194 I. .I. Radiation Oncology l Biology l Physics Volume 45, Number 3 Supplement 1999

9 1 SECOND MALIGNACIES AFTER TREATMENT OF EARLY STAGE BREAST CANCER WITH LUMPECTOMY AND RADIATION (LRT) OR MASTECTOMY WITHOUT RADIATION (MAST)

Obedian E, Fischer D, Haffty B

Yale University School of Medicine, New Haven, CT, USA

Purpose: With the increasing utilization of lumpectomy and radiation (LRT) in the managememnt of early stage of breast cancer, the safety and efficacy of such treatment must be documented through long term follow up studies. The risk of second malignancies after LRT has not been well established in the literature. The purpose of this study is to de&mine the risk of second malignancies after LRT as as function of clinical pathologic treatment parmeters, and to compare it to a similiar cohort of early breast cancer patients undergoing mastectomy without radiation (MAST).

Materials: Between January, 1970 and December, 1990, 1029 breast cancer patients at our institution underwent LRT using standard mega voltage external beam therapy to the intact breast. Wedges were routinely employed on both medial and lateral tangential fields. Regionai lymph nodes were also treated as clinically indicated, with separate internal mammary field employed in 548 patients (53%). A cohort of 1387 early stage breast cancer patients who underwent surgical treatment by simple, modified or radical mastectomy (MAST) without postoperative radiation during the same 20 year period served as a comparison group. Tumor registry data and patient records were reviewed with specific attention to the development of second malignancies. Second malignancies were categorized as contralateral breast vs non-breast including skin, soft tissue, lung, head and neck, heme/lymph, GI, Gyn, and GU. Lobular carcinoma-in-situ of the contralateral breast was excluded from anylsis. In the chart of patients undergoing LRT, detaliled analysis was carried out with respect to age, stage, smoking history, RT technique, dose. the use of chemotherapy or hormone therapy, and other clinical and/or pathological characteristics.

Results: As of March, 1999, the minimum evaluable follow-up for all patients in both cohorts was 9 years, the median follow-up was 14.6 years for the LRT group and 16 years for the MAST group. For all patients in both groups there were a total of 158 contralateral breast malignancies and 164 non-breast malignancies. The 15.year risk of any second malignancy was nearly identical for both cohorts (17.5%~~ 19%). The second breast malignancy rate at 15 years was 10% for both the MAST and LRT groups. The 15 year risk of a second non-breast malignancy was also 11% for the LRT and 10% for the MAST group. In the subset of patients< 45 years of age at treatment, the second breast and non-breast malignancy rates at 15 years were 10% and 5% for patients undergoing LRT vs 7% and 4% for patients undergoing mastectomy (p=NS). Patients developing any second malignancy had a significantly lower survival than the remainder of the cohort (55% vs 69% at 15 years p=O.O5). In the detailed analysis of LRT patients, second lung malignancies were associated with a history of tobacco use particularly if smoking continued during the RT (p=O.O57). There were fewer contralateral breast tumors in patients undergoing adjuvant hormone therapy, although this did not reach statistical significance. The adjuvant use of chemotherapy did not significantly affect the risk of second breast or non-breast malignancies. the rate of second malignancies was slightly higher in patients treated with a separate internal mammary field, (19% vs 15%) but this did not reach statistical significance(p=0.097).

Conclusion: With a median follow-up of nearly 15 years after treatment for early stage breast cancer, there appears to be no increased risk of second malignancies in patients undergoing LRT using modem techniques compared to MAST, even in the subset of patients under age 45 at the time of treatment. Although not statistically significant, adjuvant hormonal therapy appeared to lower the risk of a contralateral breast cancer and smoking increased the risk of a second lung malignancy. Chemotherapy had no adverse effect on the rate of second malignancy. Continued monitoring of these patient cohorts will be required to document that these finding are maintained with even longer-term follow-up. With nearly 15 years median follow-up, however, these data should be reassuring to women considering LRT as a treatment option.

92 LOCAL-REGIONAL CONTROL IN BREAST CANCER PATIENTS WITH A PRESUMED GENETIC PREDISPOSITION

Freedman LM, Thames HD, Hunt KK, Strom EA, McNeese MD, Heaton KM, Buchholz TA UGersity of Texas M. D. Anderson Cancer Center, Houston, 7X, USA

Purpose: Local control rates for breast cancer patients with a genetic predisposition for their disease are poorly defined. Because such a small percentage of breast cancer patients have identified germline mutations, surrogates, such as young age at diagnosis coupled with a positive family history have been used to examine this issue. We elected to study local-regional control rates following breast conservation therapy in genetically predisposed women by evaluating the outcome of patients with bilateral breast cancer and a positive breast cancer family history.

Materials and Methods: Between 1959 and 1998, 58 patients with bilateral breast cancer and a family history of breast cancer received treatment in our institution. Thirty-six of these patients had a least one breast treated with breast conservation therapy. We retrospectively reviewed the clinical characteristics, treatment, and recurrence patterns for each treatment site (116 total). The primary surgical treatment was a breast conserving procedure in 55 cases and a mastectomy in 61 cases. The family history of breast cancer included at least one primary relative in 69% and one or more secondary relatives in 3 1%. Breast cancer histology was infiltrating ductal carcinoma in 78%, invasive lobular carcinoma in 9%, and other or unknown in 13%. Tumor stage for the breast conservation cases was 0 in 4’%, I in 47%, II in 42%, and III in 7%. For the mastectomy cases the stage was 0 in 3%, I in 25%, II in 38%, III in 27%, and unknown in 8%. The median follow-up was 68 months for breast conservation therapy and 57 months for mastectomy.

Results: Eight events of local-regional recurrence were seen in the 55 breasts treated with conserving therapy. The 5- and IO-year actuarial local-regional control rates for these breasts were 86% and 76%. Nine of these patients did not have radiation as a component of breast preservation treatment. Of these, 4 had local recurrence (5. and lo-year local-regional control rates of 49%). Only 4 local recurrences occurred in the group treated with breast conserving therapy and radiation (5. and lo-year actuarial local-regional control rates of 94% and 83%). There were 18 breasts at risk for recurrence 8 or more years from diagnosis and 2 had late local recurrences (recurred at 8 years and 9 years, respectively). In the breasts treated with breast conserving surgery, an analysis of radiation use, age (under or over 40 at first diagnosis), degree of family history (primary versus secondary), margin status (negative versus other), and stage revealed that only the use of radiation was associated with

Proceedings of the 41st Annual ASTRO Meeting 195

improved local control (log rank test of radiation versus no radiation actuarial curves, p= 0.009). The lo-year actuarial rates of local-regional control following mastectomy with and without radiation were 91% and 89%, respectively.

Conclusions: Patients with a genetic predisposition to breast cancer have low S-year rates of local recurrence when treated with breast conserving surgery and radiation. However, in our series, the local failure rate exceeded 50% when radiation was omitted. Our data is consistent with the hypothesis that patients with an underlying genetic predisposition develop cancers with biologically aggressive phenotypes but which also are radiosensitive. Furthermore, these patients are likely to be at increased risk for late local recurrences, possibly due to the formation of new breast cancers.

93 THE GERMLINE p53 13964 GC MUTATION CONFERS RESISTANCE TO RADIATION IN FAMILIAL BREAST CANCER PATIENTS.

Turner BC’, Lehman TA’, Modali R’, Carbone C’, Bishop L*, Curran WJ’, Glazer PM3, Haffty BG3

Thomas Jefferson University Hospitial, Philadelphia, PA, USA’; BioSewe Biotechnologies, Laurel, MD, USA’; Yale University, New Haven, CT, USA’

Purpose: We have previously showed that breast cancer patients with constitutional mutations may be at increased risk of clonogenic and de noveau breast tumor recurrence following treatment with lumpectomy and radiation. Germline mutations in the ~53 gene are observed in patients with Li-Fraumeni Syndrome (LFS) and previously have been found in less then 1% of all women with breast cancer. Some studies have demostrated that mammalin cells with ~53 gene mutations are resistant to radation-induced cell death and lymphoblastoid cells from LFS patients are also resistant to radiation-induced apoptosis. In this study, we examined the frequency of a specific gemline noncoding intronic ~53 mutation in familial breast cancer patients treated with lumpectomy and radiation therapy and determined the functional activity of the ~53 1 39640C mutation.

Materials: From the Yale University breast cancer data base, we identified 42 breast cancer patients with strong family histories of early-onset breast cancer (index cases) and 172 sporadic breast cancer patients with no identifiable history of early-onset breast or ovarian cancer. These patients donated blood for DNA isolation from lymphocytes, were interviwed to obtain pedigrees, and signed written informed consent. The lymphocyte DNA was purified and used for polymerase chain reaction (PCR) single strand conformation polymorphism and sequencing analysis of exons 5-9 and adjacent intron sequences. The affymetrix gene chip scanner was used to scan germline ~53 mutations in exons Z-12. A PCR based genotype detection assay was developed to detect the ~53 13964 oC base change based on the loss of a HhaI restriction site that is easily detetcted on a 4% gel (3% NuSieve/l% agarose). Lymphoblastoid cells from patients with wild-type ~53 (012, VDS) and the ~53 139640C mutation (BT-16, BT-102) were immortalized using epstein-barr virus and maintained in cell culture. Cell survival, apoptosis, and cell cycle changes following treatment with cisplatinum (10 ug/ml) and varying doses of ionizing radiation (o-1200 rads) were performed using FACS.

Results: None of the 42 index cases were found to contain constitutional coding mutations in exons 2-12 of the ~53 gene. However, a germline mutation in the ~53 gene at nucleotide 13964 with a G to C base change ( 139640C) adjacent to intron 6 which contains a CpG-rich Alu repetitive element was identified in 3/42 (7.1%) hereditary breast cancer patients (index cases). Two patients were heterozygous for this mutation and one patient had a homozygous mutation. In comparison, O/172 (0%) of sporadic breast cancer patients had the ~53 139640C mutation (p=O.O003). Comprehensive BRCAI and BRCA2 nucleotide analysis from patients with the ~53 13964GC mutation revealed no concomitant deleterious BRCAl or BRCAZ mutations. Pedigree anaylsis demostrated that all three patients had strong family histories of multiple types of cancers consistent with LFS but with late age of onset. Two of the three patients with the ~53 139640C mutation had ipsilateral breast tumor recurrence (IBTR) following lumpectomy and radiation and both of these patients had bilateral breast cancer suggesting the inheritance of a dominant breast cancer susceptibility gene. Functional analysis of immortalized lymphoblastoid cell lines derived from two patients with thep53 139640C mutation demonstrated prolonged survivial and decreased apoptosis in response t’o both ionizing radiation and cisplatinum compared with ~53 wild-type lymphoblastoid cells. Immunohistochemical analysis of breast tumors from these patients revealed high levels of mutant ~53 protein also suggesting that the ~53 13964GC mutation has functional activity.

Conclusion: The germlinep53 139640C mutation is a single noncoding nonsplicing intron mutation that is prevalent in women with hereditary breast cancer and confers resisitance to radiation-induced cell death. Breast cancer patients with the ~53 1 39640C mutation treated with lumpectomy and radiation therapy may be at increased risk for IBTR.

94 PRELIMINARY ANALYSIS OF RADIOTHERAPY DATA FROM CALGB 9082: VARIABILITY OF TREATMENT FIELDS FOR LOCAL/REGIONAL BREAST CANCER AND THE IMPACT OF HIGH DOSE CHEMOTHERAPY ON THE ABILITY TO DELIVER RADIATION THERAPY

Marks LBr, Fitzgerald TJ’, Laurie F’, Glicksman AS’, Rosner GL’, Vredenburgh J’, Shpall EJ4, Crump MS, Norton L’, Peters W3

Duke Vniversiv Medical Center & CALGB, Durham, NC, USA’; QARC, Providence, RI’; CALGB3; SWOG4; NCIC*

Purpose: CALGB 9082 is a randomized trial in patients with breast cancer involving 2 10 axillary nodes without evidence of distant metastasis. Patients received CAF chemotherapy and were then randomized to receive either HD-CPB (high dose cyclophosphamide, cisplatin, BCNU with autologous marrow/stem-cell support) OR intermediate-dose CPB (ID-CPB). All patients were prescribed to get local-regional RT following chemo. Central review of the RT records was performed at QARC (Quality Assurance Review Center). We herein report the rate of common deviations in the design of RT fields in patients on this trial, and report the impact of HD-CPB on the patient’s ability to receive RT.

Materials and Methods: 783 patients were randomized between ‘91.‘98, and 653 were irradiated. The frequency of starting RT was compared in the HD-CPB and the ID-CPB groups (2.tailed chi-square). In the 645 patients with RT-data forms