Radiol Oncol 2006; 40(1): 23-8.
Locoregional control and survival after breast conserving
therapy
Mirjana Rajer, Elga Majdič
Department of Radiotherapy, Institute of Oncology Ljubljana, Slovenia,
Background. The purpose of our study was to present a 5-year survival and locoregional control rates in breast cancer patients and to establish eventual impact of the treatment and patient characteristics on lo-coregional control and survival.
Methods. From January 1998 to December 1999 564 stage 1 and 2 breast cancer patients were treated with breast conserving therapy. We evaluated the following characteristics: age, histological diagnosis, grade, size, number of metastatic lymph nodes, hormonal receptor status, extensive intraductal component (EIDC), vas-cular invasion, pathologic tumour margins, type of surgery and use of adjuvant therapy. Results. The mean age of our patients was 54.2 years. Invasive ductal carcinoma was the most common di-agnosis (82.4%), followed by invasive lobular carcinoma (10.6%). Most of the tumours were grade 2. Seventy-two % of patients had T1 tumours, 24% T2 and 3% Tis tumours. Metastatic lymph nodes were pres-ent in 44% of patients. All patients were treated with breast conserving surgery followed by radiotherapy (RT). Fifty % of patients received adjuvant chemotherapy and/ or hormonal therapy. The 5-year survival rate was 88.5%. Tumour size, number of metastatic lymph nodes, grade, hormonal receptors and vascular invasion proved to be statistically significant prognostic factors for the survival, while age and histological diagnosis were not. Local recurrence developed in 4.3% of our patients, while in 3.4% regional recurrence developed.
Conclusions. Breast conserving surgery followed by RT was associated with good rates of locoregional con-trol and survival, comparable to those reported in the literature.
Key words: breast neoplasms – surgery; survival analysis
Introduction
Received 20 February 2006 Accepted 28 February 2006
Correspondence to: Mirjana Rajer, MD, Department of Radiotherapy, Institute of Oncology Ljubljana, Zaloška 2, 1000 Ljubljana, Slovenia; Phone: + 386 41 26 99 46; Fax.: + 386 1 587 9 400; E-mail: mrajer@onko-i.si
Breast conserving therapy has been used since the 1960s and is now considered best practice in the treatment of early breast can-cer. Retrospective and prospective random-ized trials demonstrated that breast conserv-ing therapy (BCT) produces rates of survival
24
Rajer M and Majdič E / Breast conserving therapy
and locoregional control similar to those of mastectomy.1 Breast conserving surgery re-moves a detectable disease in the breast and/or regional lymph nodes, but has no ef-fect on possible undetected disease in the re-maining breast, chest wall, regional lymph nodes or distant sites.2 By combining surgery, radiotherapy and adjuvant systemic therapy we can lower the risk of locoregional and dis-tant recurrence.
In spite of this combined therapy there are still patients who develop recurrences. Local recurrence rates of 5-20% have been reported in different studies after BCT.1-3 Although the impact of local recurrence on overall survival is not well established, it has a detrimental psychological effect on the patient.1,4,5 By identifying those patients who have higher risk of developing a locoregional and/or dis-tant recurrence, we can determine the right treatment to minimize the risk.1
The purpose of this retrospective study was twofold. First of all, to present a 5-year sur-vival and locoregional control rates in patients treated at Institute of Oncology in Ljubljana and, secondly, to establish the eventual im-pact of treatment and patient characteristics on locoregional control and survival.
Methods
From January 1998 to December 1999 564 breast cancer patients stage I and II were treated at the Institute of Oncology Ljubljana with breast conserving therapy. The patient data were obtained by medical records. We evaluated the following patient and tumour characteristics: age, histological diagnosis, grade, size, number of metastatic lymph nodes, hormonal receptor status, extensive intraductal component (EIDC), vascular invasion, pathologic tumour margins, type of sur-gery and use of adjuvant therapy. We record-ed eventual locoregional and/or distant recur-rence by the retrospective analysis.
Radiol Oncol 2006; 40(1): 23-8.
Statistical univariate analyses were done using the SPSS program. A statistical signifi-cance was assessed with Log-rank, Breslow and Tarone-Ware tests.
Results
Patient and tumour characteristics
The patient’s age ranged from 28 to 77 years. The mean age was 54.2 years. The age distribution is presented on the histogram (Figure 1).
Most of the tumours were present in the upper outer quadrant (45%), followed by the outer lower quadrant (11%). Invasive ductal carcinoma (IDC) was by far the most common histological diagnosis (82.4% of tumours), fol-lowed by invasive lobular carcinoma (ILC) (10.6%) and ductal carcinoma in situ (DCIS) (2.6%). Other histological types of tumours were rare. The grade was more evenly distrib-uted. The most common was Grade 2 with 44%, followed by Grade 3 with 33% and Grade 1 with close to 26%.
Most of the tumours were T1 (72%) and T2 (42%). Tis was present in 3% of cases. Sixty-six % of patients had negative axillary lymph nodes, 25.7% had one to three metastatic lymph nodes and 7.6% had more than three metastatic nodes.
Positive estrogen receptors were detected in 66.4% of tumours and positive proges-
Figure 1. Age distribution of the patients.
Rajer M and Majdič E / Breast conserving therapy
25
Figure 2. Overall survival of patients.
Figure 3. Impact of the tumour size on survival (p = 0.000).
terone receptors in 51.1%. Vascular invasion was present in 2.6% of tumours and 15% of tumours had an EIDC.
Treatment
All patients were treated with breast conserv-ing operation, either tumorectomy or quad-rantectomy. All patients received postoperative radiotherapy with two tangential fields. The total dose was 50 Gy followed by a boost with electrons to the tumour bed of 10 to 16 Gy. In patients with more than three metasta-tic axillary lymph nodes, 50 Gy was given to the supraclavicular fossa. No patient received radiotherapy to the axilla. About half of pa-tients received adjuvant chemotherapy (54%) and hormonal therapy (50.4%).
Survival
The 5-year overall survival rate was 88.5% (Figure 2).
Prognostic factors for survival
In the context of our analysis, age and his-tology proved statistically non significant, while the influence of tumour size, nodes, grade, estrogen and progesterone receptors, vascular invasion and local and regional fail-ure was significant.
As expected, a larger tumour leads to a lower survival probability (Figure 3). In this case as well as in other presented graphs, that will follow, the difference was statistically significant.
Comparing the overall survival rate (88.5%), in patients with Grade 3 tumours it was well below 80%, while for the patients with Grade 1 tumours it was close to 100% (Figure 4).
The patients who had tumours with positive estrogen and progesterone receptors had higher survival probability (Figures 5, 6).
The patients with a local recurrence had lower survival probability (Figure 7). The same holds for regional recurrence (Figure 8). For these patients the 5-year survival was just above 20%.
Vascular invasion proved to be a statistical-ly significant prognostic factor. For patients with tumours with vascular invasion, the 5-year survival was just above 20%. We have to mention the problem of missing data for this particular prognostic factor, so we should be careful when interpreting the results.
Prognostic factors for local recurrence
The local recurrence occurred in 4.3 % of patients. Since the number of cases with local recurrence was small we used a simple cross-tabulation analysis (Table 1).
Radiol Oncol 2006; 40(1): 23-8.
26
Rajer M and Majdič E / Breast conserving therapy
Survival Functions
Survival Functions
Grade J11
3 -\-   1 -censored +   2-censored
3-censored
Time to event in months
Figure 4. Impact of the tumour grade on survival (p = 0.000).
estrogenski _— negative P| not assessed
positive -J-  negative-censored ___  not assessed-
censored
positive-censored
Time to event in months
Figure 5. Impact of the tumour estrogen receptors on survival (p= 0,004).
Tumour grade, number of metastatic lymph nodes, surgical margins and hormonal receptors were prognostic factors for development of local recurrence.
Prognostic factors for regional recurrence
The regional recurrence occurred in 3.4% of patients. The total number of recurrences was 19. The total number of patients with metastatic axillary lymph nodes was 188. One recurrence was in the axilla, one in the parasternal region and 17 in the supraclavic-ular region (Table 2).
The small number of events is a problem for statistical analysis. However, the number of metastatic lymph nodes in the axilla is a risk factor for supraclavicular recurrence in our patients.
Discussion
In the study we tried to evaluate the 5-year survival and locoregional control in our pa-tients after BCT and to establish eventual prognostic factors. The overall 5-year sur-vival was 88.5%. This result is similar to oth-er studies.1 Prognostic factors that had an impact on the survival were tumour size, nodes, grade, estrogen and progesterone re-ceptors, vascular invasion and local and regional failure.
As expected, larger tumours led to lower survival probability. There was a marked dif-ference between tumours with diameter less than 2 cm compared to those with more than 2 cm (90% versus 70% 5-year survival).
There was also a marked difference re-garding the tumour grade with lesser survival
Table 1. Prognostic factors for local recurrence	
Prognostic factor	% of Local recurrence
Grade	Grade 1 – 1.8% Grade 2 – 3.6% Grade 3 – 7.8%
Number of metastatic axillary lymph nodes	0 – 2.7% <3 – 5.5% >3 – 14%
Margins	Free margins – 2.7% Involved – 12.5%
Estrogen, progesterone	Negative receptors › higher chance of local relapse
Radiol Oncol 2006; 40(1): 23-8.
Rajer M and Majdič E / Breast conserving therapy
27
Figure 6. Impact of the tumour progesterone receptors on survival (p= 0.03).
probability in patients with less differentiated tumours.
The number of metastatic axillary lymph nodes is a well established prognostic factor2 which was also confirmed in our study. Patients with negative lymph nodes had 90% 5-year survival probability, while those with more than three metastatic nodes had only 40%.
The presence or absence of hormone re-ceptors had also an impact on the survival. Those patients who had positive receptors had higher survival probability. The same is true for patients who had tumours without vascular invasion. We have to mention that a lot of data was missing for this variable, so we
Figure 7. Impact of the tumour local recurrence on survival (p= 0,000).
Table 2. Number of metastatic lymph nodes and percentage of regional recurrence
No of metastatic	% of regional
lymph nodes	recurrence
0 1-3 >3	1. 6.3 9
can not be conclusive regarding this prognos-tic factor.
Local and regional failure influenced the survival. A recent meta-analysis confirmed the impact of local recurrence after BCT on survival which is similar to the impact of a re-currence after modified radical mastectomy. It is estimated that for 4 local relapses avoid-ed 1 life will be saved.1,2,6
Interestingly, age was not a significant prognostic factor for the survival.
The local recurrence occurred in 4.3% of patients. This is comparable to other studies, where the local failure ranges from 1.2-20%.1,7-9
We tried to identify the prognostic factors for local recurrence, but the small number of patients with local recurrences made the analysis difficult. We found some impact of grade (higher grade, more recurrences), num-ber of metastatic lymph nodes (more metasta-tic lymph nodes, more relapses), surgical re-section (patients with involved margins had
Figure 8. Impact of the tumour regional recurrence on survival (p= 0,000).
Radiol Oncol 2006; 40(1): 23-8.
28
Rajer M and Majdič E / Breast conserving therapy
more local recurrences) and the receptor status (positive receptors, better prognosis).
The regional recurrence was less common, with 3.4% of patients. Most of the regional re-currences developed in the supraclavicular fossa and only one in the axilla.
Therefore we conclude that there is no need to irradiate the axilla after an axillary dissection even if metastatic lymph nodes were found at the operation.
Conclusions
Survival and locoregional control rates in our patients are comparable to those reported in the literature. Axillary recurrence is rare after an axillary dissection even in patients with >3 metastatic lymph nodes without RT to the ax-illa.
6.   Whelan T, Clark R, Roberts R, Levine M, Foster G. Ipsilateral breast tumor recurrence post lumpecto-my is predictive of subsequent mortality: results from a randomized trial. Int J Radiat Oncol Biol Phys 1994; 30: 11-6.
7.   Straus K, Lichter A, Lippman M, Danforth D, Swain S, Cowan K, et al. Results of the national cancer institute early breast cancer trial. J Natl Cancer Inst Monogr 1992; 11: 27-32.
8.   Cabioglu N, Hunt KK, Buchholz TA, Mirza N, Singletary SE, Kuerer HM, et.al. Improving local control with breast conserving therapy: a 27 year single institute experience. Cancer 2005; 104: 20-9.
9.   Vinh-Hung V, Verschraegen C. Breast-conserving surgery with or without radiotherapy: pooled-analysis for risks of ipsilateral breast tumor recur-rence and mortality. J Natl Cancer Inst 2004; 96: 115-21.
References
1.   Mirza NQ, Vlastos G, Meric F, Bucholz TA, Esnaola N, Singletary E, et.al. Predictors of locore-gional recurrence among patients with early stage breast cancer treated with breast conserving ther-apy. Ann Sur Oncol 2002; 9: 256-65.
2.   Clarke M, Collins R, Darby S, Davies C, Elphinstone P, Evans E, et al; Early Breast Cancer Trialists’ Collaborative Group (EBCTCG). Effects of radiotherapy and of differences in the extent of surgery for early breast cancer on local recurrence and 15-year survival: an overview of the ran-domised trials. Lancet 2005; 366: 2087-106.
3.   Majdič E. Can axillary treatment in selected breast cancer patients be avoided? Radiol Oncol 2000; 34: 255.
4.   Van der Hage JA, Putter H, Bonnema J, Bartelink H, Therasse P, Van der Velde CHJ. Impact on lo-coregional treatment on the early stage breast can-cer patients: a retrospective analysis. Eur J Cancer 2003; 39: 2192-99.
5.   Dinshaw KA, Budrukkar AN, Chinoy RF, Sarin R, Badwe R, Hawaldar B, et al. Profile of prognostic factors in 1022 indian women with early stage breast cancer treated with breast conserving ther-apy. Int J Radiat Oncol Biol Phys 2005; 36: 1132-41.
Radiol Oncol 2006; 40(1): 23-8.
Slovenian abstracts
59
Radiol Oncol 2006; 40(1): 23-8.
5-letno preživetje ter pogostnost lokalne in regionalne ponovitve bolezni pri bolnicah z rakom dojk, ki smo jih zdravili z ohranitveno
operacijo in obsevanjem
Rajer M, Majdič E
Izhodišča. Ohranitveno zdravljenje je del standardne terapije bolnic z začetnim rakom dojk. V naši raziskavi smo skušali ugotoviti, kakšno je 5-letno preživetje, pogostnost lokalne in regionalne ponovitve bolezni ter ugotoviti morebitne dejavnike tveganja, ki na to vplivajo. Metode. Od januarja 1998 do decembra 1999 smo na Onkološkem inštitutu v Ljubljani zdravili 564 bolnic z rakom dojk v prvem in drugem stadiju bolezni. Pri vseh smo določili naslednje dejavnike: starost, histološko diagnozo, stopnjo malignosti, velikost tumorja, število metastatskih pazdušnih bezgavk, prisotnost hormonskih receptorjev, ekstenzivno intraduktalno komponento, vaskularno invazijo, stanje resekcijskih robov, vrsto operacije in način dopolnilnega zdravljenja.
Rezultati. Povprečna starost bolnic je bila 54,2 let. Najpogostejša histološka diagnoza je bila in-vazivni duktalni karcinom. Največ tumorjev je bilo druge stopnje malignosti. Večina tumorjev je bila T1 (72%), sledijo tumorji T2 (24%). Metastaze v pazdušnih bezgavkah je imelo 44% bolnic. Vse so bile zdravljene z ohranitveno operacijo in pooperativnim obsevanjem operirane dojke. Polovica bolnic je dobila dopolnilno kemoterapijo in hormonsko terapijo. Celokupno petletno preživetje je bilo 88,5%. Statistično značilni napovedni dejavniki so bili: velikost tumorja, število metastatskih pazdušnih bezgavk, stopnja malignosti, prisotnost hormonskih receptorjev in vaskularna invazija. Starost bolnic in histološki tip tumorja nista vplivala na preživetje. Lokalno ponovitev bolezni smo ugotovili pri 4,3% bolnic, regionalno pri 3,4%. Od 19 regionalnih ponovitev bolezni je bil ena v pazduhi, ena parasternalno in ostale v supraklavikularni loži. Zaključki. Ugotovili smo, da sta celokupno 5-letno preživetje ter pogostnost lokalne in regionalne ponovitve bolezni pri naših bolnicah primerljivi z rezultati v drugih raziskavah. Stopnja ma-lignosti, hormonski status in število pozitivnih bezgavk so napovedni dejavniki, ki napovedujejo tako celokupno preživetje kot lokoregionalno ponovitev bolezni, medtem ko se v naši raziskavi starost bolnic ni pokazala kot napovedni dejavnik.
Radiol Oncol 2006; 40(1): 57-62.