperiod and the identification of cellular components that are predictive of clinical response to therapy.

From preliminary results from 23 women (BMC Cancer. 2009;9:285), the UCSF investigators concluded that "further work is needed to identify which women may be the best candidates for such treatment for DCIS and whether best responders may safely avoid surgical intervention."

However, the pilot study is a step in the direction of active surveillance, because the investigators' "ultimate goal is to identifynonsurgical means of treatment to prevent DCIS progression toinvasive cancer, as pointed out in an editorial in the Journal of the National Cancer Institute (2008;100:228-229).

Regardless of the final findings of this pilot study, Laura Esserman, MD, MBA, professor of surgery and radiology at UCSF and an investigator in the study, thinks the time is now to discuss a change in the approach to DCIS. "We should be demanding change," she told Medscape Oncology.

One Editorial, 1 Forceful Call for Change

Dr. Esserman, who is also director of the Carol Franc Buck Breast Care Center at UCSF, recently made a forceful call for change in the naming and management of minimal-risk cancers and conditions, including DCIS, in an essay that she cowrote with colleagues for the Journal of the American Medical Association (JAMA. 2009;302:1685-1692).

"Minimal-risk lesions should not be called cancer," they write.

The JAMA editorial received widespread media coverage after the chief medical officer of the American Cancer Society made controversial remarks about breast and prostate cancer screening related to the editorial.

Lost in the media swirl was much of the substance of Dr. Esserman's essay. In it, she and her coauthors propose another term for DCIS and other low-risk lesions.

"A more appropriate term, such asindolent lesions of epithelial origin (IDLE) tumors, would helpfocus on systematically studying how to reduce or eliminatetherapeutic interventions while achieving a good outcome," they write.

"Methods exist to identifylow- and high-risk cancers. Tests for prognosis andpredictionof breast cancer are available and provide betterdiscriminatory information than clinical features alone," write Dr. Esserman and colleagues.

With DCIS, the "bulk of what we find is not high grade" Dr. Esserman explained to Medscape Oncology in an interview. She noted that only high-grade DCIS is likely to progress to invasive breast cancer.

"If it doesn't look like high-grade DCIS, we should leave it alone. We would eliminate two thirds of all biopsies if we did," Dr. Esserman said.

She also said that currently "there are sufficient data to stop and rethink the entire approach to DCIS."

Less than 5% of DCIS turns out to be "something else," including invasive cancer, said Dr. Esserman. Because a vast amount of DCIS is overtreated, a new approach to management is required. This has historical precedent, she said. "It's the story of every medical intervention — you treat a condition to the maximum extent and then you must re-evaluate your approach."

James Olson, PhD, distinguished professor of history at Sam Houston State University in Huntsville, Texas, and author of Making Cancer History: The University of Texas M.D. Anderson Cancer Center (Johns Hopkins University Press, 2009), corroborated Dr. Esserman's comments with regard to cancer.

"A key dynamic in the history of cancer treatment has been steady increases in the aggressiveness of treatments in the search for a cure, until a plateau is reached in terms of survival rates, after which there has been a search for less aggressive therapies while preserving existing survival rates," he told Medscape Oncology.

DCIS Was Rare Before Mammography

In the case of DCIS, there is a lack of convincing data that early treatment reduces mortality, Dr. Esserman said. Furthermore, finding DCIS has not led to a decrease in invasive breast cancer rates, she added.

"There are now 60,000 new cases a year of DCIS in the United States. But we haven't seen any drop in invasive cancers, despite treatment of DCIS as if it were early cancer," she explained.

In arguing for a change of approach to DCIS, Dr. Esserman said that screening for precancerous tissue works in some other cancers — it has led to a decrease in cervical cancer — but evidently not in breast cancer.

The burgeoning problem of DCIS is a result of mammography screening, said Dr. Esserman. In the days before widespread mammography, DCIS was rare. In the United States, DCIS incidence has risen from 1.87 per 100,000 in 1973 to 1975 to 32.5 in 2004, according to a recent report published online January 13 in the Journal of the National Cancer Institute by Beth Virnig, PhD, and colleagues. Dr. Virnig is professor of public health at the University of Minnesota School of Public Health in Minneapolis.

Dr. Esserman asked a basic question about breast cancer screening: "Is the purpose of mammography screening to look for DCIS? No," she answered.

"Maybe we shouldn't try so hard to find it — particularly low- and intermediate-grade DCIS. We need to take them out of the screening agenda," she added.

A Second Editorial, Less Forceful

Another call for removing the term carcinoma from DCIScomes from Carmen Allegra, MD, chief of hematology and oncology at the Shands Cancer Center of University of Florida in Gainesville.

Writing in a commentary also published online January 13 in the Journal of the National Cancer Institute, Dr. Allegra says that "strong consideration" should be given to changing the phrase DCIS to eliminate the "anxiety-producing" term carcinoma.

Dr. Allegra also writes that, with improved risk stratification, a watchful-waiting-type approach might be a good strategy for some women with DCIS, a subset "who can be monitored after biopsy in lieu of surgery or other therapies."

Dr. Allegra's proposals are especially notable because they appear in her commentary about a recent national DCIS conference. Dr. Allegra was chair of the conference — the State of the Science Conference on the Diagnosis and Management of DCIS — which was sponsored by the National Cancer Institute and National Institutes of Health.

Ann Partridge, MD, MPH, from the Dana-Farber Cancer Institute in Boston, Massachusetts, who was approached for an independent comment, agrees that the term DCIS is "confusing." Dr. Partridge is the lead author of a study that indicated that women treated for DCIS greatly overestimate the likelihood of recurrence and their risk for invasive breast cancer (J Natl Cancer Inst. 2008;100:243-251).

"Cancer implies that it can spread and be uncontrolled and kill you," she told Medscape Oncology.

" 'This is not a life-threatening problem' — that's the first thing I tell patients," she said.

Dr. Partridge agreed that DCIS is overtreated, but she noted that there is uncertainty about which patients are at highest risk of progressing to invasive breast cancer. "There are ways to risk-stratify, but they aren't great," she said.

Hysteria and Breast Cancer

Until better prognostic and predictive markers come along, overtreatment of DCIS continues. Reports that patients with DCIS are increasingly choosing bilateral mastectomy as their treatment may be the "major clinical dilemma in DCIS today," notes a recent commentary in the Journal of Clinical Oncology (2009;27:5303-5305), as reported by Medscape Oncology.

The extremism that sometimes comes into play in DCIS treatment decision making is a "cultural problem," said Dr. Partridge. "There's a hysteria around breast cancer," she added.

That "hysteria" is one of the main reasons that the strategy of active surveillance, now advancing in prostate cancer, is currently not a viable option for DCIS, said Dr. Esserman.

However, Dr. Partridge noted that men, in general, have more to potentially lose from adverse effects with radical treatments for low-risk prostate cancer than women do with the treatment for DCIS. "There is a big difference between incontinence and impotence and [removing] a piece of breast or a even whole breast," she observed. Thus men, as a group, may be more willing to watch and wait for a time to see if their condition worsens, she suggested.

The researchers have disclosed no relevant financial relationships.

J Natl Cancer Inst. 2010;102:1-9, 170-178. Abstract, Abstract

ONCOTYPE CHANGES TREATMENT DECISIONS

January 21, 2010 — Use of a gene assay that provides prognostic information about the likelihood of distant disease recurrence in low-risk, estrogen-receptor (ER)-positive, node-negative breast cancer affects both physician and patient treatment decision making, according to a new study of 89 patients published online January 11 in the Journal of Clinical Oncology.

Results from the 21-gene recurrence-score assay (Oncotype DX, Genomic Health) led medical oncologists to change their treatment recommendations in 28 patients (31.5%); 24 (27%) patients also changed their treatment decision on the basis of test results.

The study is important because it demonstrates the clinical utility of a recurrence-score assay, according to an accompanying editorial by Joseph Sparano, MD, from Albert Einstein College of Medicine and Cancer Center in the Bronx, New York, and Lawrence J. Solin, MD, from Albert Einstein Medical Center in Philadelphia, Pennsylvania.

"There is compelling evidence regarding the clinical validity of these assays in providing prognostic information about distant disease recurrence," write the editorialists. However, "there is little information about their clinical utility."

Specifically, the editorialists ask: How does the test influence clinical decision making, and do patients benefit from a change in treatment decision?

The study provides answers to both questions, suggest the editorialists and the study authors.

With regard to patient benefit, anxiety and decisional conflict were both significantly lower after recurrence-score results, report the study authors, led by Shelly Lo, MD, from Loyola University Stritch School of Medicine in Maywood, Illinois.

With regard to clinical decision marking, the biggest change caused by the test results was conversion from a medical oncologist's pretest recommendation for chemotherapy plus hormone therapy to a posttest recommendation for hormone therapy alone in 20 patients (22%).

The pretest decisions were made on "standard prognostic features," according to the study authors.

Forgoing chemotherapy in these patients is not an easy decision, suggest the editorialists. "There is unequivocal scientific evidence that adjuvant chemotherapy reduces recurrence and mortality in early-stage breast cancer," they note.

However, as Dr. Lo and her coauthors point out, the likelihood of distant recurrence in this patient population who are treated with tamoxifen is only about 15% at 10 years.

Furthermore, given the toxicities of chemotherapy, "it is critical to match each patient with the appropriate therapy for their individual tumor to maximize benefit while minimizing potential side effects," she and her colleagues write.

The test used in the study, the 21-gene recurrence-score assay, has been validated to quantify the risk for distant recurrence in tamoxifen-treated patients with lymph-node-negative, ER-positive breast cancer and to predict the magnitude of chemotherapy benefit, say the study authors.

Controversy Over Gene Expression Assays

Two multiparameter gene expression assays — the Oncotype DX and the 70-gene assay MammaPrint (Agendia BV) — have been commercially available for early-stage breast cancer for about 5 years, note the editorialists.

Several others are also now commercially available. "Some evidence suggests that these assays provide comparable prognostic information," the editorialists write.

There has been controversy about whether or not these tests have been investigated enough for the benefits and harms of the products to be sufficiently established, as reported by Medscape Oncology.

So far, in clinical practice, the use of the multiparameter tests has been limited to making treatment decisions about adjuvant chemotherapy, note Drs. Sparano and Solin.

One of the reasons for the limitation, say the editorialists, is that there are no "clinicopathologic features" obtained from patients that are predictive of chemotherapy benefit, which is not the case for endocrine therapy (i.e., ER and progesterone status) and anti-HER2 therapy (i.e., HER2 status). Also, clinical practice guidelines recommend adjuvant chemotherapy for most patients with operable breast cancer, even those that are at low risk for recurrence, the editorialists add.

Study Design and Other Results

In the study, Dr. Lo and colleagues prospectively surveyed 17 medical oncologists at 1 community and 3 academic centers. Physicians provided their treatment recommendation and patients provided their treatment choice before and after the recurrence-score results.

The average age of the 89 patients was 55 years. Roughly two thirds had intermediate-grade tumors, and the mean tumor size was smaller than 2 cm.

Before the results of the recurrence-score assay were known, the medical oncologists recommended chemotherapy plus hormonal therapy to 42 patients (47%), hormone therapy alone to 46 patients (51.7%), and either one treatment or the other to 1 patient (1.1%).

After obtaining the results of the assay, medical oncologists recommended chemotherapy plus hormonal therapy to 23 patients (25.8%) and hormone therapy alone to 60 patients (67.4%).

Six cases (6.7%) were considered to be a matter of "equipoise" or uncertainty, and patients had equal options for the various treatments. It is "unclear" what the optimal treatment is for these patients, the researchers explain, noting that they all had recurrence scores that were neither high nor low, but instead were intermediate.

The difference between the mean recurrence score on the assay for a recommendation of chemotherapy plus hormonal therapy and for a recommendation of hormonal therapy alone was significant (29 vs 16; P < .0001).

Dr. Lo and colleagues also report that the posttest treatment recommendation from the oncologists was mostly consistent with the recurrence score.

Chemotherapy was recommended to all 9 patients (100%) with high-risk recurrence scores, to 11 patients (26.2%) with intermediate-risk scores, and to 3 patients (7.9%) with low-risk scores.

Hormone therapy alone was recommended to 33 patients (87%) with low-risk scores, 27 patients (64.3%) with intermediate-risk scores, and to 0 patients with high-risk scores.

Patients reported significantly less conflict about the decision for adjuvant treatment after learning their recurrence score. Patients' mean decisional conflict score before testing was 1.99; after testing it decreased to 1.69 (P < .001).

Patients also had significantly decreased situational anxiety immediately after learning the results of the recurrence-score assay. Mean scores for situational anxiety significantly decreased over time; they were 39.6 before testing, 36.0 immediately after testing, and 34.0 a year after testing (P = .007).

Dr. Lo reports receiving research funding from Genomic Health. Dr. Sparano and Dr. Solin have disclosed no relevant financial relationships.

J Clin Oncol. Published before print January 11, 2010. Abstract, Abstract

SMOKING NSAIDs AND GERD RELATED CANCERS

NEW YORK (Reuters Health) Jan 14 - Non-steroidal anti-inflammatory drug (NSAID) use and smoking each modify the effects of acid reflux, but in different ways, new research shows.

In a study of patients with gastroesophageal reflux disease (GERD), the risk of esophageal or gastroesophageal junction malignancy was higher in smokers but lower in regular users of NSAIDs.

This is not the first time that senior investigator Dr. David Whiteman of the Queensland Institute of Medical Research in Brisbane, Australia, and his colleagues have analyzed the effect of lifestyle factors on GERD-related cancer risks. Previously they reported that obesity and gastroesophageal reflux had synergistic effects on the risk of esophageal adenocarcinoma.

In a paper in the January issue of Gut, they describe a study in which they collected health and demographic data from 365 patients with esophageal adenocarcinoma, 426 with gastroesophageal junction adenocarcinoma, 303 with esophageal squamous cell carcinoma, and 1,580 controls.

Not surprisingly, subjects who had GERD symptoms at least weekly had higher cancer risks than controls, with odds ratios of 6.4 for esophageal adenocarcinoma, 4.6 for gastroesophageal junction adenocarcinoma, and 2.2 for esophageal squamous cell carcinoma. (An association of reflux symptoms with esophageal squamous cell carcinoma has not been shown before, the researchers point out.)

When risk estimates were adjusted for combined exposure to acid reflux and smoking, however, heavy smokers (30 or more pack-years) with weekly GERD symptoms had odds ratios of 12.3 for esophageal adenocarcinoma, 13.5 for gastroesophageal junction adenocarcinoma, and 9.2 for esophageal squamous cell carcinoma, compared to controls. For never-smokers, the corresponding odds ratios were 6.8, 4.3, and 1.8, respectively.

However, when patients with at least weekly reflux symptoms took NSAIDs at least once a week, they had odds ratios of 4.8 for esophageal adenocarcinoma, 3.1 for gastroesophageal junction adenocarcinoma, and 1.0 for esophageal squamous cell carcinoma, compared to controls. In contrast, patients with frequent reflux who did not use NSAIDs had odds ratios of 13.9 for esophageal adenocarcinoma, 8.5 for gastroesophageal junction adenocarcinoma, and 2.8 for esophageal squamous cell carcinoma.

Of note, in patients with frequent GERD, the use of acid suppressants had no effect on cancer risks.

Smoking may increase esophageal cancer risk by promoting inflammation and lowering the tone of the esophageal sphincter, the researchers note.

Gut 2010:59:31-38.