Radiotherapie - bestraling na operatie bij borstkanker zinloos?

Een grote studie in 21 centra/ziekenhuizen in Japan over een periode van 6 jaar 1986 tot 1991 heeft uitgewezen dat bestraling van vrouwen die een operatie van hun borstkanker hebben ondergaan geen verschil maakt in overlevingskansen en kans op een recidief. De conclusies van de onderzoekers was dan ook en ik vertaal nu hun officiële conclusie:

"Onze bevindingen suggereren dat borstkankerpatiënten niet altijd radiotherapie nodig hebben na borstbesparende operatie, hoewel de algemeen gehanteerde voorspelling van patiënten behandeld op deze manier en het lage aantal van tumorrecidieven niet toestaat de vergelijking van de behandelmethoden als statisch bewijs te benoemen. Echter als gevolg van onze bevindingen en de feiten dat lokale tumorrecidieven normaal pas gerapporteerd worden 10 tot 15 jaar na de radiotherapie en dat andere schadelijke effecten na 20 tot 30 jaar onduidelijk zijn, de ontwikkeling van nieuwere methoden in plaats van radiotherapie zijn gewenst. In aanmerking genomen dat de gemiddelde levensverwachting van Japanse vrouwen 85 jaar bedraagt, is verbetering van een borstbesparende behandeling nodig en het wordt verwacht dat hormoontherapie (of een andere chemotherapie) die relatief vrij zijn van ongunstige effecten, een significante rol zullen gaan spelen in toekomstige behandelprotocollen"
Hier het volledige artikel zoals dat gepubliceerd is in Clin Drug Invest 21(11):775-782, 2001
Onderaan dit bericht het adres van de onderzoekers waarmee gecorrespondeerd kan worden over dit onderzoek.

Primary Breast Cancer Patients Treated with Breast-Conserving Therapy and Tamoxifen: Do They Always Require Radiotherapy?
T. Tominaga, Breast Cancer Centre, Toyosu Hospital, Showa University School of Medicine, Tokyo, Japan; H. Koyama, Department of Surgery, Osaka Medical Centre for Cancer & Cardiovascular Diseases, Osaka, Japan; Y. Nomura, Department of Mammary Glands, National Kyushu Cancer Centre, Fukuoka, Japan (up to December 1999); S. Miura Department of Breast Surgery, Aichi Cancer Centre, Aichi, Japan; T. Morimoto, School of Medical Sciences, The University of Tokushima, Tokushima, Japan (Study Group on Breast Conserving Surgery for Breast Cancer)

Abstract
Objective: To study whether Japanese breast cancer patients obtain favourable results with breast-conserving surgery if the same selection criteria as those for Western patients are used, and, simultaneously, to assess the outcome of breast-conserving surgery alone versus breast-conserving surgery followed by either irradiation of the remaining breast or adjuvant chemotherapy with tamoxifen.
Design and Setting: A prospective, randomised study and a retrospective study of postoperative adjuvant therapy for breast cancer following breast-conserving surgery were performed at 21 centres in Japan during 1989 to 1991.
Patients and Interventions: In the prospective, randomised study, 112 patients with tumours 2cm or smaller in diameter who underwent lumpectomy received either adjuvant chemotherapy with tamoxifen 20 mg/day for 2 years (Group T, n = 53) or no therapy (Group C, n = 59) postoperatively.
Results: Median 5-year disease-free survival rates after 72 months were 78.5% and 86.0%, respectively, with no significant difference between the two groups [log-rank test: p = 0.308, hazard ratio 1.47 (95% confidence interval 0.59-3.66)]. In the retrospective study, a further 70 patients who underwent similar breast-conserving surgery during the same period as the prospective study followed by radiotherapy (Group R) were compared with the patients in Groups T and C combined (n = 112). Median 5-year disease-free survival rates after 71 months were 86.6% and 82.2%, respectively, with no statistically significant difference between the two groups [log-rank test: p = 0.352, hazard ratio 0.72 (95% confidence interval 0.33-1.59)].
Conclusion: These findings suggest that breast cancer patients may not always require radiotherapy after breast-conserving surgery. [Clin Drug Invest 21(11):775-782, 2001. © 2001 Adis International Limited]

Introduction
In Japan, breast-conserving surgery is performed in approximately 30% of all breast cancer patients,[1] and results with this procedure have proved better than in Western patients.[2] If tumour-negative margins of resected specimens are confirmed, postoperative radiotherapy may not be performed, and one Japanese study has reported that the prognosis after local excision of breast tumours is similar with and without postoperative radiotherapy.[3] A high percentage of women with tumour-free margins may in fact receive unnecessary radiotherapy if breast-conserving surgery with lumpectomy plus axillary lymph node dissection plus radiotherapy is performed in all such cases. To study whether Japanese breast cancer patients obtain favourable results with breast-conserving surgery if the same selection criteria as those for Western patients are used, we initiated a prospective, randomised comparative study in 1989 and, simultaneously, a retrospective comparative study to assess the outcome of breast-conserving surgery alone versus breast-conserving surgery followed by either irradiation of the remaining breast or adjuvant chemotherapy with tamoxifen.
Patients and Methods
Patients and Surgical Procedures
Patients with stage I primary breast cancer (TNM classification) who were 75 years of age or younger were included in the study. The size of tumours located in the lateral breast had to be 2cm or smaller in diameter, while those located in the medial breast had to be 1cm or smaller in diameter. In addition, there had to be at least 3cm between the margins of the tumour and the nipple. Patients with the following tumours were excluded from the study: non-invasive breast cancer, mammographic evidence of calcification 2cm or larger in diameter, multiple lesions (tumours), bilateral breast cancer, multiple cancers, inflammatory breast cancer, and breast cancer in pregnancy or lactation. Also excluded were patients who had previously been treated for breast cancer, whether by surgery (except for biopsy), radiotherapy, chemotherapy or endocrine therapy.
As a rule, the mammary tumour was excised with its surrounding tissue at least 2cm from the tumour margin, and the skin was removed in spindle shape along the tumour width. Axillary lymph node dissection was performed up to level 1. During the operation, the margins of the resected specimens were histopathologically confirmed to be tumour-free.

We first conducted the prospective, randomised study as outlined below. In addition we collected the data for patients in the retrospective study during enrolment of the prospective study. We also compared the results of the retrospective study with those of the prospective study in addition to analysis of the data of the prospective, randomised study.

Prospective, Randomised Study
After patients had undergone lumpectomy and had given informed consent for participation in the study, they were randomly assigned via a central registration method to receive either no adjuvant therapy (Group C) or adjuvant tamoxifen (Group T), which was commenced 14 days after surgery and continued at a dosage of 20 mg/day for 2 years. The period of enrolment in this study was about 2 years, from August 1989 to August 1991. The patients underwent follow-up examinations for tumour recurrence and survival over a period of 5 years postoperatively.

Retrospective Study
Patients who received radiotherapy following breast-conserving surgery served as a (retrospective) control group (Group R). The inclusion and exclusion criteria for these patients were the same as for the prospective, randomised study, and the patients underwent identical breast-conserving surgery during the same period. As they received postoperative radiotherapy, they were not enrolled in the prospective, randomised study. This group of patients also underwent follow-up examinations for tumour recurrence and survival over a period of 5 years postoperatively.

Statistical Analysis
A 2 test was used to determine differences in the distribution of patient demographic characteristics between the groups. The Kaplan-Meier method was used to calculate overall survival, disease-free survival and recurrence-free rates. The statistical significance of differences between the groups was assessed using the log-rank test at a significance level of p < 0.05.

Results
Prospective, Randomised Study
A total of 112 patients (Group C, n = 59; Group T, n = 53) were enrolled in the prospective, randomised study. There were no differences between the two groups in the distribution of patient demo-graphic characteristics such as age, menopausal status, tumour/node classification, lymph node metastases, histological type of tumour, and ER (oestrogen receptor) status (table I).
The median follow-up period was 72 months. Five-year disease-free survival rates were 86.0% in Group C and 78.5% in Group T, with no significant difference between the two groups (p = 0.308). The hazard ratio was 1.47 [95% confidence interval (CI) 0.59-3.66] (figure 1). Similarly, there was no difference in 5-year overall survival rates of 92.5% and 94.3% between Groups C and T, respectively (p = 0.745) [figure 2]. Stratified analyses by menopausal status, lymph node metastases, and ER status demonstrated no differences in the 5-year disease-free or overall survival rates between the two groups.

Figure 1. (click image to zoom) Disease-free survival (DFS) in the prospective, randomised study. Five-year DFS was 86.0% in group C (no therapy; n = 59) and 78.5% in group T (tamoxifen; n = 53) [log-rank test: p = 0.308, hazard ratio 1.47 (95% confidence interval 0.59-3.66)].

Figure 2. (click image to zoom) Overall survival (OS) in the prospective, randomised study. Five-year OS was 92.5% in group C (no therapy; n = 59) and 94.3% in group T (tamoxifen; n = 53) [log-rank test: p = 0.745].

Retrospective Study
A total of 70 patients were available for retrospective study. Most of those patients refused to enter the prospective, randomised study and then received radiotherapy (mostly 50Gy) following breast-conserving surgery based on the clinical judgement of investigators. In one of 70 patients, node status was unknown but we judged that this patient was eligible for the analysis because the node lymph metastasis was negative and tumour size was 0.7cm.
As the 5-year disease-free and overall survival rates in Groups C and T in the prospective, randomised study were comparable, these two groups were combined (Group C + T, n = 112) and compared retrospectively with the control group who received postoperative radiotherapy (Group R, n = 70). There were no differences in the distribution of patient demographic characteristics such as menopausal status, node status, lymph node metastases, or ER status between the two groups (table II).

The median follow-up period in this study was 71 months. Five-year disease-free survival rates were 82.2% in Group C + T and 86.6% in Group R, with no significant difference between the two treatment groups (p = 0.352). The hazard ratio was 0.72 (95% CI 0.33-1.59) [figure 3]. When recurrence-free rates were compared between the two groups (C + Tvs R) according to the site of tumour recurrence (i.e. within the remaining breast, local and distant metastases), no significant differences were observed at any site (figure 4). Notably, recurrence rates in the breast were comparable: 8.0% in Group C + T (nine of 112 patients) and 7.1% in Group R group (five of 70 patients) [table III]. No differences in the 5-year overall survival rate were evident between Group C + T and Group R (p = 0.224) [figure 5].

Figure 3. (click image to zoom) Disease-free survival (DFS) in the retrospective study. Five-year DFS was 82.2% in group T + C (tamoxifen or no therapy; n = 112) and 86.6% in group R (radiotherapy; n = 70) [log-rank test: p = 0.352, hazard ratio 0.72 (95% confidence interval 0.33-1.59)]

Figure 4. (click image to zoom) Recurrence-free rates in the retrospective study in groups T + C (tamoxifen or no therapy; n = 112) and R (radiotherapy; n = 70): (a) breast tumour recurrences (log-rank test: p = 0.767); (b) local recurrences (local lymph nodes and soft tissue) [log-rank test: p = 0.750]; (c) distant metastases(log-rank test: p = 0.298).

Figure 5. (click image to zoom) Overall survival (OS) in the retrospective study. Five-year OS was 93.4% in group T + C (tamoxifen or no therapy; n = 112) and 97.0% in group R (radiotherapy; n = 70) [log-rank test: p = 0.224].

Discussion
In recent years, there has been a trend to less extensive surgical resection of breast cancers, and during 1997 in Japan, breast-conserving surgery was performed in approximately 30% of all breast cancer patients.[1] Recently, efforts have been made to identify patients who do not require axillary lymph node dissection by detection of metastases in the axillary sentinel node. Numerous reports have shown that the results of breast surgery in Japanese patients are better than in Western patients,[2] and this surgical procedure appears rather suitable for Japanese women. When performing breast-conserving surgery in Japan, surgeons attempt to completely remove the tumour, and if tumour-negative margins of resected specimens are confirmed histologically, postoperative radiotherapy may not be performed. In one study, no difference in prognosis was observed after usual local excision of mammary tumours (lumpectomy) both with and without postoperative radiotherapy.[3]
In the patients enrolled in this study, strict criteria were set for tumour size, metastases to the lymph nodes, location of tumour, and distance between the margins of the tumour and nipple. Previously, these patients could obtain more than 90 to 95% of disease-free or overall survival with total mastectomy even if pectoral muscle-conserving surgery (i.e. the Patey or Auchincloss operation) was performed.[4] However, if a surgeon fully understands the various diagnostic methods and carefully performs an adequate lumpectomy, a residual tumour in patients undergoing breast-conserving surgery can largely be avoided. To achieve this, close collaboration between surgeon and pathologist is necessary, and if such precautions are taken, malignant cells can be completely resected in at least 70 to 80% of patients.[3] Thus, only 20 to 30% of such patients will have residual tumours requiring radiotherapy or subsequent mastectomy. Currently, no antitumour agent has been reported to be adequately effective in preventing tumour recurrence.

Consequently, when breast-conserving surgery with lumpectomy plus axillary lymph node dissection plus radiotherapy is performed on patients with tumour-free margins, it may be anticipated that 70 to 80% of women will undergo unnecessary radiotherapy. The US Harvard Group[5] have described a difference in local tumour recurrence between tumour-positive and tumour-negative margins. Tumour-free margins of resected specimens are determined on the basis of histopathological findings, which are extremely important along with macroscopic examination. The evaluation can be made with frozen sections, but a more accurate evaluation can be achieved with formalin-fixed specimens.

In this investigation, patients were considered to have tumour-free margins only when a pathologist determined that cancerous lesions were completely removed. In the prospective, randomised study, patients received either breast-conserving surgery alone or breast-conserving surgery with adjuvant tamoxifen therapy, but no differences in disease-free survival or overall survival were observed between the two treatment groups.

When the effect of irradiation of the remaining breast (surgery plus radiotherapy group) on the prognosis in patients with tumour-free margins was assessed in the retrospective study by comparing these patients with the combined surgery alone and surgery plus tamoxifen group, again no significant difference in disease-free survival was observed between the two treatment groups. Furthermore, there was no difference between the groups in terms of tumour recurrences in the remaining breast or in the ipsilateral regional lymph node and soft tissue, or in distant metastases. It cannot be said, however, that the evaluation of tumour-free margins is 100% accurate, and further improvement in this procedure is necessary.

One of the possible reasons why there were no differences in disease-free survival or overall survival between the two treatment groups in the prospective, randomised study is that the number of patients included in this study was not sufficient to detect a small difference. Another possible reason is that the follow-up period of this study was too short, since recurrence of breast cancer may be observed even after 5 years. These factors were assumed to contribute to the absence of difference in this study. In the retrospective study to assess the necessity for radiotherapy after breast-conserving surgery, the wide confidence interval of the hazard ratio suggests that the number of patients included in the study was not statistically sufficient. Nevertheless, the hazard ratio obtained in this study suggests that there is no clinical advantage in conducting radiation therapy after breast-conserving surgery in terms of preventing recurrence in patients with Stage I, margin-free breast cancer.

We also attempted to assess the effect of adjuvant tamoxifen in this study, but could not statistically prove its benefit in view of the low number of tumour recurrences. Long-term follow-up examinations and accumulation of a large number of cases will be necessary to assess the effect of tamoxifen in such patients. There has, however, been a report indicating the efficacy of tamoxifen (20 mg/day for 5 years) in comparison with placebo in reducing breast cancer recurrences in women with ductal carcinoma in situ treated by lumpectomy and radiotherapy,[6] and adjuvant treatment with tamoxifen is currently considered clinically beneficial. Although radiotherapy appears to contribute to prolongation of disease-free survival by tumour dormancy for about 10 years postoperatively, it may increase ipsilateral breast tumour recurrence (IBTR) at 15 to 20 years postoperatively,[7] and its late-onset adverse effects cannot be ignored.

Recent reports have shown that although IBTR was reduced when radiotherapy was added to tamoxifen therapy compared with tamoxifen alone, overall survival was identical and the incidence of contralateral breast tumour was reduced in both groups.[8,9] Another report showed that IBTR was infrequent and overall survival was the same when radiotherapy was added to tamoxifen for patients 70 years and older compared with tamoxifen alone. These reports reveal the possibility that postoperative tamoxifen administration could reduce the incidence of IBTR and produce good overall survival, even in comparison with treatment with tamoxifen plus radiotherapy. Thus, with careful surgery even better results may be obtained

Conclusion
In conclusion, our findings suggest that breast cancer patients may not always require radiotherapy after breast-conserving surgery, although the generally favourable prognosis of patients treated in this study and the low number of tumour recurrences did not allow the equivalence of the treatment regimens to be proven statistically. However, because of this finding, and the facts that local tumour recurrences have been reported at 10 to 15 years after radiotherapy and that its late-onset adverse effects after 20 to 30 years are unclear, the development of newer treatment methods to replace radiotherapy is needed. Considering that the average life expectancy of Japanese women is over 85 years, breast-conserving therapy needs to be further improved, and it is expected that hormone therapy (or other chemotherapy), which are relatively free of adverse effects, will play a significant role in future treatment regimens.

Correspondence and offprints: Dr T. Tominaga, Breast Cancer Centre, Toyosu Hospital, Showa University School of Medicine, 4-1-18 Toyosu, Koutou-ku, Tokyo, Japan.