SAN ANTONIO-HER-2/neu testing in DCIS is controversial: Does this test provide useful information? Is HER-2 positivity a predictor of recurrence? Is the test worth the extra cost? Two posters at the San Antonio Breast Cancer Study presented studies that came to different conclusions.
The first suggested that HER-2/neu overexpression was a predictor of invasive recurrence, and the second study said it was not predictive of outcome.
On closer inspection, however, the studies had important differences, suggesting that in fact, HER-2/neu positivity may be associated with recurrence of invasive cancer. The controversy will not be resolved until more definitive studies are conducted.
At this time, HER-2/neu testing should not be considered part of routine care for ductal carcinoma in situ, said Henry Kuerer, MD, PhD, Director of Breast Surgical Oncology Training at the University of Texas M. D. Anderson Cancer Center.
DCIS is currently overtreated, he said, and he and his colleagues are looking at DCIS in a prevention setting-that is, we perform surgery, radiation, and administer tamoxifen only to prevent the development of invasive cancer, he said. The jury is still out on whether or not HER-2/neu overexpression is associated with an independent increased chance of local recurrence with invasive breast cancer following a diagnosis of DCIS.
He noted, however, that once a large database has accumulated on thousands of patients with DCIS, he would not be surprised if HER-2/neu overexpression was found to be associated with an increased risk of invasive breast cancer: After all, HER-2/neu overexpression is associated with more high-grade and estrogen receptor-negative cancer and is definitely associated with poor prognosis.
Details of Studies
The studies presented at the meeting had some important differences. The first, presented by Wedad Hanna, MD, Chief of the Anatomical Pathology Department at Sunnybrook Health Sciences Centre in Toronto, was a retrospective review of 133 cases of pure DCIS treated with breast-conserving surgery only.
HER-2/neu overexpression was found to be associated with an increased risk for recurrence; a univariate analysis showed it was the only independent variable associated with any recurrence, and a multivariate analysis found it was the only independent variable associated with any recurrence and with invasive recurrence.
After adjustment for the size of tumor and margins, the hazard ratio remained steady, showing a strong significant association between HER-2/neu and invasive recurrence. Results were
Dr. Hanna and colleagues suggested that HER-2/neu testing should be considered in all patients diagnosed with DCIS and that HER-2/neu-positive patients should be managed more aggressively than those who are HER-2/neu-negative.
To me, the cost of this test is worth the investment, she said. HER-2/neu testing may be able to spare some patients with DCIS from aggressive therapy and may prevent recurrence.
In an interview, Dr. Hanna discussed why the results of her study should be taken seriously. First, the cohort comprised patients who did not receive any additional treatment beyond surgery and therefore represent the natural course of the disease.
Figure. Henry Kuerer...Image Tools
Second, she noted, there were 41 recurrences in 133 patients (21 were invasive cancers), and these were mainly in HER-2/neu-overexpressing tumors. None of the Grade 1 DCIS patients were HER-2/neu-positive, Dr. Hanna said, and this correlates with the biology of the disease, she said.
Finally, the testing in this cohort was robust; all tumors were tested with immunohistochemistry (IHC) and those that were equivocal were confirmed by fluorescence in situ hybridization (FISH).
At a median follow-up of almost nine years, the median time to recurrence was 2.7 years for DCIS patients and 5.6 years for those with invasive breast cancer. Forty-one of 133 patients had disease recurrence; 21 recurrences were invasive.
She suggested that monoclonal antibodies directed against the HER-2/neu oncoprotein could be considered in the treatment of high-risk cases of DCIS-i.e., were HER-2/neu positive.
Figure. Wedad Hanna,...Image Tools
Different Conclusion
The second study, presented by Mordechai Gutman, MD, Associate Professor of Surgery at Meir Hospital in Kfar Saba, Israel, analyzed clinical and pathologic data from 84 patients treated for DCIS between 1990 and 1998. HER-2/neu expression was determined by analysis of archival tumor tissue.
At a median follow-up of 7.5 years (and up to 14 years in some patients), nine recurrences were reported to occur between 24 and 86 months following definitive surgery. HER-2/neu testing was found to have no effect on outcome.
HER-2/neu overexpression was not significantly associated with the rate of recurrence nor time to recurrence. The study also showed no correlation between other histopathologic markers and recurrence, with the exception of high grade, where a trend was observed toward higher recurrence rates. No correlation was seen between histopathologic markers and time to recurrence, he said.
These findings suggest that there is no need to check HER-2 routinely when a patient has DCIS. This is an unnecessary test that is expensive and gives patients information that they do not need and that may scare them, Dr. Gutman said.
HER-2/neu was correlated with negative markers, such as grading and comedo necrosis, however, and Dr. Gutman did say that the role of HER-2/neu should be studied further.
These findings are important, because today many pathologists test DCIS for HER-2/neu overexpression. HER-2/neu is costly and does not add to predictive value. Patients want to know what HER-2/neu overexpression means, and if they are told it means a bad prognosis, it adds to their distress. It is still unknown whether patients with DCIS who are HER-2/neu positive should receive trastuzumab, Dr. Gutman explained.
Type of Testing Not Known
Dr. Hanna pointed out that the type of testing was unknown in the study reported by Dr. Gutman-that is, whether standardized immunohistochemistry and/or FISH was used.
At the SABCS poster session, she said she asked Dr. Gutman about the type of testing that was used in his study, and he said that he didn't know what the pathologists at his institution used. Also, the study included patients treated with hormonal therapy and it was not clear whether radiotherapy had also been used, whereas her study focused only on patients who received no further treatment after breast-conserving surgery, Dr. Hanna explained.
Also, the number of recurrences in the Israeli study was relatively small, she noted. Dr. Gutman did not reply to requests for a second interview to follow up on these questions after the meeting.
No Change Anytime Soon
Diana Lake, MD, an attending medical oncologist at Memorial Sloan-Kettering Cancer Center called the two studies provocative, but said they are not likely to have any impact on current management of DCIS. We would not use trastuzumab in patients with DCIS, she said.
Dr. Kuerer said that he and others plan to study trastuzumab in HER-2/neu-overexpressing DCIS to determine whether this agent can downstage the disease and prevent the transition to invasive cancer.
M. D. Anderson has begun accruing patients to a feasibility trial of neoadjuvant trastuzumab. Patients with a history of cardiac disease or previous anthracycline treatment are excluded to avoid further cardiac toxicity.
In addition, the National Surgical Adjuvant Breast and Bowel Project is planning a Phase III randomized trial that will enroll about 1,000 patients with DCIS treated with breast-conservation surgery who have negative margins and HER-2/neu-overexpressing disease. Patients will be randomized to receive six weeks of whole-breast irradiation with or without concurrent trastuzumab.
Dr. Lake commented that it has been known for a while that most cases of DCIS overexpress HER-2/neu Studies showing this were done with immunohistochemistry staining, and not with FISH, the current standard.
The fact that the two studies came to different conclusions means that more study of HER-2/neu overexpression in DCIS is needed and these studies should be done using standardized testing, she said.
© 2007 Lippincott Williams & Wilkins, Inc.