| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Cancer Therapy: Clinical |
Authors' Affiliations: 1 Institute of Cancer Research and Royal Marsden NHS Foundation Trust, 2 Department of Medical Genetics, 3 Oncology Unit, Guy's and St. Thomas NHS Foundation Trust, 4 St. George's Hospital Medical School and St. George's Hospital, 5 Gray Cancer Institute, Northwood, London, United Kingdom; 6 Familial Cancer Service, Westmead Hospital, Sydney, Australia; 7 Division of Genetic Medicine, Vanderbilt University, Nashville, Tennessee; 8 Birmingham Oncology Centre, University Hospitals NHS Trust, Birmingham, United Kingdom; 9 Academic Unit of Medical Genetics and Regional Genetics Services, St. Mary's Hospital; 10 Christie Hospital, Withington, Manchester, United Kingdom; 11 Princess Anne Hospital, Wessex Regional Genetics Centre, 12 Department of Surgery, Royal South Hants Hospital, Southampton, United Kingdom; 13 Department of Radiation Oncology, University of Michigan, Ann Arbor, Michigan; and 14 Strangeways Research Laboratory, Worts Causeway, Cambridge, United Kingdom
Requests for reprints: Susan Shanley, Cancer Genetics Unit, Orchard House, Royal Marsden NHS Foundation Trust, Downs Road, Sutton SM2 5PT, United Kingdom. Phone: 44-208-661-3375; Fax: 44-208-7770-1489; E-mail: Susan.Shanley{at}rmh.nhs.uk.
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
Experimental Design: Acute and late radiation effects were evaluated in a retrospective study of 55 BRCA1 and BRCA2 mutation carriers treated with radiotherapy for breast cancer at four centers between 1983 and 2002. Individual matching with controls who had sporadic breast cancer was undertaken for age at diagnosis, time since completion of radiation, and treatment variables. Detailed assessments were undertaken by one examiner. Median follow-up was 6.75 years for carriers and 7.75 years for controls. Rates of late events (rib fractures, lung fibrosis, necrosis of soft tissue/bone, and pericarditis) as well as LENT-SOMA scores and clinical photography scores of breast size, shape, and skin telangiectasia were the primary end points.
Results: No increase in clinically significant late toxicity was seen in the mutation carriers.
Conclusions: These data add substantial weight to the evidence that the outcomes in the treated breast from radiotherapy in women with BRCA1 or BRCA2 mutations are comparable with those in women with sporadic breast cancer.
In the absence of data from prospective studies on outcomes, the current study was based on that of Pierce et al. (4), to be able to compare data sets, but was designed in more detail to minimize interobserver variation by having a single clinician perform all studies. Furthermore, in order to detect smaller differences than might be detectable by chart review alone, the clinician did interviews, clinical examinations, and photographs.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Each case was matched with a single control but more variables were used in matching than in previously published studies. These were age at diagnosis (±12 years as closer matching was not possible), stage, type of surgery, radiation fields, fractionation regime (where feasible), duration of time elapsed since radiotherapy was completed ±5 years (which was >12 months for all participants to ensure that sufficient time had elapsed in which chronic toxicity could begin to manifest), and whether or not adjuvant chemotherapy was administered. Patients were seen by one clinician (S. Shanley) for interview, clinical breast photographs, and LENT-SOMA (late effects on normal tissue—subjective, objective, management and analytical; ref. 5) evaluations. As outlined in Pavy et al. (5), this scoring is a combination of subjective patient reports of symptoms such as breast pain, objective clinical findings, e.g., arm circumference to assess lymphedema, recording of need for medical management of symptoms such as lymphedema, and an analysis of injury, e.g., degree of tissue induration.
Statistical considerations. Nonparametric methods were employed for analysis. Although individual matching was undertaken, matched pairs analysis was not employed, as missing or unknown data items would require that patients be omitted from analysis. The Kruskal-Wallis test was used for ordinal data when comparing more than two groups with a P value limit of significance of <0.05. If this overall test was significant, the Mann-Whitney test was then used to compare pairs of groups. The 
2 test was used to compare nominal data, and Fisher's exact test was used for 2 x 2 tables. No correction for multiple testing was employed (6), and P values must be interpreted with caution in view of the large number of significance tests done. P values >0.10 were shown as nonsignificant. P values between 0.10 and 0.051 are suggestive of significance and were therefore displayed. Missing or inapplicable responses were excluded when calculating P values. Unless otherwise indicated, P values in the tables refer to the comparison of all BRCA1 and BRCA2 carriers with control participants. Typical detectable differences (80% or greater power, 5% two-sided significance level) with 55 patients in each group (comparing binomial end points) were: 5% versus 25%; 10% versus 35%; 20% versus 45%; 30% versus 60%; 40% versus 70%, and 50% versus 80%.
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
–5
, but is distinguished in Tables 6
–9
by the description "c het." An additional 18 case participants had BRCA2 mutations.
|
|
|
|
|
|
|
|
|
Age at diagnosis and menopausal status. The median and range of ages at diagnosis of cases and controls were similar as they were matched within a window of 12 years (Table 1). Postmenopausal status was defined as the absence of menses for 1 year or bilateral oophorectomy (with or without hormone replacement therapy). Corresponding to the age matching, equivalent numbers of cases and controls were premenopausal at diagnosis (75% in each group).
Tumor and nodal status. Tumors were matched as closely as possible for tumor size (T) and nodal (N) stages at diagnosis (Table 1). Up to one T stage difference was accepted if the nodal status matched. If nodal status in a case was unknown (NX), a control was considered acceptable if the nodal status was 0 or 1.
Surgical procedure. Forty-three cases and 46 controls had wide local excisions. Nine in each group had mastectomy. There were three cases in which surgery was not undertaken prior to radiotherapy, and these were matched with controls in which wide local excisions were undertaken. They were included in the analyses of radiation toxicity but were excluded from analyses of tumor outcomes (rates of ipsilateral and contralateral tumor development) due to the atypical nature of their treatment. Furthermore, an additional three cases and their corresponding controls were excluded from tumor outcome evaluations due to differences in staging.
Contralateral mastectomy was undertaken in 7 of 48 (15%) cases evaluable for tumor outcomes and in 2 of 48 (4%) controls. To correct for this effect on the degree of exposure (exposure = number of breasts at risk x number of months at risk) to cancer risk experienced by cases versus controls, the total and median exposures were calculated for each group. Median exposure for the cases as a whole was 153 breast-months versus 177 in the controls, which was not significantly different.
Chemotherapy. Thirty-four cases in each group (62%) had chemotherapy in addition to radiation.
Endocrine therapy. A similar proportion of cases and control participants received endocrine manipulation, but the balance differed between the use of oophorectomy and tamoxifen (Table 2).
Duration of follow-up. Figures were equivalent for BRCA1 carriers and controls (with medians of 
8 years), but was shorter for BRCA2 carriers (median of 4 years), despite efforts to match within an accuracy of 5 years. For all carriers considered together, their median follow-up was 6.75 years. Because follow-up for 10 years is regarded as ideal in assessing late effects of radiation, the proportion of cases and controls with this extent of assessment were calculated and found to be equivalent with 17 of 55 (31%) in this category versus 18 of 55 (33%) of controls.
Radiation variables. Radiotherapy matching was based on the level of available data at enrollment, whereby sites were coded as breast or chest wall. Data on which patients had additional fields [supraclavicular fossa (SCF), axilla, or internal mammary nodes] were collected after the interview and examination of each participant (Table 3). The frequency of boost sites was similar in the two groups with an insignificant excess among the controls. SCF and axillary sites were more common in the case group. A significant excess of axillary irradiation occurred in the node-positive cases (8 of 24) versus 1 of 30 node-positive controls (P = 0.007). Median total dose at each site and fractionation size were not significantly different between carriers and controls (Table 4).
Other potential confounders. Potential confounders included fair skin type with minimal tanning ability, fair or red hair type, breast size greater than C cup, and concurrent illnesses such as hypertension, diabetes, or connective tissue disease. No significant differences were seen.
Histopathology. The pathologic features of the tumors reflect the significant underrepresentation of in situ carcinoma in BRCA mutation carriers (Table 5). Data on grade and hormone receptor status were missing despite vigorous efforts to locate them in 27 cases and 18 controls. Although the differences were not significant, there was a trend towards higher grade and estrogen receptor–negative status in BRCA carriers in those from which these data were accessible.
Acute toxicity
The only difference in acute toxicity was a trend towards increased recall of acute pain in BRCA carriers (Table 6), but this was not accompanied by an increase in signs of inflammation as determined by either patient recall or review of their notes. Only one case and one control had their radiotherapy interrupted due to brisk skin reactions; in a second control, boosts were reduced due to nausea, whereas in a second case, radiation therapy was delayed due to sepsis attributed to chemotherapy.
Late radiation toxicity
Frequency of late events. Rib fracture occurred in only 3 of 55 (5%) of the cases and controls (Table 7). Pulmonary fibrosis was documented in one carrier and no controls. Only one case of soft tissue necrosis was reported in a case participant in whom a mastectomy was done for recurrence postradiotherapy, resulting in some necrosis of the mastectomy flaps. No reports of cardiac effects postradiation occurred in either cases or controls.
LENT-SOMA scoring. Edema, fibrosis, telangiectasia, and atrophy did not vary significantly between groups. Chronic pain of moderate to severe degree was slightly (but insignificantly) more prevalent in the case group, but there were no reports of regular medication being required. Detectable lymphedema (determined by circumferential arm measurements) was not different in the two groups. More women among the cases required ongoing use of elastic stockings (10 of 52 = 19% versus 8%), although this was not significant. No study participant had ulceration of the breast (Table 8).
Photographic scoring. Five of 30 cases (17%) evaluable for breast distortion and 4 of 30 controls (13%) were found to have severe changes in breast size or shape (Table 9). One case and two controls had severe telangiectasia. Mild to moderate changes in breast size and shape occurred in 70% of carriers versus 66% of controls.
Cancer outcomes. Only three ipsilateral tumors were found in the cases—two in BRCA1 carriers and one in a BRCA2 carrier, whereas six occurred in the controls. Contralateral tumors, however, occurred in four cases and only one control. The low rate of tumor events meant that it was not possible to draw conclusions on the effect of therapy on cancer rates.
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
The necessary relaxation of some matching criteria (age ±12 years, time since follow-up ±5 years, and T stage ±1 provided that nodal status was matched) is acknowledged to have introduced some potential selection bias. There is a trend toward higher T and overall stage in the cases and an excess of additional treatment fields (SCF and axillary), which would be expected to bias the study in favor of increased toxicity in BRCA carriers.
An increase in chronic toxicity was not seen in BRCA1/2 mutation carriers versus controls as measured by late event rates of rib fracture, soft tissue necrosis, symptomatic cardiac, or symptomatic lung fibrosis, which occurred at similar rates to those reported by Pierce et al. (4). Carriers reported a nonsignificant increase in the use of lymphedema control measures but no demonstrable increase in arm edema was documented at their examinations. If this were indicative of a real effect on edema, it would be unlikely that axillary radiation is contributory, as only 1 of the 10 women in this group had received radiation treatment to the axilla. Most informative, however, were the objective clinical evaluations by LENT-SOMA and scoring of photographs. According to the literature on LENT-SOMA studies, the rates of toxicity of breast cancer vary, which is expected, given the variety of variables that may influence outcome. For example, severe fibrosis has been seen at rates <2% up to 16%, and severe telangiectasia at <2% up to 18%, with some of the variation being attributable to different regimes and durations of follow-up (7, 8). Hoeller et al. (7) highlights the issue of interobserver variability, hence, our decision to have all LENT-SOMA analyses done by a single observer.
Our photographic scoring was undertaken on two occasions (by independent observers blinded to mutation status) for breast shape and size and skin changes, and did not show any increase in severity of cosmetic outcomes in the BRCA1/2 mutation carriers. There is overall good agreement in our study between the scoring of breast size and shape in this photographic assessment and the assessment of atrophy in LENT-SOMA scores undertaken at the time of patient examination. Due to the differences in scale used in LENT-SOMA versus photographic scoring, there are differences in the interpretation of the severity of telangiectasia within groups (the LENT-SOMA evaluation scoring more mild and moderate telangiectasia compared with the photographic scoring) between these two aspects of our evaluations, but the consistent lack of difference between cases and controls in both forms of evaluation in our study supports the robustness of our findings.
Survival and metastatic disease were not outcome measures of this study as deceased patients were excluded, but local recurrences and contralateral tumors were recorded, although not as primary outcome measures. Low tumor event rates in cases and controls, however, limited the power of this study to assess the effects of radiation on tumor formation.
In summary, this study is the first to make detailed analyses with rigorous evaluations of late radiation treatment toxicities in BRCA1 and BRCA2 mutation carriers treated for breast cancer in a United Kingdom setting. Despite the power limitations of our study, our results add to the literature from North America and Canada on radiation outcomes in BRCA carriers, in that they exclude a major increase (in the order of a 20% increase) in severe late toxic effects in BRCA1/2 mutation carriers treated according to standard protocols in the United Kingdom during the study period in comparison to women with sporadic disease.
|
Footnotes |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Received 5/23/06; revised 8/ 4/06; accepted 9/15/06.
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. Santarosa, L. Del Col, E. Tonin, A. Caragnano, A. Viel, and R. Maestro Premature senescence is a major response to DNA cross-linking agents in BRCA1-defective cells: implication for tailored treatments of BRCA1 mutation carriers Mol. Cancer Ther., April 1, 2009; 8(4): 844 - 854. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Shanley, K. McReynolds, A. Ardern-Jones, R. Ahern, I. Fernando, J. Yarnold, G. Evans, D. Eccles, S. Hodgson, S. Ashley, et al. Acute Chemotherapy-Related Toxicity Is Not Increased in BRCA1 and BRCA2 Mutation Carriers Treated for Breast Cancer in the United Kingdom Clin. Cancer Res., December 1, 2006; 12(23): 7033 - 7038. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
