Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Journal of the Egyptian National Cancer Institute (2014) 26, 31–35 Cairo University Journal of the Egyptian National Cancer Institute www.nci.cu.adu.eg www.sciencedirect.com Full Length Article Prognostic value of lymph node ratio in node-positive breast cancer in Egyptian patients Tawfik R. Elkhodary Nermeen A. Niazy c a b c a,* , Mohamed A. Ebrahim a, Elsayed E. Hatata b, Mansoura University, Oncology Center, Faculty of Medicine, Mansoura, Egypt Mansoura University, Specialized Medical Hospital, Faculty of Medicine, Mansoura, Egypt Mansoura University, Public Health & Community Medicine, Faculty of Medicine, Mansoura, Egypt Received 12 June 2013; accepted 7 October 2013 Available online 9 November 2013 KEYWORDS Breast cancer; Lymph node ratio; Prognosis; Egypt Abstract Background: Breast cancer in Egypt is the most common cancer among women and is the leading cause of cancer mortality. Traditionally, axillary lymph node involvement is considered among the most important prognostic factors in breast cancer. Nonetheless, accumulating evidence suggests that axillary lymph node ratio should be considered as an alternative to classical pN classification. Materials and methods: We performed a retrospective analysis of patients with operable nodepositive breast cancer, to investigate the prognostic significance of axillary lymph node ratio. Results: Five-hundred patients were considered eligible for the analysis. Median follow-up was 35 months (95% CI 32–37 months), the median disease-free survival (DFS) was 49 months (95% CI, 46.4–52.2 months). The classification of patients based on pN staging system failed to prognosticate DFS in the multivariate analysis. Conversely, grade 3 tumors, and the intermediate (>0.20 to 60.65) and high (>0.65) LNR were the only variables that were independently associated with adverse DFS. The overall survival (OS) in this series was 69 months (95% CI 60–77). Conclusion: The analysis of outcome of patients with early breast cancer in Egypt identified the adverse prognostic effects of high tumor grade, ER negativity and intermediate and high LNR on DFS. If the utility of the LNR is validated in other studies, it may replace the use of absolute number of axillary lymph nodes. ª 2013 Production and hosting by Elsevier B.V. on behalf of National Cancer Institute, Cairo University. * Corresponding author. Address: Faculty of Medicine, Mansoura University, PO Box 35516, Mansoura, Egypt. Tel.: +20 1007291213. E-mail address: [email protected] (T.R. Elkhodary). Peer review under responsibility of The National Cancer Institute, Cairo University. Background In Egypt, breast cancer is estimated to be the most common cancer among females accounting for 37.7% of their total with 12,621 new cases in 2008. It is also the leading cause of cancer related mortality accounting for 29.1% of their total with 6546 Production and hosting by Elsevier 1110-0362 ª 2013 Production and hosting by Elsevier B.V. on behalf of National Cancer Institute, Cairo University. http://dx.doi.org/10.1016/j.jnci.2013.10.001 32 deaths [1]. These estimates are confirmed in many regional Egyptian cancer registries [2,3]. There are variations in breast cancer patterns between developed and developing countries. Higher tumor stage at diagnosis due to the lack of widely implemented national screening programs in Egypt just started few years ago [4]. Also younger age is perhaps more seen in developing nations too and this could be well explained by the different population pyramids being predominantly young in the developing countries [5]. Traditionally, axillary lymph node involvement and the increasing absolute number of positive lymph nodes (pN) are considered among the most important prognostic factors in breast cancer [6–8]. Nonetheless, in a recent report using the Geneva Cancer Registry data, the authors established that certain identified cutoff points for axillary lymph node ratio (LNR; ratio of positive over excised lymph nodes) predicted breast cancer survival more accurately than the classical pN classification [9]. The report recommended that LNR should be considered as an alternative to the pN staging. Similarly, a systematic review of 24 articles including 32,299 patients, also concluded that the LNR was superior to the number of involved nodes as a prognostic indicator [10]. New evidence has continued to accrue confirming the prognostic significance of nodal ratios in various worldwide population settings [11]. The analysis of the survival outcome of mostly young patients with early breast cancer in Saudi Arabia identified the adverse prognostic effects of young age, high grade of tumor and intermediate and high LNR on DFS [12]. Prompted by these reports and influenced by the above allotted differences in breast cancer patterns in Egypt compared with that in Western countries, we planned this analysis to examine the prognostic role of LNR on disease free survival (primary objective) and overall survival (secondary objective) of patients with breast cancer. We also intended to highlight the clinicopathologic features of breast cancer in our locality. Materials and methods We conducted a retrospective review of 1290 breast cancer patients treated at the Oncology center, Mansoura University between April 2006 and May 2012. Data were retrieved from the electronic hospital-based cancer registry database of the Department of Oncology at Mansoura University Oncology Center, Mansoura, Egypt, until May 2013. Only patients with invasive non-metastatic breast cancer, who underwent axillary lymph node dissection and had one or more positive axillary lymph nodes, were considered eligible. Patients must not have received neoadjuvant chemotherapy. The final number available was 500 patients, representing the study population. A database was constructed to record the following information for each patient: patient characteristics, diagnostic methods, histological features of the tumor including immunohistochemistry, number of dissected and positive lymph nodes, LNR, treatment details, recurrence data, survival, and cause of death. For the survival analysis, we used the LNR cutoff values as identified by Vinh-Hung et al. [9]. The endpoints for the survival analyses were disease-free (DFS) and overall survival (OS). For patients who were lost to follow-up, their survival status at last contact was used. T.R. Elkhodary et al. Definitions Disease stage was defined according to the criteria laid down by the International Union Against Cancer, with group clinical and pathological staging according to the American Joint Committee on Cancer. OS was estimated from date of diagnosis to date of last follow-up or death from breast cancer. DFS was defined as the time between date of surgery and date of last follow-up or date of best available evidence of the first unfavorable event: local recurrence, locoregional recurrence, distant metastases, or death. Loco-regional recurrence was defined as any recurrence in the ipsilateral, chest wall, or lymph node. Contralateral breast recurrence was considered as distant metastases. Statistical methods Survival functions were estimated applying the method of Kaplan and Meier [13], while the statistical procedure of Brookmeyer–Crowley was used to estimate the 95% confidence interval (CI) of median survival [14]. The log-rank test was used to assess the significance of unadjusted differences in survival [15]. Exploring variables for their independent prognostic effect on survival was carried out using the multivariate stepwise Cox’s proportional regression hazard model [16,17], and the goodness of fit was judged by plotting the cumulative baseline hazard function for residuals [18]. We also compared the survival functions for variables after stratifying for baseline differences in additional variables [19]. A two-tailed p value of <0.05 was considered statistically significant. The software IBM-SPSS version 20 was used for all statistical evaluations. The study was approved by the Institutional Review Board. Results Five-hundred patients were considered eligible for the analysis. None of the patients had sentinel node procedure. Table 1 depicts patient and disease characteristics. The mean age was 50 years, with 12% (60 patients) and 88% (440 patients) below and above 35 years, respectively. Patients in this series showed a relatively high prevalence of poor prognostic features: 90.6% had more than T1 tumor, 92.8% presented with grade 2 or 3 tumors, 63% and 30.4% had 4 or more and 10 or more positive lymph nodes respectively, around 34.3% had negative ER, and 70% had intermediate or high LNR. Moreover, the median tumor size was 3.7 cm, with 24% and 8% of patients having tumor P5 and P10 cm, respectively. The median follow-up was 35 months (95% CI 32–37 months), the minimum follow up was 12 months while the maximum follow up was 82 months. Disease-free survival The median DFS was 49 months (95% CI 46.4–52.2 months). Table 2 shows the univariate analysis of the influence of various factors on DFS. Patient age 635 years, T2, T3 or T4 tumor, tumor that was grade 2 or 3, tumor that was N2 or N3, tumor with intermediate or high LNR, and ER or PR Prognostic value of lymph node ratio in node-positive Table 1 33 Patients and disease characteristics. Parameter No. % Mean (SD) 635 >35 50 (11) 60 440 12.0 88.0 T stage T1 T2 T3 T4 47 276 156 21 9.4 55.2 31.2 4.2 Tumor grade GI GII GIII 36 369 95 7.2 73.8 19.0 LN stage pN1 pN2 pN3 185 163 152 37.0 32.6 30.4 LN ratio Median (range) 60.2 0.2 & 60.65 >0.65 0.4 (0.04–1.0) 150 190 160 30.0 38.0 32.0 ER (496 cases) ER-ve ER + ve 170 326 34.3 65.7 PR (496 cases) PR-ve PR + ve 163 333 32.9 67.1 Her-2 status (277 cases) Her-2Her-2+ 215 62 77.6 22.4 Surgery MRM BCS 467 33 93.4 6.6 Adjuvant chemotherapy No chemotherapy Anthracycline-based Anthracycline & taxane Others 14 263 211 12 2.8 52.6 42.2 2.4 Adjuvant endocrine therapy No endocrine therapy Tamoxifen Aromatase inhibitor Tamoxifen and aromatase inhibitor 185 257 19 39 37.0 51.4 3.8 7.8 Adjuvant radiotherapy No radiotherapy Adjuvant radiotherapy 126 374 25.2 74.8 Relapse site (205 cases) Skin & LNs Bone Lung Liver Multiple Brain 43 39 28 32 64 7 Age (years) negative tumors were adversely associated with inferior DFS. The LNR delineates DFS differences between the three risk categories (Fig. 1). Variables found significant in the univariate analysis were tested in the multivariate Cox regression analysis. Table 3 shows that grade 3 tumors, ER negativity, and the intermediate and high LNR categories were the only variables that were independently associated with adverse DFS. In that multivariate analysis, compared with patients within the low LNR risk group, the hazard ratio of breast cancer recurrence risk was 2.2 (95% CI 1.5–4.9) for patients in the intermediate LNR risk group and 3.2 (95% CI 1.9–5.5) for patients within the high LNR risk group. Classifying patients according to pN1, pN2, or pN3 status was not predictive for DFS. Overall survival The OS of the 500 patients in this series was 69 months (95% CI 60–77). Only ER ve tumor was shown to adversely influence OS with 2-fold (95% CI 1.4–2.9), an increase in the risk of dying of breast cancer. Discussion Recognizing that axillary lymph node status is the most powerful predictor of outcome for women with breast cancer [6,7], the new American Joint Committee on Breast Cancer staging 34 Table 2 T.R. Elkhodary et al. Univariate analysis of the effects of prognostic variables on DFS. Variable No. Median DSF (months) 3 Year DFS (%) Age p Hazard ratio 95.0% CI 1.55 1 1.09–2.29 0.03 635 >35 60 440 34 51 61 79 T1 T2 T3 T4 47 276 156 21 58 51 42 31 81 77 50 38 Grade I Grade II Grade III 36 369 95 62 47 30 74 62 54 N1 N2 N3 185 163 152 58 53 38 79 64 49 60.2 0.2 & 60.65 >0.65 150 190 160 68 50 28 85 63 43 ve +ve 170 326 26 55 52 87 ve +ve 163 333 30 53 58 79 T stage <0.001 1 1.55 2.75 3.91 0.93–2.79 1.51–4.81 1.96–7.83 1 2.46 4.82 1.39–4.02 2.42–9.75 <0.001 <0.001 <0.001 1.959 3.479 1 1.362–2.819 2.457–4.926 <0.001 <0.001 <0.001 2.807 4.480 1 1.856–4.245 2.984–6.725 0.11 0.001 <0.001 <0.001 Grade 0.007 <0.001 <0.001 LN stage LN ratio ER PR 2.56 1 1.93–3.33 2.04 1 1.55–2.69 <0.001 Table 3 Multivariate analysis of the effects of prognostic variables on disease free survival. Variable p Grade 0.004 Grade I Grade II Grade III LN ratio 60.2 0.2 and 60.65 >0.65 ER +ve ve Figure 1 ratio. DFS of studied cases classified according to the LN group patients with increasing absolute number of positive nodes into increasingly higher nodal stages: pathologic pN1, one to three positive nodes; pN2, four to nine positive nodes; pN3, 10 or more positive nodes [8]. Thus, the likelihood of finding all of the positive nodes in the axilla is dependent on the extent of axillary dissection and the extent of pathologic examination. There is evidence that the likelihood of finding positive nodes in the axilla increases with the number of nodes removed. Therefore, in the current staging system, the prognostic value of the absolute number of nodes removed for predicting disease burden in the axilla is confounded by the number of nodes removed. 0.11 0.004 0.01 .01 0.003 0.01 Hazard ratio 95.0% CI 1 2.2 3.4 0.8–3.8 2.1–7.1 1 2.2 3.2 1.5–4.9 1.9–5.5 1 2.1 1.4–3.3 In a developing country like Egypt, a proportion of breast surgeries was performed at low-volume community hospitals with less than optimal axillary lymph nodes retrieval [20]. In our series where some patients are operated prior to referral to our center, 93 cases (18.4%) had inadequate axillary dissection, i.e. <10 lymph nodes. LNR may be considered as an alternative prognostic factor that may overcome that limitation. The utility of this variable has been tested and proven in several published reports and summarized in the meta-analysis of Woodward et al. [10]. In exploring the prognostic value of LNR in our series, we elected to use the same cutoff points identified by Vinh-Hung et al. [9], despite as the authors indicated, that different authors have used different cutoff points to classify patients into several risk groups [21–23]. Prognostic value of lymph node ratio in node-positive In the current series, LNR together with ER negativity and tumor grade were the only factors that were found to influence DFS in the multivariate analysis. Patients with intermediate and high LNRs were associated with a hazard ratio of recurrence of 2.2 and 3.2, respectively as compared with those in the low risk category. Analysis of the OS only identified ER- tumor as the only prognostic factor associated with an increased risk of dying from breast cancer. It is conceivable that the infrequent number of deaths, so far, may account of the failure of identifying the prognostic value of other variables. Perhaps, reanalysis of this series after a longer follow-up may reveal the influence of additional variables. This was clear in a similar retrospective Lebanese study that included patients diagnosed with breast cancer over a longer period (21 years). On univariate analysis, both the absolute number of positive lymph nodes and the LNR were significant predictors of OS. On multivariate analysis, only the LNR remained an independent predictor of OS, with a 2.5-fold increased risk of dying at an LNR of P0.25 [24]. Conclusion The analysis of survival outcome of Egyptian patients with early breast cancer identified the adverse prognostic effects of high tumor grade, ER negativity, and intermediate and high LNR on DFS. Our findings reinforce the importance of LNR rather than the pN staging as an important prognostic variable in breast cancer and it may replace the use of absolute number of axillary lymph nodes. Conflict of interest We have no conflict of interest to declare. References [1] GLOBOCAN 2008 v1.2. Cancer Incidence and Mortality Worldwide: IARC Cancer base no. 10. International Agency for Research on Cancer, http://globocan.iarc.fr/; 2008 [accessed June 2013]. [2] The National Cancer Registry Program of Egypt (NCRPE). Reports and Statistics: Aswan, Damietta & El-Minia; 2013 http://www.cancerregistry.gov.eg/reports.aspx [accessed June 2013]. [3] The Gharbiah Population-based Cancer Registry (GPCR). Cancer in Egypt, Gharbiah; 2007 http://www.emro.who.int/ ncd/pdf/cancer_registry_Egypt.pdf [accessed June 2013]. [4] Salem DS, Kamal RM, Helal MH, et al. Women health outreach program; a new experience for all Egyptian women. J Egypt Natl Cancer Inst 2008;20:313–22. [5] Akarolo-Anthony SN, Ogundiran TO, Adebamowo CA. Emerging breast cancer epidemic: evidence from Africa. Breast Cancer Res 2010;12(Suppl. 4):S8. 35 [6] Masood S. Prognostic/predictive factors in breast cancer. Clin Lab Med 2005;25:809–25 (viii). [7] Kapoor A, Vogel VG. Prognostic factors for breast cancer and their use in the clinical setting. Expert Rev Anticancer Ther 2005;5:269–81. [8] Singletary SE, Allred C, Ashley P, et al. Revision of the American joint committee on cancer staging system for breast cancer. J Clin Oncol 2002;20:3628–36. [9] Vinh-Hung V, Verkooijen HM, Fioretta G, et al. Lymph node ratio as an alternative to pN staging in node-positive breast cancer. J Clin Oncol 2009;27:1062–8. [10] Woodward WA, Vinh-Hung V, Ueno NT, et al. Prognostic value of nodal ratios in node-positive breast cancer. J Clin Oncol 2006;24:2910–6. [11] Vinh-Hung V, Nguyen NP, Cserni G, et al. Prognostic value of nodal ratios in node-positive breast cancer: a compiled update. Future Oncol 2009;5:1585–603. [12] Ibrahim EM, Elkhodary TR, Zekri JM, et al. Prognostic value of lymph node ratio in poor prognosis node-positive breast cancer patients in Saudi Arabia. Asia-Pac J Clin Oncol 2010;6:130–7. [13] Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958;53:457–81. [14] Brookmeyer R, Crowley JA. Confidence interval for the median survival time. Biometrics 1982;38:29–41. [15] Mantel N. Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chemother Rep 1966;50:163–70. [16] Cox DR, Oakes D. Regression models and life tables (with discussion). J Roy Stat Soc 1997;B34:187–220. [17] Marubini E, Valsecchi MG. Analysing data from clinical trials and observational studies. New York, NY: Wiley; 1995. [18] Kay R. Proportional hazard regression models and the analysis of censored survival data. Appl Stat 1977;26:227–37. [19] Kalbfleisch JD, Prentice RL. The statistical analysis of Failure time data. New York, NY: Wiley; 1980. [20] Al Jiffry B, Ezzat A, Hildell J, et al. Surgical management of breast cancer in Saudi Arabia: a call for improvement. Ann Saudi Med 1998;18:531–3. [21] Tsuchiya A, Kanno M, Abe R. The impact of lymph node metastases on the survival of breast cancer patients with ten or more positive lymph nodes. Surg Today 1997;27:902–6. [22] Overgaard M, Hansen PS, Overgaard J, et al. Postoperative radiotherapy in high-risk premenopausal women with breast cancer who receive adjuvant chemotherapy. Danish Breast Cancer Cooperative Group 82b Trial. N Engl J Med 1997;337:949–55. [23] Knoop AS, Bentzen SM, Nielsen MM, Rasmussen BB, Rose C. Value of epidermal growth factor receptor, HER2, p53, and steroid receptors in predicting the efficacy of tamoxifen in highrisk postmenopausal breast cancer patients. J Clin Oncol 2001;19:3376–84. [24] Hatoum HA, Jamali FR, El-Saghir NS, et al. Ratio between positive lymph nodes and total excised axillary lymph nodes as an independent prognostic factor for overall survival in patients with nonmetastatic lymph node-positive breast cancer. Indian J Surg Oncol 2010;1:305–12.