Long-Term Oncologic Outcome of Breast Cancer in Southern Iran: A Retrospective Cohort Study

Introduction: Breast cancer remains the most frequent and lethal cancer among women worldwide. This study aimed to investigate the characteristics, prognostic factors and outcome of breast cancer patient treated and followed-up in Shiraz, southern Iran from 2000 to 2005. Methods: In this retrospective study, 1,024 patients with breast cancer who were treated in Namazi hospital of Shiraz University of Medical Sciences between 2000 and 2005 were included. Survival analysis was performed to determine potential factors influencing disease free-and overall survival in these patients. Results: Median age of the patients at diagnosis was 47 (range 19-83) years. Median follow-up for surviving patients was 68 months. The majority of patients presented at stage II (42%) and stage III (41%). Additionally, most (61%) patients had positive axillary node. The 5-, 10- and 15-year overall survival rates were 83.8%, 61.5% and 56.6% respectively. On multivariate analysis for overall survival, histologic type, Tumor stage, node stage, disease stage, M stage, hormone receptor status, adjuvant radiotherapy, and chemotherapy regimen remained independent prognostic factor for overall survival. Conclusion: in this study, the poor long-term oncologic outcome of the patients with breast cancer may be due to the higher rate of locally advanced disease; as well as the lack of modern systemic therapies in the study period.


Long-Term Oncologic Outcome of Breast Cancer in Southern Iran: A Retrospective Cohort Study
may influence the survival duration in different studies (de Boer et al., 2010;Metcalfe et al., 2019;Saad et al., 2019). In a study on more than 170,000 breast cancer patients in 2015, T-stage had reverse correlation with survival duration (Saadatmand et al., 2015). In females with BRCA2 mutation positive estrogen receptor causes shorter survival (Metcalfe et al., 2019). Regarding to an 18-year cohort study in 2016 (based on Danish cancer registry), lung, liver and brain metastasis respectively comes after bone metastasis as the most common sites of breast cancer metastasis. Patients with brain and liver metastasis had highest mortality rate during the first year after metastasis incidence (Ording et al., 2017). Radiotherapy and systemic treatments such as chemotherapy, hormone therapy, targeted therapy and immunotherapy are considered as non-surgical treatments for breast cancer (Chambers et al., 2019;Waks et al., 2019).
There is a paucity evidence regarding the long-term oncologic outcome and prognostic factors of patients with breast cancer in southern Iran. This study aimed to investigate the characteristics, prognostic factors and longterm oncologic outcome of 1,024 breast cancer patient referred to the Namazi hospital of Shiraz University of 3878 Medical sciences from 2000 to 2005.

Materials and Methods
This retrospective study performed on all women with pathologically proven invasive breast carcinoma referring to the Namazi hospital of Shiraz University of Medical Sciences. Study cases selected since January 2000 till January 2005 with follow up till January 2021. Patients had regular follow up records. We evaluate patients' status using their phone numbers and reviewing their medical records. In case of uncertainty about patients' status, we searched local databases such as insurance information. Noninvasive breast cancer such as "carcinoma in situ" and pathologies other than adenocarcinoma such as "lymphoma", "sarcoma", and "phyllodes tumor" were excluded. Tumor, Node, Metastasis (TNM) staging was defined using the 8th edition of the American Joint Committee on Cancer (AJCC) TNM staging system (Amin et al., 2017).
Study variables were Age at presentation, TNM stage, histologic subtype, estrogen and progesterone receptor status, Her2-Neu receptor status, dissected lymph nodes and positive dissected LNs ratio, Chemotherapy regimen, radiation dose, metastasis location and time, overall survival (OS) and disease free survival (DFS). Patients categorized with age cut-off point of 40 years. Date of breast cancer pathology report considered as disease presentation time. Used chemotherapy regimens were, CMF (Cyclophosphamide, Methotrexate and Fluorouracil), CAF (Cyclophosphamide, Adriamycin and Fluorouracil), CAF followed by paclitaxel and TAC (Docetaxel, Adriamycin and Cyclophosphamide). Disease Free Survival considered as the time period between surgery and the first occurred metastasis.
Data inserted in SPSS version 21 and checked for repetitive data. In the statistical tests p-value under 0.05 assumed as statistical significance. Mean and standard deviation used for quantitative data description. Student t-tests and ANOVA or non-parametric substitutes used for comparison between different groups. Kaplan-Meier survival analysis done for age groups, histology, hormone receptor status, chemotherapy, radiotherapy, M-Stage, N-Stage, stage group, positive dissected LNs. Overall and disease free survival duration compared between subgroups using Log-Rank test. Cox regression analysis assumptions met and it performed to evaluate the possible effect of mentioned factors on overall and disease free survival.

Results
January 2000 to January 2005, 1,059 patients with documented breast cancer referred to our center for nonsurgical treatments. Thirty five patients were excluded due to male gender and rare pathologic subtypes so we performed analysis on 1024 women in this study. Median age of the patients at diagnosis was 47 (range 19-83) years and 274 of them (26.8%) were under 40 years. Ductal carcinoma was the most frequent (88.6%) histologic subtype followed by medullary (5.2%) and lobular (4.5%) carcinoma. The majority of patients presented at stage II (42%) and stage III (41%). TNM staging frequencies are shown in figure 4. Additionally, most (61%) patients had a clinically positive axillary node at the time of diagnosis. In the current study, the rate of stage IV at presentation was only 1.5% which might be underestimated due to inadequate staging workup. Due to limited resources and financial problems, 27 percent of patients with stage III disease did not have CT scans or bone scans. Only 26 patients (2.8%) did not receive adjuvant chemotherapy in their treatment course. During the follow-up period, approximately one-third of the patients developed recurrences, of which, only 2% had local recurrences, and the rest having distant relapse. Bone (43%), lung (22%), brain (9%), and liver (8%) were the most common foci of the metastasis. Chemotherapy regimen of CMF including cyclophosphamide, methotrexate and 5-FU (60.1%) and CAF including cyclophosphamide, doxorubicin and 5-FU (31.9%) were the most frequent chemotherapy protocol    (Figure 1). Likewise, the 1-, 2-, 3-, 5-, 10and 15-year OS rates were 99.3%, 96.7%, 93%, 83.8%, 61.5% and 56.6% respectively (Figure 2). In the univariate analysis, age (P = 0.043), hormone receptor status (P <0.001), chemotherapy regimen (P <0.001), number of positive lymph nodes (P <0.001), node stage (P <0.001), and disease stage (P <0.001) were significant prognostic factors for disease-free survival. Likewise, Univariate analysis showed metastatic site (P <0.001), chemotherapy regimen (P <0.001), hormone receptor status (P <0.001),  number of positive lymph nodes (P <0.001), node stage (P <0.001), and disease stage (P <0.001) were significant prognostic factors for overall survival (Table 1). Figure  3 illustrates the impact of the most important prognostic factors including disease group, T-stage, N-stage, and hormone receptor status on overall survival. On multivariate COX-Regression analysis for overall survival, medullary and lobular carcinoma histology, adjuvant radiotherapy, chemotherapy regimen, hormone receptor status, node stage, disease stage and more than 3 positive dissected lymph node had affect overall survival chance independently (Table 2). In multivariate analysis to find potential risk factors for visceral and bone metastasis, more positive dissected lymph nodes (HR=1.1, p-value= 0.001) and less dissected lymph nodes (HR=0.94, p-value=0.006) are risk factors for visceral

Discussion
In this retrospective study, we found a high rate of node positivity and locally advanced disease stage in patients with breast cancer treated in southern Iran. Higher node and disease stage, hormone receptor status, histologic subtypes, and adjuvant therapies independently have negative impact on disease-free and overall survival rates in these patients.
Breast cancer is one of the major health problems in developing countries with limited resources. In Iran, breast cancer imposes a heavy financial burden on patients and the insurance system (Farhood et al., 2018). Adjuvant treatments such as radiotherapy and systemic treatments include a significant portion of these costs. Patients with more advanced diseases incur a higher cost of treatment. Therefore, recognizing prognostic factors and early diagnosis of the disease can be very useful in saving patients' lives and reducing the cost of cancer treatment (Rezapour et al., 2021). Most major reports from Iran mainly investigated the incidence and prevalence of the disease and risk factors or clinical features of the patients and lack clinical information to fully evaluate the survival rate and prognostic factors; as well as, long-term follow-up of the patients with breast cancer (Shamshirian et al., 2020).
As far as we know, this is the second largest series with the longest follow-up period of the patients with breast cancer reported in southern Iran (Rezaianzadeh et al., 2009). In this study, the median age of the patients was 47 years old, which is about a decade younger than reports from developed countries. This finding and other clinical findings such as higher rate node positivity and locally advanced disease of this study is compatible with  Abedi et al., 2016;Foroozani et al., 2020).
In developing countries such as Iran in which there is no screening program for breast cancer detection, the presentation of the disease in advanced stages may be due to a delay in diagnosis (Dianatinasab et al., 2018). In a multi-center study by Foroozani et al, factors of delay in diagnosis and advanced stage at presentation were evaluated among Iranian patients with breast cancer. They found more than 3 months delay in diagnosis in one-third of patients. Older age, low family income, lack of health insurance, rural residency, never married, postmenopausal status, history of breast disease, unawareness of breast self-examination, asymptomatic tumors, obesity, and underlying comorbidities were associated factors with the delay in diagnosis (Foroozani et al., 2020).
In most studies, age is known as a prognostic factor. Young women tend to have higher rate of local recurrence and lower survival rate than older women (Bollet et al., 2007;He et al., 2017;Langman et al., 2021). In the present study, patients less than 40 years old had lower survival rates. This finding is compatible with the most reported series in the literature (Dai et al., 2019;Zhang et al., 2018). However, in contrary to our results, some reports found that patients less than 40 years old had a higher cure rate than older ones (Mojtaba Meshkat et al., 2020). In a recent study, Kadkhodaei et al reported treatment outcomes and prognostic factors among young patients with non-metastatic breast cancer in southern Iran. In the multivariate COX regression model, conserving breast surgery and higher disease stages remained independent prognostic factors for disease-free and overall survival. (Kadkhodaei et al., 2022).
In the current study, the histological type was recognized as the prognostic factor in both univariate and multivariate analysis. Histological typing is a known prognostic factor in breast cancer. College of American Pathologists has been considered histologic type as Long-Term Oncologic Outcome of Breast Cancer in Southern Iran: A Retrospective Cohort Study Category I among all prognostic factors in breast cancer (Fitzgibbons et al., 2000). The extent of axillary lymph node involvement and disease stage are the most important and persistent prognostic variables in patients with breast cancer (Soerjomataram et al., 2008). The finding of the present study was consistent with most literature evidence and showed independent prognostic impact of the node and disease stage on DFS and OS in patients with breast cancer. In a similar study by Meshkat et al, found the number of involved lymph nodes and disease stage as independent prognostic factor in patients with breast cancer (Meshkat et al., 2020). In another study on 101 breast cancer patients, N-stage and perineural invasion were independent factors affecting disease free survival (Singh et al., 2022). In our study, higher number of dissected lymph nodes and less number of positive lymph nodes were independent risk factors for distant metastasis. In other studies factors such as histological subtype, negative hormone receptor, larger tumor size, older age and higher clinical stage mentioned as distant metastasis risk factors (Pulido et al., 2017;Slimane et al., 2004;Yao et al., 2019).
In addition, they found the overall 1, 5, 10, and 15-year survival rate were 95%, 75%, 60%, 47% respectively. Overall survival rates of the current study were consistent with study of Meshkat et al., (2020) and other reports from Iran (Ghasem Abedi et al., 2016;G. Abedi et al., 2016;Nafissi, 2008). Although there is a paucity regarding the state of DFS of breast cancer patients in Iran, DFS rates of the present study were in agreement with study of Mousavi et al they found 3-and 5-year DFS were 71.28 and 63.09 respectively (Mousavi et al., 2019).
In general, the cause of advanced stage presentation and poor survival of patients with breast cancer in our country is multifactorial and various factors such as socio-economic, cultural, nutrition, limited resource, and health behaviors determinants play an important contribution (Dianatinasab et al., 2018;Foroozani et al., 2020;Sobri et al., 2021).
Breast cancer decrease age-related life expectancy in our patients comparing to normal Iranian females with same age group (more dominant in younger cases). We followed our patients for 21 years but the expected life expectancy were far more than our follow up duration in younger age groups. Although none of our patients in younger age groups lived more than their expected life expectancy, significant number of them were alive at the end of study. On the other hand, more than half of the older patients had survival duration equal or more than their expected life expectancy. In conclusion, to determine the effect of breast cancer on life expectancy of younger patients longer follow up duration required.
Limitations of the study include the retrospective nature of the study, non-uniform pathologic reports of patients' medical records, the lack of data regarding some pathological features such as tumor grade, and the molecular status of the tumor, such as ER, PR and HER2.
Strengths of this study include large population sample size and long-term follow up, recognizing more evidence of prognostic factors, early diagnosis of the disease and reducing the cost of breast cancer treatment in southern Iran.

Author Contribution Statement
All authors contributed equally in this study.