Online First/Online Only
Articles/items published ahead of print or only online.
Sunday, March 09, 2014
BY ED SUSMAN
The addition of carboplatin to standard chemotherapy for women diagnosed with triple-negative breast cancer appears to significantly increase the percentage of women who achieve a pre-surgery pathologic complete response, researchers reported at the San Antonio Breast Cancer Symposium.
In the Cancer and Leukemia Group B (CALGB) 40603 trial, 46 percent of the 212 women who did not receive carboplatin achieved a pCR – the primary endpoint of the trial -- compared with 60 percent of the 221 women who did receive carboplatin, reported William M. Sikov, MD, Associate Professor of Medicine at Warren Alpert Medical School of Brown University.
For the secondary endpoint, 41 percent of women who did not receive carboplatin achieved a pCR in the breast and axillary lymph nodes compared with 54 percent of patients who received the platinum-containing drug.
“In Stage II-III triple-negative breast cancer, the addition of carboplatin to standard neoadjuvant chemotherapy significantly increased the pathologic complete response rates in the breast and in the breast and axilla,” he said at a news conference. “These preoperative data are encouraging.
“Should carboplatin be routinely added to neoadjuvant chemotherapy for stage II-III triple-negative breast cancer? With the caveat that we do not have long-term results -- we don’t know if increasing the pathologic complete response rate will result in significant improvements in recurrence free or overall survival -- I think that if you are looking in the neoadjuvant setting, the answer would be yes.”
The trial simultaneously found benefit in treatment of triple-negative breast cancer with the anti-angiogenesis drug bevacizumab, he added. Among the 218 women who were not treated with bevacizumab, 48 percent achieved a pCR in the breast compared with 59 percent of the 215 women who did receive bevacizumab.
Among the women who were not treated with bevacizumab, 44 percent had a pCR in the breast and axilla compared with 52 percent of the women who received bevacizumab.
“The addition of bevacizumab significantly increased the pathologic complete response rate in the breast but not in the breast and axilla,” Sikov said. “Because bevacizumab increases toxicities, I don’t think it should be routinely added to neoadjuvant chemotherapy.
“Triple-negative breast cancer accounts for about 20 percent of invasive breast cancers in the United States. We know this is an aggressive type of cancer and that it is more common in blacks, Hispanics, younger women, and BRCA1 gene mutation carriers.
“Although there have been significant advances in treating women with this disease in the past 20 years, we know that women with early-stage triple-negative disease remain at higher risk of early recurrence and that prognosis for patients with advanced-stage disease is very poor,” he added.
In advanced triple negative -- estrogen receptor-negative, progesterone receptor-negative, and HER2-negative – disease, the median overall survival is less than one year.
Sikov noted that at the 2012 San Antonio Breast Cancer Symposium, a meta-analysis was reported showing that with standard neoadjuvant chemotherapy about 34 percent of patients diagnosed with triple-negative breast cancer achieve a pathological complete response – i.e., with no viable residual disease in either the breast or the lymph nodes at surgery: “These patients do significantly better than women who have residual disease at surgery. One of the clinical goals of our study was to try to increase the percentage of patients to achieve a pathologic complete response.”
In the CALGB 40603 study, a control group of women with triple-negative disease were treated with paclitaxel at 80 mg/m2 weekly for 12 weeks followed by dose-dense doxorubicin and cyclophosphamide for four cycles -- a regimen that has been shown to benefit women diagnosed with triple-negative breast cancer. A second group of women were given the same chemotherapy regimen along with bevacizumab at a dose of 10 mg/kg every two weeks for a total of nine doses.
“We decided to investigate bevacizumab based on improvements in time to progression in patients with metastatic triple-negative disease,” he said.
A third group of women received the same chemotherapy regimen, but with carboplatin dose to an area under the curve (AUC) of six every three weeks for four cycles.
A fourth group of women received chemotherapy, plus carboplatin plus bevacizumab. There was one on-treatment death in this protocol.
Sikov explained that the study was designed as a Phase II trial with a 2-by-2 randomization. The trial was designed to study the two arms of patients who received carboplatin with the two arms who did not receive the platinum-containing drug; and the two arms that contained bevacizumab with those women who did not receive the biologic agent.
“The study was not powered to compare the individual treatment arms,” he said. The trial was not powered to determine statistically significant differences in progression-free survival or in overall survival.
After the chemotherapy was administered, women then were treated with surgery and/or radiotherapy at the discretion of the treating physician. The trial did not include any planned adjuvant systemic therapy -- also leaving that decision to the treating physician.
Women were eligible if they were diagnosed in clinical Stage II-III disease; had 10 percent or less estrogen plus progesterone receptor positivity and were HER2-negative. Patients also could not have a contraindication to bevacizumab. They underwent baseline evaluation including breast magnetic resonance imaging or breast ultrasound imaging for tumor measurement.
‘Good Representation of Black Patients’
A total of 443 women were enrolled in the study over a period of 36 months. About 60 percent of the women were between the ages of 40 and 59; about 72 percent of the participants were white, and 22 percent were black: “We reached out to the black community to enroll in the study, and had a good representation of black women in the trial,” Sikov said.
About 68% of the women were diagnosed in clinical Stage II, and 32 percent were in clinical Stage 3. About 66 percent were diagnosed with T2 tumors, and 20 percent had T3 tumors. About 45 percent of the women were clinical node-negative, and 55 percent were node positive.
The toxicities reflected the known profiles of the treatments, he said: Neutropenia was common, with 22 percent of the control arm patients experiencing neutropenia, increasing to 27 percent with bevacizumab added. This increased even further to 56 percent if carboplatin was added to chemotherapy, and to 67 percent if the patient was given bevacizumab plus carboplatin plus the standard chemotherapy.
Thrombocytopenia occurred in the range of three to four percent, unless carboplatin was added to treatment and then 20 percent of the women experienced thrombocytopenia; the combination of chemotherapy, carboplatin, and bevacizumab was associated with a 26 percent rate of thrombocytopenia. “However, only the groups that received both carboplatin and bevacizumab experienced higher rates of febrile neutropenia,” Sikov said.
Hypertension occurred in about 12 percent of the women in the bevacizumab arms, but was uncommon in women not taking the anti-angiogenesis drug. Bleeding and thrombosis were also seen more other with bevacizumab- treated patients.
“When we turn to serious adverse events, there were increases with carboplatin but the most concerning adverse events were seen among those women taking bevacizumab,” he said. Those adverse events were mainly infections and thrombosis and a higher rate of postsurgical complications, even though bevacizumab therapy had ended at least six weeks prior to surgery.
The patients who received both drugs had the highest pathologic complete response rate, 67 percent, in the breast; and the highest pathologic complete response rate, 60 percent, in the breast and axilla. However, Sikov said, treatment with both drugs did not show a synergistic effect.
Critical Questions Still Unanswered
Sikov said the results of CALGB 40603 trial and the earlier GeparSixto trial provide evidence that carboplatin improves outcomes in triple-negative breast cancer, but still leaves critical questions unanswered, including the optimal dose and schedule to enhance efficacy and reduce adverse events.
If a clinician believes a women with triple-negative breast cancer is a candidate for neoadjuvant chemotherapy based on tumor size or a desire to improve the chances for breast-conservation surgery, “then I think it makes sense to add carboplatin, and that you can comfortably do so with acceptable additional toxicity,” Sikov said.
The lead investigator of the GeparSixto trial, Gunter von Minckwitz, MD, PhD, Chairman of the German Breast Group, cautioned, though, about comparing his trial and the CALGB trial: “The regimen for GeparSixto was completely different. We had different combinations; we used an 18-week schedule instead of 24 weeks.
“We had to reduce the dose of carboplatin from AUC 2 to AUC 1.5. We have now done further safety analyses of the study and we have seen that the AUC 1.5 dose is much better tolerated.”
Sikov was less enthusiastic about use of bevacizumab. “There now have been three studies that indicate that the use of bevacizumab increases the pathologic complete response rate, but none of these trials are powered to look at progression-free survival or overall survival, and there are increases in toxicities such as febrile neutropenia, bleeding, clotting, and post-surgical complications,” he noted.
“Correlative studies are pending to see if we can identify markers predictive of response -- or resistance -- to the standard and investigational regimens.”
The trial received funding from the National Cancer Institute, Genentech, the Breast Cancer Research Foundation, and the American Recovery and Reinvestment Act of 2009.
Sunday, March 09, 2014
BY ED SUSMAN
Women diagnosed with breast cancer who are able to achieve a pathological complete response (pCR) to treatment with HER2-receptor antagonists appear to have significantly longer overall and event-free survival, according to a study reported at the San Antonio Breast Cancer Symposium.
Presenting the results of the Phase III Neoadjuvant Lapatinib and/or Trastuzumab Treatment Optimization (NeoALTTO) trial, Martine Piccart-Gebhart, MD, PhD, Chair of the Breast International Group (BIG) and Professor of Oncology and Director of Medicine at Jules Bordet Institute in Brussels, Belgium, explained that she and her colleagues had previously reported that the combination therapy resulted in twice as many patients having a pCR.
And now in the updated report, achievement of a pCR showed benefits in event-free survival. Of the 137 patients who achieved pathological complete responses, there were 19 relapses during the four years of follow-up – a four-year event-free survival rate of 86 percent.
But for the 274 patients who did not achieve a pCR, there were 79 events, or a 72 percent event-free survival rate at four years. The difference, she reported, was highly significant (P=0.0003), and translated to a 63 percent reduction in the risk of having an event.
The study assessed whether use of two anti-HER2 drugs would induce higher rates of pCR in women with HER2-positive breast cancer that were two centimeters or larger compared with use of single HER2-receptor blockade with trastuzumab or lapatinib.
For the 427 women who underwent surgery, events were defined as primary breast cancer recurrence, second primary cancers, or death. For the 28 who chose not to have surgery, an event was defined as regional or distant disease progression during neo-adjuvant treatment or death.
Much of the difference in outcome related to women who were diagnosed with HER2-positive/hormone receptor (HR)-negative breast cancer, Piccart-Gebhart said. In these women, those who achieved a pathological complete response had an 85 percent event-free survival rate at four years compared with 65 percent for patients who did not achieve a pCR.
About 87 percent of patients with HER2-positive/HR-positive breast cancer who achieved a pCR had a four-year event-free survival compared with 78 percent of women who did not have a complete response.
“The results provide further evidence that HER2-positive/HR-negative and HER2-positive/HR-positive breast cancer subgroups are two different diseases,” Piccart-Gebhart said.
In the original report of the study three years ago, the researchers observed that about 61 percent of women with HR-negative breast cancer were able to achieve a pathological complete response with dual HER2-receptor blockade; overall 50 percent more women achieved a pCR with dual blockade.
“Our new analysis shows that the improved pathologic complete response rates seen in the hormone receptor negative/HER2-positive subgroup seem to translate into better long-term outcomes for patients,” she said.
“I believe that these results if confirmed in the large ALTTO [Adjuvant Lapatinib and/or Trastuzumab Treatment Optimization] adjuvant trial will impact the process of drug development in the field of early HER2-positive breast cancer.”
Those results in the trial that enrolled 8,300 women are expected later this year, she said.
‘Very Valid Surrogate Endpoint’
Speaking at a news conference that featured the NeoALTTO study, the moderator, Jennifer Litton, Associate Professor of Breast Medical Oncology at the University of Texas MD Anderson Cancer Center, said, “What Dr. Piccart-Gebhart has shown is that pathological complete response continues to be a very valid surrogate endpoint and continues to be valid for the development of new drugs so that we can get them to patients quicker, with smaller numbers of patients involved and less cost.
“For me, pathological complete response standing out as a strong clinical endpoint is the continued importance of this study,”
Piccart-Gebhart said that when she and her colleagues analyzed the results, a similar pattern was observed with overall survival. The women who achieved a pCR had a 94 percent overall survival rate at four years compared with 87 percent for the patients who did not achieve a pCR, a relative risk reduction in death of 65 percent.
There was little difference in overall survival among the HR-positive breast cancer patients – 96 versus 93 percent for those who did not achieve a pCR, but there was a major difference among the women with HR-negative breast cancer. Women who had a pCR also had a 94 percent overall survival rate at four years compared with 80 percent for the women with HR-negative disease who were unable to achieve a pCR.
“Patients who achieved a pathological complete response had significantly better event-free survival and overall survival compared with patients who did not achieve a complete response irrespective of the treatment arm,” she said.
Overall survival by treatment arm showed a 22 percent relative risk reduction in death for those women treated in the dual HER2-blockade arm – a 95 percent survival rate over four years, compared with 93 percent for patients treated with lapatinib and 90 percent for those on trastuzumab. The difference, though, failed to achieve statistical significance, she said.
Piccart-Gebhart said the NeoALTTO trial was powered to detect pCR differences between the treatment arms, but underpowered to detect moderate differences in event-free and overall survival between the treatment arms. Another follow-up analysis will be performed in two and a half years, she added.
No adverse events were observed that were inconsistent with the known safety profile of lapatinib and/or trastuzumab, she said.
The study, managed under the overall umbrella of the Breast International Group, was supported by funds from GlaxoSmithKline; and Piccart-Gebhart disclosed that she has received honoraria from Roche and that her institution has received research funding from GlaxoSmithKline.
Sunday, March 09, 2014
BY ROBERT H. CARLSON
Children with low-grade gliomas are known to have superior five- and 10-year overall survival rates compared with adults with the disease. Now, a new study confirms that adult survivors of pediatric low-grade glioma also have a low incidence of glioma-related death, even 20 years after diagnosis.
In the study, now available online ahead of print in Pediatric Blood & Cancer (DOI: 10.1002/pbc.24958), the overall cancer-specific survival rate was 87 percent, in the analysis of the National Cancer Institute's Surveillance Epidemiology and End Results (SEER) database of 4,040 children.
“This highlights that pediatric low-grade gliomas are very unlikely to undergo malignant transformation resulting in tumor-related death, which contrasts with the natural history of adult low-grade gliomas,” said the researchers – senior author was Peter Manley, MD, a pediatric neuro-oncologist at Dana-Farber/Boston Children’s Cancer and Blood Disorders Center.
But the analysis also showed that children treated with radiotherapy had a reduced overall survival and greater risk of disease death than those not receiving radiotherapy -- approximately 70 percent at 20 years.
Cranial radiotherapy has largely fallen out of favor, but it was a common component of regimens for pediatric low-grade glioma in the 1970s and 1980, and 18 percent of the patients in the cohort received radiation as part of their treatment.
Interestingly, tumor grade did not have an impact on survival. “Grade 1 did better [than grade 2], but even grade-2s had overall survival rates well over 70 percent [at 20 years],” Manley said. “That was very eye opening to us.”
Children treated with radiation therapy had a reduced overall survival and greater risk of disease death compared with those not treated with radiation therapy, with a hazard ratio of 3.9. This held true whether surgical resection was total or partial.
Patients with subtotal or no resection had a small but statistically significant increase in risk of disease-related death compared with those with total resection on multivariate analysis, with a hazard ratio of 1.5, but this was not statistically significant, he said.
But children who had subtotal resection and no radiation still had a significantly superior outcome compared with those who had total resection who underwent radiotherapy.
Although the SEER database did not show the incidence of radiation-induced secondary malignancies, the hazard ratio of death due to non-glioma-related causes was 2.4 in patients treated with radiation, suggesting that radiation-induced mortality may have accounted for some of the deaths that were not directly due to glioma, the authors reported.
Deaths were classified simply as “death due to glioma” or “death not due to glioma.” Other causes of death such as secondary malignancies and radiation-induced tumors, or radiation necrosis, were not classified as deaths due to glioma.
The analysis covered pediatric patients up to 19 years of age, who were diagnosed with a grade 1 or 2 glioma between 1973 and 2008; median age at diagnosis was nine.
Gross Total Resection Not Necessary
Manley, who is also Director of the Stop & Shop Family Pediatric Neuro-Oncology Outcomes Clinic, which focuses on survivorship issues, said that whether resection was total or partial did not affect overall survival.
“It appears that low-grade gliomas in children have a different biology than in adults -- children don't transform into more malignant tumor as adults do,” he said in an interview. “The difference in the biology of adult and pediatric biology low-grade gliomas is one of the most important thing to understand.”
He noted that the study's first author, Pratiti Bandopadhayay, MD, PhD, a pediatric neuro-oncologist at Dana-Farber, is now investigating the molecular basis of gliomas to understand the differences between adult and pediatric disease.
Treatment for pediatric low-grade glioma varies, Manley said. Observation is appropriate for some patients after a gross total resection, while those with residual disease or progression might be retreated with chemotherapy. The most common regimens are vincristine-carboplatin, and the “UCSF” regimen of 6-thioguanine-procarbazine-CCNU-vincristine, he said.
Sicker Patients More Likely to Receive Radiotherapy?
Asked for her opinion for this article, Yasmin Khakoo, MD, Program Director in the Pediatric Neuro-Oncology Fellowship and Associate Attending Pediatric Neurologist at Memorial Sloan Kettering Cancer Center, said it is not surprising that patients who received radiation did not do as well in the long term -- possibly due to sicker patients getting radiation, to different radiation techniques, and perhaps because second malignancies were not as amenable to treatment as they are today.”
A limitation of the study, she said, is that it does not indicate what patients died of, which the authors do acknowledge in the paper. The researchers were also unable to determine progression-free survival from the
“The SEER data base is good, but is not the most robust because it does not include the data we collect now to assess patient outcome, and it does not include how many patients received other modalities such as chemotherapy,” Khakoo said.
“And obviously if you are saying that patients who were radiated had a poor outcome you may be self-selecting for patients who had more aggressive disease. They may have had as many as six chemotherapy regimens before they were randomized to radiation or not radiation and might not have survived the tumor to begin with.”
The data may not be entirely relevant today considering the quality of radiation given in 1973, she noted. “Today's more focused modalities would have fewer side effects such as radiation vasculopathy and radiation necrosis which were more frequent in the past. We see patients today who were radiated back then and now have side effects.”
One small group of patients was not discussed in the study – she pointed out: those with neurofibromatosis (NF). “In the 1970s, patients with NF who had optic pathway tumors or low-grade gliomas were receiving radiation, and we now know that patients with NF have a high second malignancy potential as well as vasculopathy potential with radiation, so we reserve radiation for refractory patients. We don't know what percentage of patients in this study had NF.”
Another question concerned classifications: “We now have subclassifications of grade 1 and grade 2 based on molecular as well as other histopathological features that we didn't have in the ‘70s and ‘80s,” she said. “For example, we have juvenile pilocytic astrocytoma [JPA], which is grade 1, but recently we've been noticing a new classification called a pyelomyxoid astrocytoma, which in the 1970s looked the same as a JPA, but we know now it's a subset of JPA that is more aggressive. “
JPAs have molecular signatures such as BRAF mutation, and now there are drugs that target that lesion, Khakoo said. “We wouldn't irradiate a patient now with a BRAF mutation because there is better targeted therapy.
She said the study is interesting but probably will not change practice because clinicians generally try to avoid cerebral radiation in pediatric patients. “We learned a lot from those studies done earlier about the effects of radiation.”
Friday, March 07, 2014
BY SARAH DIGIULIO
A new report from the American Cancer Society estimates that among U.S. children and adolescents (age 19 and younger), 15,780 new cases of cancer will be diagnosed this year and 1,980 deaths from cancer will occur.
The article, now available online ahead of print in CA: A Cancer Journal for Clinicians (DOI: 10.3322/caac.21219), and also disseminated as a Special Section of the Society’s “Cancer Facts & Figures 2014” (cancer.org/statistics), summarizes the most recent and comprehensive data on cancer incidence, mortality, and survival from the National Cancer Institute, the Centers for Disease Control and Prevention, and the North American Association of Central Cancer Registries.
“We’ve been frustrated with the aggregated numbers because they don’t tell the full picture of where we’re making progress and where we still have shortfalls,” Rebecca Kirch, the ACS’s Director of Quality of Life and Survivorship and Cancer Control and an editor of the report, explained in an interview.
“This report is intended as a roadmap to help prioritize where extra attention needs to be paid for cancers that haven’t seen progress. It is essential to help us plan out and prioritize the research agenda and the advocacy agenda—so that we can develop better clinical outcomes, both in terms of cure rates and quality of life measures.”
She noted that although ACS’s annual report always includes data on childhood and adolescent cancer survivors, this is the first report in approximately a decade to include incidence and mortality rates broken down by cancer subtype, as well as information about these survivors’ long-term and late effects from treatments.
“Progress in childhood cancer has been dramatic for some sites, but we cannot let that blind us from the fact that progress has been disappointingly slow for other sites, and that cancer remains the second leading cause of death in children,” Otis W. Brawley, MD, the ACS’s Chief Medical Officer, said in a news release. “There is much work to be done to improve outcomes, to reduce side effects, and, we hope, to understand more about the molecular events that lead to childhood cancer in order to come up with ways to prevent or detect it early.”
Key statistics and findings from the report are the following:
The annual incidence of cancer from birth to age 19 is 18.8 per 100,000 individuals;
Approximately one in 285 children will be diagnosed with cancer before age 20;
Today about one in 530 young adults between age 20 and 39 is a childhood cancer survivor;
The most common cancers in children 14 and younger are acute lymphocytic leukemia (26%), brain and central nervous system (CNS) cancers (21%), neuroblastoma (7%), and non-Hodgkin lymphoma (6%);
The most common cancers in children 15 to 19 are Hodgkin lymphoma (15%), thyroid carcinoma (11%), brain and CNS cancers (10%), and non-Hodgkin lymphoma (6%);
Non-Hispanic white (white) and Hispanic children have the highest incidence rates for childhood and adolescent cancers;
Although incidence rates are substantially lower for non-Hispanic black (African American) children and adolescents than for whites and Hispanics, death rates are similar due to lower survival rates in African Americans;
The overall incidence of pediatric cancer in the U.S. from 1975 to 2010 increased by an average of 0.6 percent per year, with incidence rates increasing specifically for four cancer types (acute lymphocytic leukemia, acute myeloid leukemia, non-Hodgkin lymphoma, and testicular germ cell tumors);
Incidence rates decreased for Hodgkin lymphoma and remained stable for other cancers;
Overall five-year survival rates across all cancer sites increased from 63 percent between 1975 and 1979, to 83 percent between 2003 and 2009; and
Death rates for all childhood and adolescent cancers combined steadily declined from 1975 to 2010 by an average of 2.1 percent per year (resulting in an overall decline in mortality of more than 50%).
The report includes additional information on risk factors, symptoms, treatment, and long-term and late effects for the most common cancers types in children and adolescents by site. Data in the report also show that some cancers still do remain deadly, such as diffuse intrinsic pontine glioma—for which the median survival time after diagnosis is less than one year, Kirch added. “We have no available treatment—nothing we can do.”
Other challenges detailed in the report are that only a small proportion of childhood cancers have known or preventable causes; and early diagnosis of cancer in children is often difficult because of the similarity of symptoms to more common diseases in children. The clinical scene for children and adolescents with cancer looks very different than it does for adults—from diagnosis onward, Kirch said. “We don’t have a lot of information yet about what’s causing some of these cancers. And, we don’t have screening tools for the types of cancers that kids get like we do for adults.”
Another challenge: Even though advances in survival have been made for many types of cancers, many children treated for cancer still have high risks of long-term health issues associated with treatment (see OT’s coverage of unmet survivorship needs of childhood cancer survivors in the 9/25/13 issue).
In acute lymphocytic leukemia there has been progress in treatment and improvements in cure rates, but emerging literature shows that long-term side effects are still problematic for these patients, Kirch explained. “The toxicities of our treatment are a significant concern. We still need to make sure people’s lives, across the life course, are pain- and symptom-free to the extent that is possible.”
Remembering the 17%
Asked to comment about the report for this article, Leslie L. Robison, PhD, Chair of the Department of Epidemiology and Cancer Control and Associate Director for Cancer Prevention and Control at St. Jude Children’s Research Hospital, noted that the report highlights the improvements in survival rates for childhood and adolescent patients with cancer in the past four decades. “But, it is important to also focus on the work that remains to cure the other 17 percent of patients [the report notes the overall survival rate across all cancers for this patient population is 83%], and to understand the long-term consequences of treatment in the cured population who have 60 to 70 years of life ahead of them.”
LESLIE L. ROBISON, PHD
Another key take from the report, he added, is the need for prevention research: “This report emphasizes—after decades of research—how little we know about the causes of the cancers that occur in this young population.”
Thursday, March 06, 2014
BY KURT SAMSON
Two premalignant abnormalities on breast biopsies that for decades have been viewed as having very different behaviors appear to have the same long-term cancer potential and progression risk, according to new data from researchers at the Mayo Clinic.
It is generally believed that atypical ductal hyperplasia (ADH) is a direct precursor of breast cancer and therefore portends breast cancer in the same breast, while atypical lobular hyperplasia (ALH) has an equal risk of cancer in both breasts and may not be a direct precursor of breast cancer. In the study, however, published in Cancer Prevention Research (2014;7;211-217) and led by
Lynn C. Hartmann, MD, Professor of Oncology, cancer was twice as likely to occur in the breast having the biopsy, and this was true for both ADH and ALH.
Moreover, the findings run counter to current understanding that ALH primarily leads to lobular cancer. Instead, the researchers discovered that ALH was associated predominantly with later ductal cancers of the breast -- also similar to cancers after ADH.
Both types resulted in invasive ductal cancers, and 69 percent were of intermediate or high grade. Further, about 25 percent had spread to the lymph nodes. Although the numbers were not statistically significant, the study also found that more ductal cancer in situ occurred following ADH than after ALH – 25 percent versus 13 percent. In addition, women with ADH were more likely to have an ipsilateral breast cancer in the first five years after diagnosis than in later years.
The study included 698 women with biopsies showing atypical hyperplasia: 330 with ADH, 327 with ALH, and 32 with both types. Over an average follow-up of 12.5 years, 143 women developed breast cancer, but at the same rates regardless of the type of atypia. And in confirmation of the high-risk nature of atypia, 29 percent of these women developed breast cancer by 25 years of follow up.
“Our observations do not support present assumptions that ADH and ALH have substantively different behaviors,” Hartmann said. “Ours is the first report with sufficient numbers of both types of atypia and long-term follow-up that was able to examine the type of subsequent breast cancer, when these occurred, and in which breast.
“We showed that even though the two types of atypia look different histologically, they behave quite similarly in terms of later breast cancers in patients.”
Importantly, she added, many have believed that women who have both atypia and a positive family history of breast cancer have a risk that is substantially higher than that of someone with atypia without a family history.
“In our large study of women with atypia, we compared the later breast cancer risk of 257 women who had a family history and 372 without a family history, and saw no difference in their later risk. How can this be? In fact, one cause for the development of atypia is having a family history – so the risk associated with family history is already accounted for by the presence of the atypia itself.
Better Risk Assessment
More than one million women undergoing a breast biopsy in the United States each year have benign findings, and about 10 percent of these biopsies reveal atypical hyperplasia, a premalignant finding with a proliferation of abnormal cells, which have some but not all the features of a breast cancer.
“Most clinicians have viewed ADH as a direct precursor to breast cancer, arguing that it requires complete surgical excision, while others have maintained that ALH serves as an indicator of heightened and equal risk of breast cancer across both breasts,” Hartmann explained.
“Whether or not ALH on a core biopsy requires surgical excision remains a topic of research investigation. Moreover, some experts have argued that women with atypia develop ‘better risk’ breast cancers -- meaning low-grade cancers with a good prognosis.”
Improve Clinical Management
Findings from this study could improve clinical management of patients with breast tissue abnormalities, she said.
“Given the high-risk nature of atypia, with a risk of breast cancer of close to 30 percent at 25 years after biopsy, we recommend that such women be seen at a breast center for recommendations about surveillance and preventive therapy options.
“What we are providing is absolute risk data for a sizable cohort of women with atypia. The medical community has known for years that the relative risk of breast cancer for women with atypia is four times greater than for other women, but that does not translate into an actual number for an individual with abnormal biopsy findings. Hopefully our paper will raise awareness of the risk and encourage the most sophisticated surveillance, likely to include MRI screening, and the use of chemoprevention drugs such as tamoxifen.”
Many practitioners already encourage women with atypia to consider chemoprevention drugs, she added.
“I think that women are increasingly aware that these are options, but nevertheless, many women remain more concerned about the possible side effects associated with these agents. They also need good data about the breast cancer risks.”
She and her colleagues are now preparing a review article that may help to guide development of new guidelines reflecting their findings, she said.
Asked for his opinion for this article,
Stuart J. Schnitt, MD, Professor of Pathology at Harvard Medical School and Director of Anatomic Pathology at Beth Israel Deaconess Medical Center, said some of the observations in this study confirm data from previous reports but that there are also new data that add to what is understood about atypical hyperplasia.
In a 2007 study published in Cancer (2007;109:180-187), he and his colleagues (first author was Laura C. Collins, MD) analyzed the magnitude and laterality of breast cancer risk based on the histologic type of atypical hyperplasia among women in the Nurses’ Health Study. Like the Mayo group, they found that about 60 percent of subsequent cancers occurred in the ipsilateral breast. However, that study also suggested that the risk associated with ADH and ALH depended upon the menopausal status of the patient at the time of the benign breast biopsy, something that was not addressed in the Mayo study.
In addition, in a 2003 study in The Lancet (2003;361:125-129), researchers at Vanderbilt University, led by David L. Page, MD, looked at ALH as a unilateral predictor of breast cancer risk in 250 women in the Nashville Breast Cancer Studies cohort.
Schnitt noted that comparisons among these studies is somewhat difficult since the Nurses’ Health Study benign breast disease analysis was based on a nested case-control study whereas the Mayo and Vanderbilt studies are cohort studies. “Nonetheless, despite the methodologic differences in these three studies, there are many similarities in the findings regarding atypical hyperplasias,” he said.
“We now recognize that both ADH and ALH are associated with a substantially increased risk of breast cancer, that both lesions appear to represent direct cancer precursors as well as markers of increased risk. I think the similarities rather than the differences between ADH and ALH is one of the main take-home messages from the Mayo study. The time has come to find ways to better stratify risk and identify which women with atypical hyperplasia are at highest risk.”
And, although treatment and monitoring guidelines from the American Society of Clinical Oncology or the American Cancer Society would be helpful, the extant data are not sufficiently robust to permit consensus guideline recommendations, he said.
“Guidelines need to be evidence-based, and the available evidence from individual studies is based on very small numbers in some subgroups. Perhaps attempting to combine data from the Nurses’ Health Study, the Vanderbilt Study, and the Mayo Clinic study would be a worthwhile undertaking and provide more statistical power than any of these studies alone.”