The Incidence of Lung Carcinoma after Surgery for Breast Carcinoma with and without Postoperative Radiotherapy

Transcription

1 1362 The Incidence of Lung Carcinoma after Surgery for Breast Carcinoma with and without Postoperative Radiotherapy Results of National Surgical Adjuvant Breast and Bowel Project (NSABP) Clinical Trials B-04 and B-06 Melvin Deutsch, M.D. 1 Stephanie R. Land, Ph.D. 2,3 Mirsada Begovic, M.D, Ph.D. 4 H. Samuel Wieand, Ph.D. 2,3 Norman Wolmark, M.D. 5 Bernard Fisher, M.D. 5 1 Department of Radiation Oncology, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania. 2 National Surgical Breast and Bowel Project (NSABP) Biostatistical Center, Pittsburgh, Pennsylvania. 3 Department of Biostatistics, University of Pittsburgh, Pittsburgh, Pennsylvania. 4 Department of Medicine, University of Pittsburgh, Pittsburgh, Pennsylvania. 5 National Surgical Breast and Bowel Project (NSABP) Operations Center, Pittsburgh, Pennsylvania. See editorial on pages , this issue. The National Surgical Breast and Bowel Project (NSABP) B-04 and B-06 trials were supported by Public Health Service Grants NCI-U10-CA and NCI-U10-CA from the National Cancer Institute, National Institutes of Health, Department of Health and Human Services. The authors thank Barbara C. Good, Ph.D., for editorial assistance. Address for reprints: Melvin Deutsch, M.D., Department of Radiation Oncology, University of Pittsburgh Medical Center Presbyterian, 200 Lothrop Street, Pittsburgh, PA 15213; Fax: (412) ; deutschm@msx.upmc.edu Received April 21, 2003; revision received June 19, 2003; accepted June 24, BACKGROUND. In the current study, the authors compared the incidence of subsequent primary lung carcinoma in patients with breast carcinoma who received radiotherapy as part of their treatment and in those patients who did not. The patients were participants in two large National Surgical Adjuvant Breast and Bowel Project (NSABP) breast carcinoma trials, B-04 and B-06, which prospectively randomized women to either undergo surgery alone or to undergo surgery and postoperative radiotherapy. METHODS. The NSABP trial B-04 ( ) randomized patients to undergo radical mastectomy versus total (simple) mastectomy and radiotherapy to the chest wall, axilla, and supraclavicular and internal mammary lymph node areas. For patients with a clinically uninvolved axilla, there was a third randomization arm: total mastectomy without radiotherapy. The B-06 trial ( ) randomized patients between those undergoing total mastectomy versus lumpectomy versus those undergoing lumpectomy and breast irradiation, with all patients undergoing an axillary lymph node dissection. The records of all patients who developed a recurrence in the lung or a new primary lung tumor were reviewed to determine the incidence and laterality of confirmed and probable primary lung carcinoma. RESULTS. For the 1665 evaluable patients on the NSABP B-04 trial (mean follow-up of 21.4 years), there was a total of 23 subsequent confirmed and probable ipsilateral or contralateral primary lung carcinomas. In those patients who had received comprehensive postmastectomy radiotherapy, there was a statistically significant increase in the incidence of these new primary tumors (P 0.029). With regard to the development of confirmed new primary ipsilateral lung carcinoma alone, the incidence was statistically significantly increased (P 0.013) in those patients who had received radiotherapy as part of their treatment, and when confirmed and probable ipsilateral lung carcinomas were analyzed, there was a strong trend toward a statistically significant increase in those patients who had received radiotherapy (P 0.066). For the 1850 evaluable patients on the NSABP trial B-06 (mean follow-up of 19.0 years), there was a total of 30 second primary lung carcinomas but no increase in either ipsilateral or contralateral primary tumors of the lung in those patients who had received radiotherapy. CONCLUSIONS. Extensive postmastectomy irradiation of the chest wall and regional lymphatic node areas, with consequent exposure of a greater volume of lung to higher doses as administered in the NSABP B-04 trial compared with postlumpectomy breast irradiation in the NSABP B-06 trial, was associated with an increased incidence of subsequent primary lung tumors, both ipsilateral and contralateral. Cancer 2003;98: American Cancer Society American Cancer Society DOI /cncr.11655

2 Lung Ca after Breast Ca Treatment/Deutsch et al The increasing use of breast conservation therapy (lumpectomy and radiotherapy) for Stage I and II invasive carcinoma and for ductal carcinoma in situ, plus the renewed enthusiasm for postmastectomy chest wall irradiation and regional lymph node irradiation in patients with involved axillary lymph nodes, suggests that an increasing number of women with breast carcinoma will be exposed to pulmonary irradiation. A carcinogenic effect of radiotherapy on the lungs in women with breast carcinoma has been suggested in population-based studies and case control studies from cancer registries, 1 5 but other retrospective studies from individual institutions comprised of large numbers of breast carcinoma patients treated with radiotherapy after breast conservation surgery report conflicting results To our knowledge, the relation between radiotherapy for breast carcinoma and the development of subsequent lung carcinoma is uncertain. In the current study, we reviewed data regarding new primary lung carcinomas from two large, prospective, randomized National Surgical Adjuvant Breast and Bowel Project (NSABP) clinical trials for breast carcinoma (NSABP B-04 and NSABP B-06), in which patients were randomized between treatment with surgery alone versus treatment with surgery and postoperative radiotherapy. To our knowledge, there has been no other report on this subject from trials of this magnitude and with two decades of follow-up. MATERIALS AND METHODS The Clinical Trials The B-04 study (Fig. 1A) accrued 1765 patients between 1971 and 1974, 1665 of whom were judged eligible based on protocol requirements. Patients were randomized according to the preoperative clinical assessment of their axillary lymph node status. Those with a clinically negative axilla were randomized to one of three arms: 1) radical mastectomy without radiation; 2) total mastectomy and radiotherapy to the chest wall, axilla, supraclavicular region, and internal mammary lymph node regions; or 3) total mastectomy without radiotherapy. Patients with clinically involved axillary lymph nodes were randomized only between radical mastectomy versus total mastectomy and radiotherapy. For those undergoing total mastectomy, the intended surgery was removal of the entire breast but not the underlying pectoralis muscles or the ipsilateral axillary lymph nodes. Eligibility criteria, details regarding the surgical procedures, and radiotherapy guidelines have been described previously. 11 KEYWORDS: breast carcinoma, lung carcinoma, mastectomy, postmastectomy irradiation, second malignancy. FIGURE 1. (A) Schema of the National Surgical Breast and Bowel Project (NSABP) B-04 trial. (B) Schema of the NSABP B-06 trial. LNs: lymph nodes; RT: radiotherapy; Dis: dissection; Mx: mastectomy; Ax Dis: axillary lymph node dissection; Irrad: irradiation. For patients randomized to treatment with radiotherapy, postoperative radiotherapy was administered to the chest wall via 2 parallel opposed tangential fields to a dose of 5000 centigrays (cgy) in 25 fractions calculated at a depth of two-thirds the distance between the skin of the chest wall and the base of the tangential fields at midseparation. The supraclavicular and infraclavicular regions were treated with a single anterior field angled 15 degrees laterally to administer 4500 cgy at a depth of 3 cm. The inferior border of this field usually was at the level of the second costalchondral junction. A smaller posterior axillary field was used to bring the midaxilla dose to 5000 cgy in 25 fractions. The internal mammary lymph node chain was either included in the tangential fields or, more commonly, treated with 4500 cgy in 25 fractions at a depth of 3 cm by a separate anterior field usually 6 cm wide (Fig. 2). All radiotherapy was administered with cobalt-60 or supervoltage X-ray equipment. Electron beams were not used in this study. In addition to the above, patients with a clinically positive axilla were to be treated with an additional cgy per 5 10 fraction boost using a direct appositional field di-

3 1364 CANCER October 1, 2003 / Volume 98 / Number 7 nor regional lymph node irradiation was permitted. Patients with pathologically involved axillary lymph nodes subsequently received 2 years of treatment with 5-fluorouracil and phenylalanine mustard. Eligibility criteria, details regarding the randomization of patients, and guidelines for the surgical procedure and radiotherapy have been described previously. 12 For both the NSABP B-04 and B-06 studies, informed consent was obtained from all patients. FIGURE 2. Example of postmastectomy radiotherapy fields in the National Surgical Breast and Bowel Project (NSABP) B-04 trial. rected toward the palpable axillary lymph nodes. All radiotherapy was administered daily, 5 days per week. The B-06 trial (Fig. 1B) accrued 2163 patients between 1976 and 1984, of whom 1850 were judged eligible and evaluable with respect to second primary tumors. All patients were randomized to one of three arms: 1) total mastectomy and axillary lymph node dissection without breast radiotherapy; 2) lumpectomy and axillary lymph node dissection without radiotherapy; or 3) lumpectomy and axillary lymph node dissection with postoperative radiotherapy. Patients randomized to Group 3 were treated with radiotherapy to the ipsilateral breast alone via two opposed parallel tangential fields. The administered dose was 5000 cgy in 25 fractions calculated at a depth of twothirds the distance between the skin overlying the breast and the base of the tangential fields at midseparation. All radiotherapy was administered with either cobalt-60 gamma radiation or megavoltage X-rays. Neither a supplemental boost to the operative area Evaluation of New Primary Lung Carcinomas All records of patients reported as having developed disease recurrence in the lungs or a new primary lung tumor were reviewed by two physicians. Using radiographic, pathologic, and clinical reports, a determination was made for each patient as to whether the reported findings in the lung represented a definite or probable new lung carcinoma versus metastatic breast carcinoma or even benign changes. Patients were considered to have a confirmed lung primary tumor if a lung biopsy or cytology revealed a different histology from the primary breast carcinoma such as squamous cell carcinoma, large cell carcinoma, or small cell carcinoma. Patients also were considered to have a confirmed primary tumor of the lung if radiographic findings were compatible with this diagnosis and there was a histologic diagnosis of adenocarcinoma considered by the institution s pathologist to be suggestive of a new primary tumor versus metastatic disease. Patients considered to have an unconfirmed but probable primary lung tumor were those with a solitary lung nodule hilar or mediastinal lymphadenopathy and a histologic diagnosis of adenocarcinoma that the institution s pathologist considered to be either primary lung carcinoma or metastatic breast carcinoma. Also, patients with a nonbiopsied solitary lung nodule with or without hilar or mediastinal lymphadenopathy were considered to have an unconfirmed but probable primary lung carcinoma. Patients were considered to have metastatic breast carcinoma in their lungs if there were multiple lung nodules and/or a pleural effusion with a histology similar to that of the primary breast carcinoma. Statistical Analysis Follow-up data as of March 31, 2002 were included in the current analysis. The follow-up time among surviving patients in the NSABP B-04 trial ranged from a minimum of 1 year to a maximum of 30.5 years, with a mean of 21.4 years. There were only 2 patients with follow-up of 8 years. In the NSABP B-06 trial, the follow-up time among surviving patients ranged from a minimum of 0.9 years to a maximum of 29.7 years,

4 Lung Ca after Breast Ca Treatment/Deutsch et al TABLE 1 Incidence of Subsequent Lung Carcinomas According to Breast Carcinoma Treatment NSABP B-04 NSABP B-06 RM n 654 TM n 365 TM RT TM n 646 P value a n 589 L n 633 L RT n 628 P value a Confirmed ipsilateral lung carcinoma Confirmed and probable ipsilateral lung carcinoma All confirmed and probable ipsilateral plus contralateral lung carcinomas NSABP: National Surgical Breast and Bowel Project; RM: radical mastectomy; TM: total mastectomy; RT: radiotherapy; L: lumpectomy. a P value refers to comparison of the radiotherapy arm with the treatment arms without radiotherapy. with a mean of 19.0 years. There were 12 patients with a follow-up 5 years. The primary comparisons were performed with a binomial exact test, comparing the number of confirmed and probable second primary lung carcinomas that occurred in the irradiated groups with the number expected based on the total follow-up time in each group (radiotherapy and no radiotherapy) for each study. FIGURE 3. Cumulative incidence of all lung carcinomas in the National Surgical Breast and Bowel Project (NSABP) B-04 trial (ipsilateral and contralateral, confirmed and unconfirmed). RESULTS The 25-year and 20-year disease-free and overall survival data for the NSABP B-04 and B-06 studies, respectively, were reported recently. 13,14 Of the 1665 evaluable patients on the NSABP B-04 study, 23 (1.4%) subsequently developed what was considered to be confirmed or probable primary lung carcinoma. There was a statistically significant increase in the total number of lung carcinomas, histologically confirmed and unconfirmed, both ipsilateral and contralateral, among the patients randomized to receive postmastectomy comprehensive radiotherapy (P 0.029) and a statistically significant increase in confirmed ipsilateral lung carcinoma among the irradiated patients (P 0.013). However, when the unconfirmed ipsilateral lung carcinoma cases were included, there was only a strong trend toward an increased incidence of ipsilateral lung carcinoma among the irradiated patients (P 0.066) (Table 1). The cumulative incidence plots for all subsequent lung carcinomas, both ipsilateral and contralateral, is shown in Figure 3. Figures 4 and 5 show the cumulative incidence plots for 1) ipsilateral confirmed and 2) ipsilateral confirmed and unconfirmed lung carcinomas, respectively, for the surgery without radiotherapy and the surgery with radiotherapy arms. The majority of new lung carcinoma cases were diagnosed 5 years from the time of diagnosis of the breast carcinoma. The rate ratios for a primary lung carcinoma in an irradiated patient compared with a nonirradiated patient were 2.61, 5.03, and 2.79, respectively, for all lung carcinomas, ipsilateral confirmed lung carcinomas, and ipsilateral confirmed and unconfirmed but likely lung carcinomas (Table 2). Table 3 shows the histologic diagnoses of lung carcinoma on the NSABP B-04 study, based on the institution s pathology report for all confirmed and probable cases. Of the 1850 evaluable patients in the NSABP B-06 study, 30 (1.6%) developed a new primary lung carcinoma, confirmed or probable, with no significant differences noted between patients who received radiotherapy and those who did not (Table 1). DISCUSSION Although to our knowledge the majority of previous reports regarding lung carcinoma after treatment of breast carcinoma have been based on reviews of non-

5 1366 CANCER October 1, 2003 / Volume 98 / Number 7 TABLE 2 Rates of Subsequent Lung Carcinomas per 10,000 Patient-Years of Follow-Up and Rate Ratios of Irradiated versus Nonirradiated Patients in NSABP Trial B-04 Surgery RT Surgery alone Ratio All confirmed and probable ipsilateral contralateral lung carcinoma Confirmed ipsilateral lung carcinoma Confirmed and probable ipsilateral lung carcinoma FIGURE 4. Cumulative incidence of ipsilateral confirmed lung carcinoma cases in the National Surgical Breast and Bowel Project (NSABP) B-04 trial. NSABP: National Surgical Breast and Bowel Project; RT: radiotherapy. TABLE 3 Histology of Lung Carcinoma by Breast Carcinoma Treatment in NSABP Trial B-04 Histology Surgery Surgery RT Total FIGURE 5. Cumulative incidence of ipsilateral confirmed and unconfirmed lung carcinoma cases in the National Surgical Breast and Bowel Project (NSABP) B-04 trial. Ipsilateral confirmed Adenocarcinoma Squamous cell Large cell Small cell Ipsilateral probable No histology a Adenocarcinoma Undifferentiated Contralateral confirmed Adenosquamous Squamous cell Contralateral probable No histology a NSABP: National Surgical Breast and Bowel Project; RT: radiotherapy. a Diagnosis by radiographic studies. randomized populations from tumor registries or single institutions, the advantage of the current study is that all patients we reviewed were enrolled in two large clinical trials in which they were randomly assigned to receive radiotherapy versus no radiotherapy. In addition, long-term follow-up exists for these patients, and the reviewers were blinded as to administered treatment. However, there was no central review of pathologic or radiographic material, and because this was a retrospective study, it was necessary to rely on institutional reports when making a determination of primary lung carcinoma or metastatic breast carcinoma. As noted earlier, cases of second primary lung carcinoma were considered to be either confirmed or probable on the basis of a comprehensive, detailed review of the radiographic, pathologic, and clinical records in addition to the assessment by the individual institution s physicians. In some cases, physicians from the individual treating facilities were unable to distinguish between metastases and second primary tumors. We suspect that this same difficulty and uncertainty in making an accurate determination of a new primary lung carcinoma versus metastatic breast carcinoma exists in other retrospective reviews. Several large retrospective reviews from tumor registries have indicated an increased risk of primary lung carcinoma in patients who received postmastectomy radiotherapy. 1 3 The exact details of the administered radiotherapy were not usually known. The majority of these patients were treated with orthovoltage radiation, and a lesser number with cobalt-60. In their review of breast carcinoma patients from the Con-

6 Lung Ca after Breast Ca Treatment/Deutsch et al necticut tumor registry treated between 1935 and 1971 with radiotherapy (mostly orthovoltage), Inskip et al. calculated that the average dose to the ipsilateral lung was 1520 cgy. 3 They estimated that the excess relative risk of developing lung carcinoma in the ipsilateral lung increased 20% per 100 cgy. With the use of electron beam radiotherapy, computed tomography planning, and linear accelerators, the volume of irradiated lung and the administered dose to the lung can be reduced substantially. As previously reported, with breast-only irradiation, only a narrow portion of the anterior ipsilateral lung, usually 3 cm in width, is within the high-dose volume. 15 The radiotherapy techniques used in the NSABP B-04 protocol allowed a maximum thickness of up to 5 cm of lung in the tangential beams used to treat the chest wall. Electron beams were not used, and therefore the anterior fields used to treat the supraclavicular and internal mammary lymph node areas delivered a substantial dose to the underlying lung. The data from the literature regarding subsequent primary lung carcinoma after lumpectomy and radiotherapy are less certain. Galper et al. reported a statistically significant increase in second primary lung carcinomas over what was expected after 5 years of follow-up in women who underwent lumpectomy and radiotherapy and were age 50 years at the time of diagnosis of their breast carcinoma. 8 In addition to breast irradiation, 50% of their patients received supraclavicular irradiation. Obedian et al. also showed an increased incidence of lung carcinoma after lumpectomy and radiotherapy, mainly in smokers. 9 However, as in the series by Galper et al., the majority of patients also underwent irradiation of the supraclavicular region and approximately 50% received internal mammary lymph node irradiation in addition to breast irradiation. Patients who continued to smoke at the time of radiotherapy had a higher incidence of subsequent lung carcinomas versus those who were former smokers versus those who were never smokers (P 0.06). 9 Neugut et al., in a case control study from a Connecticut tumor registry, also reported an increased risk of lung carcinoma 10 years after radiotherapy, with a markedly increased risk of lung carcinoma in women who, in addition to receiving radiotherapy, were cigarette smokers. 4 However, to our knowledge, there were no details concerning the administered radiotherapy to the breast or the chest wall in this series. In a review of 1253 women who were treated with lumpectomy and breast irradiation and followed for a median of 8.9 years, Fowble et al. found a 1% incidence of second lung carcinomas at 10 years (2% in smokers and 0.3% in nonsmokers). Approximately 22% of their patients also received supraclavicular irradiation. The difference in the incidence of lung carcinomas between smokers and nonsmokers was not found to be statistically significant (P 0.16). 7 As is the case with a majority of other reports, the current series lacks precise information regarding the history of cigarette smoking, and we are unable to determine the effect, if any, of smoking on the development of lung carcinoma after treatment for breast carcinoma, with or without radiotherapy. A review by Lavey et al. of patients from a single institution followed after lumpectomy or mastectomy and breast/chest wall irradiation demonstrated no increased incidence of second primary lung carcinomas. 6 It is interesting to note that 74% of those who received radiotherapy also received treatment to the supraclavicular and internal mammary lymph node areas. Similarly, in our review of patients on the NSABP B-06 study, in which patients received postlumpectomy breast irradiation, there was no increased incidence of subsequent primary lung carcinomas reported. The majority of the reports concerning lung carcinoma after breast carcinoma include few patients treated with chemotherapy. Therefore, it is not possible to make a statement regarding how chemotherapy would impact on the incidence of postradiotherapy lung carcinoma. Our data from the NSABP B-06 study, in which patients with involved axillary lymph nodes received chemotherapy, did not demonstrate an increased incidence of second lung carcinomas in this population. However, radiotherapy in that protocol was administered just to the breast, with a small volume of underlying lung included in the high-dose volume. Current recommendations for the administration of postmastectomy radiotherapy or postlumpectomy radiotherapy in patients with positive lymph nodes include guidelines that would substantially reduce the volume of lung irradiated and the absorbed dose in the lung compared with techniques such as those used in the older NSABP B-04 study Therefore, if lung volume and dose received are risk factors, the subsequent incidence of ipsilateral lung carcinoma in currently irradiated patients should be lower than what we reported from the NSABP B-04 study. REFERENCES 1. Schenker JG, Levinsky R, Ohel G. Multiple primary malignant neoplasms in breast cancer patients in Israel. Cancer. 1984;54: Harvey EB, Brinton LA. Second cancer following cancer of the breast in Connecticut, Natl Cancer Inst Monogr. 1985;68:

7 1368 CANCER October 1, 2003 / Volume 98 / Number 7 3. Inskip PD, Stovall M, Flannery JT. Lung cancer risk and radiation dose among women treated for breast cancer. J Natl Cancer Inst. 1994;86: Neugut AI, Murray T, Santos J, et al. Increased risk of lung cancer after breast cancer radiation therapy in cigarette smokers. Cancer. 1994;73: Travis LB, Curtis RE, Inskip PD, Hankey BF. Lung cancer risk and radiation dose among women treated for breast cancer. J Natl Cancer Inst. 1995;87: Lavey RS, Eby NL, Prosnitz LR. Impact of radiation therapy and/or chemotherapy on the risk for a second malignancy after breast cancer. Cancer. 1990;66: Fowble B, Hanlon A, Freedman G, Nicolaou N, Anderson P. Second cancers after conservative surgery and radiation for stages I-II breast cancer: identifying a subset of women at increased risk. Int J Radiat Oncol Biol Phys. 2001;51: Galper S, Gelman R, Recht A, et al. Second nonbreast malignancies after conservative surgery and radiation therapy for early-stage breast cancer. Int J Radiat Oncol Biol Phys. 2002;52: Obedian E, Fischer DB, Haffty BG. Second malignancies after treatment of early-stage breast cancer: lumpectomy and radiation therapy versus mastectomy. J Clin Oncol. 2000;18: Kurtz JM, Amalric R, Brandone H, Ayme Y, Spitalier JM. Contralateral breast cancer and other second malignancies in patients treated by breast-conserving therapy with radiation. Int J Radiat Oncol Biol Phys. 1988;15: Fisher B, Montague E, Redmond C, et al. Comparison of radical mastectomy with alternative treatments for primary breast cancer. Cancer. 1977;39: Fisher B, Bauer M, Margolese R, et al. Five-year results of a randomized clinical trial comparing total mastectomy and segmental mastectomy with or without radiation in the treatment of breast cancer. N Engl J Med. 1985;312: Fisher B, Jeong JH, Anderson S, Bryant J, Fisher ER, Wolmark N. Twenty-five year follow-up of a randomized trial comparing radical mastectomy, total mastectomy, and total mastectomy followed by irradiation. N Engl J Med. 2002;347: Fisher B, Anderson S, Bryant J. Twenty-five year follow-up of a randomized trial comparing total mastectomy, lumpectomy, and lumpectomy plus irradiation for the treatment of invasive breast cancer. N Engl J Med. 2002;347: Deutsch M, Bryant J, Bass G. Radiotherapy review on National Surgical Adjuvant Breast and Bowel Project (NSABP) Phase III Breast Cancer Clinical Trials. Is there a need for submission of portal/simulation films? Am J Clin Oncol. 1999;22: Kuske RR. Adjuvant chest wall and nodal irradiation: maximize cure, minimize late cardiac toxicity. J Clin Oncol. 1998; 16: Conte G, Nascimben O, Turcato G. Three-field isocentric technique for breast irradiation using individualized shielding blocks. Int J Radiat Oncol Biol Phys. 1988;14: Hartsell WF, Murthy AK, Kiel KD, Kao M, Hendrickson FR. Technique for breast irradiation using custom blocks conforming to the chest wall contour. Int J Radiat Oncol Biol Phys. 1990;19: Arthur DW, Arnfield MR, Warwicke LA, Morris MM, Zwicker RD. Internal mammary node coverage: an investigation of presently accepted techniques. Int J Radiat Oncol Biol Phys. 2000;48: