Pregnancy-associated breast cancer in rural Rwanda: the experience of the Butaro Cancer Center of Excellence

Pregnancy-associated breast cancer in rural Rwanda: the experience of the Butaro Cancer Center of... Background: Breast cancer is the most common malignancy encountered during pregnancy. However, the burden of pregnancy-associated breast cancer (PABC) and subsequent care is understudied in sub-Saharan Africa (SSA). Here, we describe the characteristics, diagnostic delays and treatment of women with PABC seeking care at a rural cancer referral facility in Rwanda. Methods: Data from female patients aged 18–50 years with pathologically confirmed breast cancer who presented for treatment between July 1, 2012 and February 28, 2014 were retrospectively reviewed. PABC was defined as breast cancer diagnosedinawomanwho waspregnantorbreastfeeding. Numbers and frequencies are reported for demographic and diagnostic delay variables and Wilcoxon rank sum and Fisher’s exact tests are used to compare characteristics of women with PABC to women with non-PABC at the alpha = 0.05 significance level. Treatment and outcomes are described for womenwithPABConly. Results: Of the 117 women with breast cancer, 12 (10.3%) had PABC based on medical record review. The only significant demographic differences were that women with PABC were younger (p =0.006) and more likely to be married (p = 0.035) compared to women with non-PABC. There were no significant differences in diagnostic delays or stageatdiagnosis between womenwithPABCand womenwithnon-PABCwomen.Elevenofthe women with PABC received treatment, three had documented treatment delays or modifications due to their pregnancy or breastfeeding, and four stopped breastfeeding to initiate treatment. At the end of the study period, six patients were alive, three were deceased and three patients were lost to follow-up. Conclusions: PABC was relatively common in our cohort but may have been underreported. Although patients with PABC did not experience greater diagnostic delays, most had treatment modifications, emphasizing the potential value of PABC-specific treatment protocols in SSA. Larger prospective studies of PABC are needed to better understand particular challenges faced by these patients and inform policies and practicestooptimizecarefor womenwithPABCinRwanda and similar settings. Keywords: Breast cancer, Pregnancy, Breastfeeding, Africa * Correspondence: jmvduse@gmail.com Jean Marie Vianney Dusengimana and Vedaste Hategekimana contributed equally to this work. Partners In Health/Inshuti Mu Buzima, P.O.Box 3432, Kigali, Rwanda Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Dusengimana et al. BMC Cancer (2018) 18:634 Page 2 of 8 Background at presentation [21]. In this study, we focus on women Breast cancer is the most common type of malignancy with PABC seeking care at BCCOE to better understand encountered during pregnancy [1–5]. Despite the grow- their current demographic and clinical characteristics, ing body of literature on pregnancy-associated breast diagnostic delays, how they are treated and their out- cancer (PABC), most often defined as breast cancer di- comes at the end of the study period to inform the agnosed during a pregnancy or within one to two years provision of care for patients with PABC in LMICs. after delivery with some definitions extending up to five years after delivery [6], the implications of PABC remain Methods poorly understood. Reports on prevalence, optimal man- Study design and setting agement, prognosis and survival of women with PABC This retrospective cohort study included women with versus women with non-PABC are conflicting and this breast cancer presenting for treatment at BCCOE. has had repercussions for how PABC is managed [7–12]. BCCOE was opened by the Rwanda Ministry of Health While at one time considered to be untreatable, recent in July 2012, with the support of the organizations Part- reports suggest that mothers and babies can do well fol- ners In Health, Dana-Farber Cancer Institute and Brig- lowing treatment during pregnancy if it is delivered in a ham and Women’s Hospital in Boston, USA [33]. timely manner [5, 7]. For operable disease, surgery is Patients are referred to BCCOE from across Rwanda considered the best treatment during pregnancy and and adjacent countries because of the availability and af- chemotherapy can be safely delivered in the second and fordability of pathology services, chemotherapy, surgery third trimesters [2, 8, 13–16]. and clinicians with training in cancer care. As there are Breast cancer is a growing public health concern in no full-time oncologists at BCCOE [37], cancer treat- low- and middle-income countries (LMICs), including in ment is facilitated through the implementation of sub-Saharan Africa (SSA), where both breast cancer inci- nationally-endorsed protocols based on international dence and mortality rates are rising [17–20]. Women standards [33–36]. Medical oncology care is provided by with breast cancer in SSA are more likely to die of their a team of general physicians, internists, pediatricians disease than women with breast cancer in high-income and nurses who follow the protocols and consult regu- regions in part because of delays in diagnosis leading to larly with international oncology specialists. Surgical advanced stages at presentation [21]. It is likely that the care is provided by general surgeons. As of July 2017, relative burden of PABC is higher in SSA than in North there was no radiation available in Rwanda. America and Europe because of the region’s higher fer- The existing breast cancer treatment protocols provide tility rates, younger population, and younger median age simplified algorithms for early stage, locally advanced at breast cancer diagnosis [22–25]. Women with PABC and metastatic disease. Since there are no specific guide- in SSA may face even longer delays to diagnosis and lines for women with PABC, treatment follows general treatment than their counterparts with non-PABC be- breast cancer protocols but can be modified based on cause of low levels of breast cancer awareness, the possi- clinician judgment, with expert consultation available bility of breast cancers being misidentified as normal from one of the partnering institutions. breast modifications that occur during pregnancy or lac- tation, providers delaying diagnostic workup until after Study population and data collection pregnancy or lactation is complete, the competing prior- The study population included female patients with ities of pregnant women or new mothers, and the lack of pathologically confirmed breast cancer who presented at alternatives to breastfeeding which make it difficult for a BCCOE between July 1, 2012 and February 28, 2014. mother if she is advised to stop for treatment [9, 26–31]. Only women between 18 and 50 years old were included However, despite the likely higher burden of PABC in in the analysis to allow us to focus on the experiences of SSA and the additional challenges women with PABC women of reproductive age and to increase the compar- face, little is known about prevalence, presentation and ability of women with PABC to women with non-PABC. management of PABC in the region [7, 8, 17, 24, 27]. Clinical data were extracted from an existing database. Rwanda is a small low-income country in East Africa This database included information from the medical re- with one of the highest population densities in the con- cords, such as cancer diagnosis and staging information, tinent. Breast cancer is the most common cancer diag- for all BCCOE patients who presented with a breast nosed in women at national referral hospitals and is the complaint during the study period. Staging of breast most common adult cancer encountered at the Butaro cancer patients at BCCOE was conducted according to Cancer Center of Excellence (BCCOE), a specialized the American Joint Committee on Cancer (AJCC) sta- cancer facility in the rural northwest [32–36]. Women at ging system [38], but treatment protocols were based on BCCOE often present with late stage disease and both whether patients fell into simplified categories of “early,” patient and system delays are associated with later stage “locally advanced,” or “metastatic” disease. In Table 1 Dusengimana et al. BMC Cancer (2018) 18:634 Page 3 of 8 Table 1 Subset of breast cancer patients with demographic, Table 1 Subset of breast cancer patients with demographic, diagnostic delays and staging data available; comparing women diagnostic delays and staging data available; comparing women with and without pregnancy associated breast cancer with and without pregnancy associated breast cancer (Continued) PABC Non-PABC p-value (N = 12) (N = 62) PABC Non-PABC p-value n (%) unless n (%) unless (N = 12) (N = 62) otherwise noted otherwise noted n (%) unless n (%) unless otherwise noted otherwise noted Demographics Unknown 1 (8.3) 6 (9.7) Age - Median (IQR) 37 (35,38) 43 (37, 47) 0.006 ECOG status Marital Status 0 10 (83.3) 41 (66.1) 0.499 Married/In 11 (91.7) 32 (51.6) 0.035 relationship 1 0 10 (16.1) Widow/Divorced 1 (8.3) 15 (24.2) 2 1 (8.3) 5 (8.1) Single 0 (0) 15 (24.2) 30 0 Education level Unknown 1 (8.3) 6 (9.7) None 2 (16.7) 12 (19.4) 0.911 PABC Pregnancy associated breast cancer Fisher’s exact test Primary 6 (50.0) 28 (45.2) Wilcoxon rank sum test General national medical insurance Secondary 3 (25.0) 11 (17.7) Rwanda Assurance Maladie – Employed national medical insurance University 1 (8.3) 11 (17.7) Health insurance status patients are grouped as having non-metastatic and meta- d b Mutuelle 9 (75.0) 46 (74.2) 0.879 static disease to increase power to explore differences in these two groups. A focused chart review of all patients RAMA 3 (25.0) 11 (17.7) meeting the study criteria was conducted to confirm Other private 0 3 (4.8) who had PABC. For our purposes, PABC was defined as Not applicable 0 2 (3.2) breast cancer diagnosed during pregnancy or when a Distance from home to health facility woman was breastfeeding. Breastfeeding was included Less than 1 h 7 (58.3) 35 (56.5) > 0.999 within our definition because the date of last delivery 1–2 h 3 (25.0) 16 (25.8) was not always documented in cancer patients’ medical records, making it impossible to calculate the precise More than 2 h 2 (16.7) 11 (17.7) length of time since delivery. Since the median duration Care seeking patterns and clinical presentation of breastfeeding in Rwanda is approximately 2.5 years Healthcare provider visited first (N = 10 for PABC, N =44 [36, 39], we deemed it reasonable to use breastfeeding as for non-PABC) a proxy for being within two to three years postpartum Health center 8 (80.0) 33 (75.0) 0.883 [6]. Any patients that met our definition were classified Hospital outside 0 2 (4.6) as having PABC, and data on their treatment and out- Rwanda comes was collected. Private hospital 2 (20.0) 6 (13.6) For the focused chart review, a data collection form Referral hospital 0 3 (6.8) designed by the research team included a series of Time between symptom onset and first presentation to a health yes/no questions to prompt the data collector to rec- facility (days) ord treatment information and to note any indica- Median (IQR) 109 (6.5, 325.5) 139.5 (28, 402) 0.367 tions of treatment delays or modifications due to Time between first presentation at health facility and diagnosis (days) PABC (Additional file 1: Appendix 1). The form also asked “Did the patient stop breastfeeding while re- Median (IQR) 212 (58.5, 362) 145 (59, 328) 0.907 ceiving breast cancer treatment? If yes, why?” Treat- Stage at diagnosis ment and outcomes data were collected at the time Non-metastatic 9 (75.0) 47 (75.8) > 0.999 of chart review - March 31, 2015. Outcomes were Metastatic 3 (25.0) 14 (22.6) based on treatment notes documenting patients’ sta- Unknown 0 1 (1.6) tus as of this date and categorized as alive and in Hormone receptor status care, lost to follow-up and deceased. A patient was considered lost to follow-up if her last recorded Positive 8 (66.7) 33 (53.2) 0.798 clinic visit was six months or more prior to March Negative 3 (25.0) 23 (37.1) 31, 2015 and there was no indication in her chart that she had finished treatment. Dusengimana et al. BMC Cancer (2018) 18:634 Page 4 of 8 For detailed information on demographics, diagnostic for women with PABC and 145 days (IQR: 59, 328) for delays and staging, we consulted a second database de- women with non-PABC (p = 0.907). Three (25.0%) of the veloped from surveys conducted as part of a separate women with PABC had metastatic disease and 8 (66.7%) study of diagnostic delays [21]. All women ages 21 or had tumors that were hormone receptor status positive. older with a breast complaint who were available at the We did not identify any significant differences between time of survey administration were included in this women with PABC and women with non-PABC in terms study. This database was linked to the clinical database of diagnostic delays or stage at presentation (Table 1). using unique patient identification numbers. For the current analysis, we used survey responses from women PABC treatment and outcomes with pathologically confirmed cancer. Patient delay was Among the 12 women identified as having PABC, 3 were calculated as the time between breast symptom onset pregnant at the time of diagnosis, and 9 were breastfeed- and first presentation to a doctor or nurse and system ing. Eleven patients (91.7%) underwent at least one treat- delay was calculated as the time between the first visit to ment modality and one woman (who was pregnant at a doctor or nurse and the date of the first pathology re- diagnosis) was lost to follow-up before starting treat- port confirming breast cancer. ment (Table 2). As first course of treatment, four (36.4%) women received endocrine therapy, five (45.5%) Analysis women received neoadjuvant IV chemotherapy, one We described demographic characteristics, diagnostic (9.1%) woman received endocrine therapy and neoadju- delays and stage at diagnosis for the women ages 21 and vant IV chemotherapy and one (9.1%) woman underwent older contained in the secondary dataset. We used Wil- a mastectomy. Of the 11 patients who received any coxon rank sum and Fisher’s exact tests to compare treatment, 3 women (27.2%) had indication of a delay or characteristics of women with PABC to women with modification in treatment due to pregnancy or breast- non-PABC at the alpha = 0.05 significance level. We de- feeding. Of these, two women delayed the start of treat- scribed treatment and outcomes among women with ment because of pregnancy and one woman delayed the PABC only, indicating any modifications made to treat- start of treatment because she was breastfeeding. Four ment due to pregnancy or breastfeeding. Statistical ana- women (36.3%) were documented to have stopped lyses were performed using Stata v.13 (College Station, breastfeeding in order to start treatment. One other TX: Stata Corp LP). woman stopped breastfeeding during her treatment, but the reasons were not specified. For the other 3 women Results who were breastfeeding at diagnosis and received treat- Characteristics of women with PABC ment, cessation or continuation of breastfeeding was not There were 252 women that presented to BCCOE with documented in the record. confirmed breast cancer within the study window, and As of March 31, 2015, six (50.0%) of the women with 117 (46.4%) were between 18 and 50 years old (Fig. 1). PABC were alive, of which four (66.7%) were still receiv- Of these 117 patients, 12 (10.3%) met our criteria for ing chemotherapy or endocrine therapy, one (16.7%) was having PABC. All 12 (100%) patients with PABC and 62 on palliative care and one (16.7%) completed treatment. (59.0%) patients with non-PABC were contained in the Three (25.0%) patients were deceased, all of which had secondary dataset (Table 1). The median age for women metastatic disease at diagnosis and three (25.0%) patients with PABC was 37 years (interquartile range [IQR]: 35, were lost to follow-up. 38). Most women with PABC were married or in a rela- tionship (n = 11, 91.7%) and had a primary school level Discussion of education or less (n = 6, 50.0%). Nine women (75.0%) We identified 12 women with PABC who enrolled at had Mutuelle de Sante, the national health insurance, BCCOE between July 1, 2012 and February 28, 2014, and 7 (58.3%) lived within one hour of the nearest health representing 10.3% of women ages 18–50 years diag- center. The only significant demographic differences be- nosed with breast cancer. The proportion of women tween women with PABC and women with non-PABC with PABC is somewhat higher than rates documented were that women with PABC were significantly younger in Europe (for example, 7.1% of women under 45 with (p = 0.006) and more likely to be married (p = 0.035). breast cancer in Sweden [40]) but lower than rates docu- The median time from first onset of symptoms to first mented in other sub-Saharan sites (for example, 21.2% visit to a health facility (patient delay) was 109 days of premenopausal women with breast cancer at a Niger- (IQR: 6.5, 325.5) for women with PABC and 139.5 days ian hospital were diagnosed during or within two years (IQR: 28, 402) for women with non-PABC (p = 0.367). after pregnancy [6]). However, we believe that our study The median time from first visit to a health facility and underestimates the true rate of PABC prevalence at diagnosis (system delay) was 212 days (IQR: 58.5, 362) BCCOE since pregnancy and breastfeeding status in our Dusengimana et al. BMC Cancer (2018) 18:634 Page 5 of 8 Fig. 1 A flowchart describing sample included in PABC analysis cancer patients is not routinely documented. Women Table 2 Treatment of women with PABC (N = 12) with PABC did experience considerable delays prior to n% diagnosis, with a median time of 109 days from symp- Type of PABC (n = 12) tom onset to first visit at a health facility, and 212 days from first visit to a health facility to initial diagnosis. Pregnant at diagnosis 3 25.0 Based on our review of the literature, we hypothesized Breastfeeding at diagnosis 9 75.0 that women with PABC would experience longer diag- Patient received any treatment nostic delays and as a result have more advanced disease Yes 11 91.7 at diagnosis [5, 25, 30, 41]. We did not find this in our No (lost to follow-up prior to treatment) 1 8.3 population, which may be attributable to our small sam- Limited to individuals who received treatment (n = 11) ple size which limited our power to detect differences between women with and without PABC. However, since Type of treatment initially received long diagnostic delays have been observed in patients Endocrine therapy 4 36.4 with PABC in other settings [3, 9, 27, 28, 42–44] and be- Neoadjuvant IV chemotherapy 5 45.5 cause most patients with breast cancer at BCCOE ex- Endocrine therapy and neoadjuvant IV chemotherapy 1 9.1 perience long diagnostic delays regardless of pregnancy Mastectomy 1 9.1 and lactation status, [21] educational interventions for Any indication of treatment delays or modifications due to pregnancy patients and providers in Rwanda are important for pro- or breastfeeding? moting earlier breast cancer detection, and should in- No 9 81.8 clude the fact that breast cancer can occur during pregnancy and lactation. Leveraging existing health care Yes 3 27.2 interactions that occur for pregnant and breastfeeding Delayed start of treatment because of pregnancy 2 18.2 women could make pregnancy an opportunity for early Delayed start of treatment because of breastfeeding 1 9.1 diagnosis instead of a risk factor for delay and late stage Stopped breastfeeding to start treatment 4 36.4 presentation [20, 21, 45, 46]. Dusengimana et al. BMC Cancer (2018) 18:634 Page 6 of 8 In our study, three women with PABC had docu- settings with more barriers to cancer care, the challenges mented modifications made to their treatment plan be- facing women with PABC could be even more acute. cause of pregnancy or breastfeeding. Treatment Further research both in our settings and in others will considerations in PABC are complex and often require help identify the range of issues that need to be ad- an individualized approach, though key principles (such dressed across the region. as avoiding chemotherapy in the first trimester) should be disseminated in breast cancer treatment protocols Conclusion [31]. Effective and safe treatment for patients with PABC This study highlights that PABC is an important clinical also requires multidisciplinary collaboration [31]. At challenge among patients diagnosed with breast cancer in BCCOE, this often entailed email or phone consultations Rwanda. As the burden of breast cancer rises in LMICs, with surgical or medical oncologists and obstetricians further research is needed to illuminate the scope of this based in Rwanda’s capital city or in the United States. issue, understand its medical, cultural and psychosocial im- Further investigations into how prognosis is associated plications for patients, and help develop context-specific with delayed or modified treatment because of preg- management protocols in sub-Saharan Africa. nancy or breastfeeding in our setting are needed. In addition to the psychosocial challenges faced by any Additional file young woman with a cancer diagnosis, patients in SSA and other low-income regions face particular challenges Additional files 1: Appendix 1, Portion of data collection form designed by researchers to record delays, deviations and modifications in treatment when diagnosed with PABC. Four patients stopped for PABC cohort. This form was used during the medical record abstraction breastfeeding because of cancer treatment. While this process to identify treatment delays and modifications due to pregnancy or can be important for the health of the baby, stopping breastfeeding for patients with PABC. (DOCX 17 kb) breastfeeding is a particular challenge in Rwanda be- cause of the expense of appropriate alternatives (e.g. in- Abbreviations AJCC: American Joint Committee on Cancer; BCCOE: Butaro Cancer Center of fant formula) with resulting consequences for babies’ Excellence; IQR: Interquartile range; LMICs: Low- and middle-income coun- health. These challenges must be addressed by oncology tries; PABC: Pregnancy-associated breast cancer; RAMA: Rwanda Assurance programs in Rwanda and similar settings, through finan- Maladie (Employed National Medical Insurance); SSA: Sub-Saharan Africa; USA: United States of America cial support for procurement of formula. Additionally, for women who are pregnant at diagnosis, therapeutic Acknowledgements abortion may be the preferred choice, and in some set- We acknowledge the Partners In Health/Inshuti Mu Buzima Intermediate Operational Research Training Program under which this study was tings therapeutic abortion may be legally, logistically, developed and funded.. We acknowledge the leadership and clinicians for and culturally difficult to pursue. Research focused spe- facilitating data collection. We also acknowledge Michel Abewe, who cifically on managing PABC in SSA and similar reviewed the patient charts for PABC indications. low-resource settings that have limited diagnostic and Funding treatment capacity will be vital to understand the com- The Intermediate Operational Research Training Program, including some of plex needs of these patients and address the increasing the data collection costs, was supported by Partners In Health/Inshuti Mu Buzima-Rwanda and the Breast Cancer Research Foundation, grant BCRF-17- burden of the disease [40]. 147. The sub-cohort data collection, analysis and manuscript writing for the This study has several limitations. First, the small delays study was funded by Dana-Farber Cancer Institute and the Global number of women with PABC limited our ability to de- Women’s Health Fellowship at Brigham and Women’s Hospital. tect differences in delays and stage at presentation. Sec- Availability of data and materials ond, PABC status is likely underreported because The data that supports the findings of this study is available from the pregnancy or breastfeeding status is not systematically Rwanda Ministry of Health and Partners In Health/Inshuti Mu Buzima, but restrictions apply to the availability of the data, which was used with documented in our medical record system. Our findings permission for the current study and therefore not publicly available. Data is underscore the importance of comprehensive and sys- however available from the corresponding author upon reasonable request tematic data collection and management systems in and with permission of Partners In Health/Inshuti Mu Buzima and the Ministry of Health. LMICs in order to accurately assess patient diagnoses, treatment and outcomes. Third, due to a lack of suffi- Authors’ contributions cient follow-up time to have key outcomes data, we did Conception/Design: LP, VH, JMVD, TM, LS and ST; Data collection: VH, JMVD, TM, IN, TF, and LP; Data analysis: VH, JMVD, LP, RB and BHG; Results not compare outcomes of patients with PABC patients interpretation: All co-authors Manuscript writing: JMVD, VH, LP, RB, BHG to outcomes of patients with non-PABC. We plan to and NG; Final approval of manuscript: All co-authors. study this in the future, when we have more individuals and longer follow-up. Our findings are not necessarily Ethics approval and consent to participate There were two data sources included in this study; both received ethics generalizable to other settings in sub-Saharan Africa, approval from the Rwanda National Ethics Committee (Rwanda) and Partners particularly facilities that are not rural, or hospitals that Institutional Review Board (Boston, USA). The first data source was a chart are not cancer referral facilities. We suspect that in extraction of routinely collected clinical data. Because there was no direct Dusengimana et al. BMC Cancer (2018) 18:634 Page 7 of 8 contact with patients and minimal risks, both ethics committees waived the 17. DeSantis CE, Bray F, Ferlay J, Lortet-Tieulent J, Anderson BO, Jemal A. need for consent for that aspect of the project. The second data source was International variation in female breast cancer incidence and mortality rates. collected as part of an ongoing study [21]; women were consented prior to Cancer Epidemiol Biomark Prev. 2015;24(10):1495–506. https://doi.org/10. data collection as part of the study. The consent obtained was written. 1158/1055-9965. 18. Kantelhardt EJ, Cubasch H, Hanson C. Taking on breast cancer in East Africa: Competing interests global challenges in breast cancer. Curr Opin Obstet Gynecol. 2015;27(1): The authors declare that they have no competing interests 108–14. https://doi.org/10.1097/GCO.0000000000000139. 19. Brinton LA, Figueroa JD, Awuah B, Yarney J, Wiafe S, Woon SN, et al. Breast cancer in sub-Saharan Africa: opportunities for prevention. Breast Cancer Publisher’sNote Res and Treat. 2014;144(3):467–78. https://doi.org/10.1007/s10549-014-2868-z. Springer Nature remains neutral with regard to jurisdictional claims in 20. Anderson BO, Ilbawi AM, El Saghir NS. Breast cancer in low and middle published maps and institutional affiliations. income countries (LMICs): a shifting tide in global health. Breast J. 2015; 21(1):111–8. https://doi.org/10.1111/tbj.12357. Author details 21. Pace LE, Mpunga T, Hategekimana V, Dusengimana JM, Habineza H, 1 2 Partners In Health/Inshuti Mu Buzima, P.O.Box 3432, Kigali, Rwanda. Ministry Bigirimana JB, et al. Delays in breast cancer presentation and diagnosis at of Health, Butaro District Hospital, Butaro, Rwanda. Harvard Medical School, two rural cancer referral centers in Rwanda. Oncologist. 2015;207(7):780–8. Boston, MA, USA. Brigham and Women’s Hospital, Boston, MA, USA. https://doi.org/10.1634/theoncologist.2014-0493. 5 6 University of Pennsylvania, Philadelphia, PA, USA. Dana Farber Cancer 22. Donkor A, Lathlean J, Wiafe S, Vanderpuye V, Fenlon D, Yarney J, et al. Institute, Boston, MA, USA. Factors contributing to late presentation of breast cancer in Africa: a systematic literature review. Arch Med. 2015;8:2. Received: 15 September 2017 Accepted: 21 May 2018 23. Gewefel H, Sarhia B. Breast cancer in adolescent and young adult women. Clinical Breast Cancer. 2014;14(6):390–5. https://doi.org/10.1016/j.clbc.2014.06.002. 24. Mair LP. General characteristics of African family and marriage in African References marriage and social change. New York: Routledge; 2013. p. 1–8. 1. American Cancer Society. Treating Breast Cancer During Pregnancy. Last 25. Miranda JJ, Kinra S, Casas JP, Davey Smith G, Ebrahim S. Non-communicable revised 12/01/2014. http://www.cancer.org/cancer/breastcancer/ diseases in low- and middle-income countries: context, determinants and moreinformation/pregnancy-and-breast-cancer. Accessed 20 July 2016. health policy. Trop Med Int Health. 2008;13(10):1225–34. https://doi.org/10. 2. Van Calsteren K, Heyns L, De Smet F, Van Eycken L, Gziri MM, Van Gemert 1111/j.1365-3156. W, et al. Cancer during pregnancy: An analysis of 215 patients emphasizing 26. Akarolo-Anthony SN, Ogundiran TO, Adebamowo CA. Emerging breast the obstetrical and the neonatal outcomes. J Clin Oncol. 2009;28(4):683–9. cancer epidemic: evidence from Africa. Breast Cancer Res. 2010;12(Suppl 4): https://doi.org/10.1200/JCO.2009.23.2801. S8. https://doi.org/10.1186/bcr2737. 3. Litton JK, Theriault RP, Gonzalez-Anjulo AM. Breast cancer diagnosis during 27. Chiedozi LC, Iweze FI, Aboh IF, Ajabor LN. Breast cancer in pregnancy and pregnancy. Women’s Health (Lond Engl). 2009;5(3):243–9. https://doi.org/10. lactation. Trop Geogr Med. 1988;40(1):26–30. 2217/whe.09.2. 28. Al-Amri AM. Clinical presentation and causes of the delayed diagnosis of 4. Loibl S, von Minckwitz G, Gwyn K, Ellis P, Blohmer JU, Schlegelberger B, et breast cancer in patients with pregnancy associated breast cancer. J Family al. Breast carcinoma during pregnancy: international recommendations from Community Med. 2015;22(2):96–100. https://doi.org/10.4103/2230-8229. an expert meeting. Cancer. 2006;106(2):237–46. 5. Anderson JM. Mammary cancers and pregnancy. Br Med J. 1979;1(6171): 29. Sule EA, Ewemade F. Management of pregnancy associated breast cancer 1124–7. with chemotherapy in a developing country. Int J Surg Case Rep. 2015;17: 6. Hou N, Ogundiran T, Ojengbede O, Morhason-Bello I, Zheng Y, Fackenthal J, 117–20. https://doi.org/10.1016/j.ijscr.2015.10.008. et al. Risk factors for pregnancy-associated breast cancer: a report from the 30. Ibrahim NA, Oludara MA. Socio-demographic factors and reasons associated Nigerian breast Cancer study. Ann Epidemiol. 2013;23(9):551–7. with delay in breast cancer presentation: a study in Nigerian women. Breast. 7. Amant F, von Minckwitz G, Han SN, Bontenbal M, Ring AE, Giermek J, et al. 2012;21(3):416–8. https://doi.org/10.1016/j.breast.2012.02.006. Prognosis of women with primary breast cancer diagnosed during 31. Shachar SS, Gallagher K, McGuire K, Zagar TM, Faso A, Muss HB, et al. pregnancy: results from an international collaborative study. Journal Clin Multidisciplinary Management of Breast Cancer during Pregnancy. Oncol. 2013;31(20):2532–9. https://doi.org/10.1200/JCO.2012.45.6335. Oncologist. 2017;22:324–34. 8. Theriault RL, Litton JK. Pregnancy during or after breast cancer diagnosis: 32. Murthy SS, Tapela N, Muhimpundu MA, Balinda JP, Musabyemariya F, Kirby what do we know and what do we need to know? J Clin Onc. 2013;31: K. A national framework for breast cancer control: a report on Rwanda’s 2521–2. https://doi.org/10.1200/JCO.2013.49.7347. inaugural symposium on the management of breast cancer. J Cancer Policy. 9. Woo JC, Yu T, Hurd TC. Breast cancer in pregnancy: a literature review. Arch 2015;6:3–7. Surg. 2003;138:91–8. discussion 99 33. Tapela NM, Mpunga T, Hedt-Gauthier B, Moore M, Mpanumusingo E, Xu MJ, 10. Bonnier P, Romain S, Dilhuydy JM, Bonichon F, Julien JP, Charpin C, et al. et al. Pursuing equity in cancer care: implementation, challenges and Influence of pregnancy on the outcome of breast cancer: a case-control preliminary findings of a public cancer referral center in rural Rwanda. BMC study. Int J Cancer. 1997;72:720–7. Cancer. 2016;16:237. https://doi.org/10.1186/s12885-016-2256-77. 11. Hassan I, Muhammed I, Attah MM, Maboqunje O. Breast cancer during 34. Shulman LN, Mpunga T, Tapela N, Wagner CM, Fadelu T, Binagwago A. pregnancy and lactation in Zaria, Nigeria. East Afr Med J. 1995;72(5):280–2. Bringing cancer care to the poor: experiences from Rwanda. Nat Rev 12. Polivka J Jr, Altun I, Golubnitschaja O, et al. Pregnancy associated breast Cancer. 2014;14(12):815–21. https://doi.org/10.1038/nrc3848. Cancer: the risky status quo and new concepts of predictive medicine. 35. Mubiligi JM, Hedt-Gauthier B, Mpunga T, Tapela N, Okao P, Harries AD, et al. EPMA J. 2018;9(1):1–13. https://doi.org/10.1007/s13167-018-0129-7. Caring for patients with surgically resectable cancers: experience from a 13. Amant F, Vandenbroucke T, Verheecke M, Fumagalli M, Halaska MJ, Boere I, specialized center in rural Rwanda. Public Health Action. 2014;4(2):128–32. et al. Pediatric outcome after maternal cancer diagnosed during pregnancy. https://doi.org/10.5588/pha.14.0002. N Engl J Med. 2015;373:1824–34. https://doi.org/10.1056/ NEJMoa1508913. 36. O’Neil DS, Keating NL, Dusengimana JMV, Hategekimana V, Umwizera A, 14. Meisel JL, Economy KE, Calvillo KZ, Schapira L, Tung NM, Gelber S, et al. Mpunga T, et al. Quality of breast cancer treatment at a rural cancer center Contemporary multidisciplinary treatment of pregnancy-associated breast in Rwanda. J Global Onc Published online before print May 12, 2017. doi: cancer. Springerplus. 2013;2(1):297. https://doi.org/10.1186/2193-1801-2-297. https://doi.org/10.1200/JGO.2016.008672. 15. Assi HA, Khoury KE, Dbouk H, Khalil LE, Mouhieddine TH, Epidemiology ESNS. Prognosis of breast cancer in young women. J Thorac Dis. 2013; 37. Rubagumya F, Greenbergy L, Manirakiza A, DeBoer R, Park PH, Mpunga T, 5(Suppl 1):S2–8. https://doi.org/10.3978/j.issn.2072-1439.2013.05.24. Shulman LN. Increasing global access to cancer care: models of care with 16. Murphy CG, Mallam D, Stein S, Patil S, Howard J, Sklarin N, et al. Current or non-oncologists as primary providers. Lancet Oncol. 2017;18:1000–2. th recent pregnancy is associated with adverse pathologic features but not 38. Edge SB, Compton CC. The American joint committee on Cancer: the 7 impaired survival in early breast cancer. Cancer. 2012;118(13):3254–9. edition of the AJCC Cancer staging manual and the future of TNM. Ann https://doi.org/10.1002/cncr.26654. Surg Oncol. 2010;17(6):1471–4. Dusengimana et al. BMC Cancer (2018) 18:634 Page 8 of 8 39. National Institute of Statistics of Rwanda, Ministry of Health Rwanda, and ICF International. Rwanda Demographic and Health Survey 2010. Calverton: National Institute of Statistics of Rwanda, Ministry of Health Rwanda, and ICF International; 2012. 40. Andersson TM, Johansson AL, Hsieh CC, Cnattingius S, Lambe M. Increasing incidence of pregnancy-associated breast cancer in Sweden. Obstet Gynecol. 2009;114(3):568–72. https://doi.org/10.1097/AOG.0b013e3181b19154. 41. Yang YL, Chan KA, Hsieh FJ, Chang LY, Wang MY. Pregnancy-associated breast cancer in Taiwanese women: potential treatment delay and impact on survival. PLoS One. 2014;9(11):e111934. https://doi.org/10.1371/journal. pone.0111934. 42. Anderson BO, Petrek JA, Byrd DR, Senie RT, Borgen PI. Pregnancy influences breast cancer stage at diagnosis in women 30 years of age and younger. Ann Surg Oncol. 1996;3:204–11. 43. Ishida T, Yokoe T, Kasumi F, Sakamoto G, Makita M, Tominaga T, Simozuma K, Enomoto K, Fujiwara K, Nanasawa T, et al. Clinicopathologic characteristics and prognosis of breast cancer patients associated with pregnancy and lactation: analysis of case–control study in Japan. Jpn J Cancer Res. 1992;83: 1143–9. https://doi.org/10.1111/j.1349-7006.1992.tb02737.x. 44. Basaran D, Turgal M, Beksac K, Ozyuncu O, Aran O, Beksac MS. Pregnancy- associated breast cancer: Clinicopathological characteristics of 20 cases with focus on identifiable causes of diagnostic delay. Breast Care. 2014;9:355–9. https://doi.org/10.1159/000366436. 45. Galukande M, Kiguli-Malwadde E. Rethinking breast cancer screening strategies in resource-limited settings. Afr Health Sci. 2010;10(1):89–92. 46. Adebamowo CA, Akarolo-Anthony S. Cancer in Africa: opportunities for collaborative research and training. Afr J Med Med Sci. 2009;38(Suppl 2):5–13. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png BMC Cancer Springer Journals

Pregnancy-associated breast cancer in rural Rwanda: the experience of the Butaro Cancer Center of Excellence

Free
8 pages

Loading next page...
 
/lp/springer_journal/pregnancy-associated-breast-cancer-in-rural-rwanda-the-experience-of-rohBciLw4o
Publisher
Springer Journals
Copyright
Copyright © 2018 by The Author(s).
Subject
Biomedicine; Cancer Research; Oncology; Surgical Oncology; Health Promotion and Disease Prevention; Biomedicine, general; Medicine/Public Health, general
eISSN
1471-2407
D.O.I.
10.1186/s12885-018-4535-y
Publisher site
See Article on Publisher Site

Abstract

Background: Breast cancer is the most common malignancy encountered during pregnancy. However, the burden of pregnancy-associated breast cancer (PABC) and subsequent care is understudied in sub-Saharan Africa (SSA). Here, we describe the characteristics, diagnostic delays and treatment of women with PABC seeking care at a rural cancer referral facility in Rwanda. Methods: Data from female patients aged 18–50 years with pathologically confirmed breast cancer who presented for treatment between July 1, 2012 and February 28, 2014 were retrospectively reviewed. PABC was defined as breast cancer diagnosedinawomanwho waspregnantorbreastfeeding. Numbers and frequencies are reported for demographic and diagnostic delay variables and Wilcoxon rank sum and Fisher’s exact tests are used to compare characteristics of women with PABC to women with non-PABC at the alpha = 0.05 significance level. Treatment and outcomes are described for womenwithPABConly. Results: Of the 117 women with breast cancer, 12 (10.3%) had PABC based on medical record review. The only significant demographic differences were that women with PABC were younger (p =0.006) and more likely to be married (p = 0.035) compared to women with non-PABC. There were no significant differences in diagnostic delays or stageatdiagnosis between womenwithPABCand womenwithnon-PABCwomen.Elevenofthe women with PABC received treatment, three had documented treatment delays or modifications due to their pregnancy or breastfeeding, and four stopped breastfeeding to initiate treatment. At the end of the study period, six patients were alive, three were deceased and three patients were lost to follow-up. Conclusions: PABC was relatively common in our cohort but may have been underreported. Although patients with PABC did not experience greater diagnostic delays, most had treatment modifications, emphasizing the potential value of PABC-specific treatment protocols in SSA. Larger prospective studies of PABC are needed to better understand particular challenges faced by these patients and inform policies and practicestooptimizecarefor womenwithPABCinRwanda and similar settings. Keywords: Breast cancer, Pregnancy, Breastfeeding, Africa * Correspondence: jmvduse@gmail.com Jean Marie Vianney Dusengimana and Vedaste Hategekimana contributed equally to this work. Partners In Health/Inshuti Mu Buzima, P.O.Box 3432, Kigali, Rwanda Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Dusengimana et al. BMC Cancer (2018) 18:634 Page 2 of 8 Background at presentation [21]. In this study, we focus on women Breast cancer is the most common type of malignancy with PABC seeking care at BCCOE to better understand encountered during pregnancy [1–5]. Despite the grow- their current demographic and clinical characteristics, ing body of literature on pregnancy-associated breast diagnostic delays, how they are treated and their out- cancer (PABC), most often defined as breast cancer di- comes at the end of the study period to inform the agnosed during a pregnancy or within one to two years provision of care for patients with PABC in LMICs. after delivery with some definitions extending up to five years after delivery [6], the implications of PABC remain Methods poorly understood. Reports on prevalence, optimal man- Study design and setting agement, prognosis and survival of women with PABC This retrospective cohort study included women with versus women with non-PABC are conflicting and this breast cancer presenting for treatment at BCCOE. has had repercussions for how PABC is managed [7–12]. BCCOE was opened by the Rwanda Ministry of Health While at one time considered to be untreatable, recent in July 2012, with the support of the organizations Part- reports suggest that mothers and babies can do well fol- ners In Health, Dana-Farber Cancer Institute and Brig- lowing treatment during pregnancy if it is delivered in a ham and Women’s Hospital in Boston, USA [33]. timely manner [5, 7]. For operable disease, surgery is Patients are referred to BCCOE from across Rwanda considered the best treatment during pregnancy and and adjacent countries because of the availability and af- chemotherapy can be safely delivered in the second and fordability of pathology services, chemotherapy, surgery third trimesters [2, 8, 13–16]. and clinicians with training in cancer care. As there are Breast cancer is a growing public health concern in no full-time oncologists at BCCOE [37], cancer treat- low- and middle-income countries (LMICs), including in ment is facilitated through the implementation of sub-Saharan Africa (SSA), where both breast cancer inci- nationally-endorsed protocols based on international dence and mortality rates are rising [17–20]. Women standards [33–36]. Medical oncology care is provided by with breast cancer in SSA are more likely to die of their a team of general physicians, internists, pediatricians disease than women with breast cancer in high-income and nurses who follow the protocols and consult regu- regions in part because of delays in diagnosis leading to larly with international oncology specialists. Surgical advanced stages at presentation [21]. It is likely that the care is provided by general surgeons. As of July 2017, relative burden of PABC is higher in SSA than in North there was no radiation available in Rwanda. America and Europe because of the region’s higher fer- The existing breast cancer treatment protocols provide tility rates, younger population, and younger median age simplified algorithms for early stage, locally advanced at breast cancer diagnosis [22–25]. Women with PABC and metastatic disease. Since there are no specific guide- in SSA may face even longer delays to diagnosis and lines for women with PABC, treatment follows general treatment than their counterparts with non-PABC be- breast cancer protocols but can be modified based on cause of low levels of breast cancer awareness, the possi- clinician judgment, with expert consultation available bility of breast cancers being misidentified as normal from one of the partnering institutions. breast modifications that occur during pregnancy or lac- tation, providers delaying diagnostic workup until after Study population and data collection pregnancy or lactation is complete, the competing prior- The study population included female patients with ities of pregnant women or new mothers, and the lack of pathologically confirmed breast cancer who presented at alternatives to breastfeeding which make it difficult for a BCCOE between July 1, 2012 and February 28, 2014. mother if she is advised to stop for treatment [9, 26–31]. Only women between 18 and 50 years old were included However, despite the likely higher burden of PABC in in the analysis to allow us to focus on the experiences of SSA and the additional challenges women with PABC women of reproductive age and to increase the compar- face, little is known about prevalence, presentation and ability of women with PABC to women with non-PABC. management of PABC in the region [7, 8, 17, 24, 27]. Clinical data were extracted from an existing database. Rwanda is a small low-income country in East Africa This database included information from the medical re- with one of the highest population densities in the con- cords, such as cancer diagnosis and staging information, tinent. Breast cancer is the most common cancer diag- for all BCCOE patients who presented with a breast nosed in women at national referral hospitals and is the complaint during the study period. Staging of breast most common adult cancer encountered at the Butaro cancer patients at BCCOE was conducted according to Cancer Center of Excellence (BCCOE), a specialized the American Joint Committee on Cancer (AJCC) sta- cancer facility in the rural northwest [32–36]. Women at ging system [38], but treatment protocols were based on BCCOE often present with late stage disease and both whether patients fell into simplified categories of “early,” patient and system delays are associated with later stage “locally advanced,” or “metastatic” disease. In Table 1 Dusengimana et al. BMC Cancer (2018) 18:634 Page 3 of 8 Table 1 Subset of breast cancer patients with demographic, Table 1 Subset of breast cancer patients with demographic, diagnostic delays and staging data available; comparing women diagnostic delays and staging data available; comparing women with and without pregnancy associated breast cancer with and without pregnancy associated breast cancer (Continued) PABC Non-PABC p-value (N = 12) (N = 62) PABC Non-PABC p-value n (%) unless n (%) unless (N = 12) (N = 62) otherwise noted otherwise noted n (%) unless n (%) unless otherwise noted otherwise noted Demographics Unknown 1 (8.3) 6 (9.7) Age - Median (IQR) 37 (35,38) 43 (37, 47) 0.006 ECOG status Marital Status 0 10 (83.3) 41 (66.1) 0.499 Married/In 11 (91.7) 32 (51.6) 0.035 relationship 1 0 10 (16.1) Widow/Divorced 1 (8.3) 15 (24.2) 2 1 (8.3) 5 (8.1) Single 0 (0) 15 (24.2) 30 0 Education level Unknown 1 (8.3) 6 (9.7) None 2 (16.7) 12 (19.4) 0.911 PABC Pregnancy associated breast cancer Fisher’s exact test Primary 6 (50.0) 28 (45.2) Wilcoxon rank sum test General national medical insurance Secondary 3 (25.0) 11 (17.7) Rwanda Assurance Maladie – Employed national medical insurance University 1 (8.3) 11 (17.7) Health insurance status patients are grouped as having non-metastatic and meta- d b Mutuelle 9 (75.0) 46 (74.2) 0.879 static disease to increase power to explore differences in these two groups. A focused chart review of all patients RAMA 3 (25.0) 11 (17.7) meeting the study criteria was conducted to confirm Other private 0 3 (4.8) who had PABC. For our purposes, PABC was defined as Not applicable 0 2 (3.2) breast cancer diagnosed during pregnancy or when a Distance from home to health facility woman was breastfeeding. Breastfeeding was included Less than 1 h 7 (58.3) 35 (56.5) > 0.999 within our definition because the date of last delivery 1–2 h 3 (25.0) 16 (25.8) was not always documented in cancer patients’ medical records, making it impossible to calculate the precise More than 2 h 2 (16.7) 11 (17.7) length of time since delivery. Since the median duration Care seeking patterns and clinical presentation of breastfeeding in Rwanda is approximately 2.5 years Healthcare provider visited first (N = 10 for PABC, N =44 [36, 39], we deemed it reasonable to use breastfeeding as for non-PABC) a proxy for being within two to three years postpartum Health center 8 (80.0) 33 (75.0) 0.883 [6]. Any patients that met our definition were classified Hospital outside 0 2 (4.6) as having PABC, and data on their treatment and out- Rwanda comes was collected. Private hospital 2 (20.0) 6 (13.6) For the focused chart review, a data collection form Referral hospital 0 3 (6.8) designed by the research team included a series of Time between symptom onset and first presentation to a health yes/no questions to prompt the data collector to rec- facility (days) ord treatment information and to note any indica- Median (IQR) 109 (6.5, 325.5) 139.5 (28, 402) 0.367 tions of treatment delays or modifications due to Time between first presentation at health facility and diagnosis (days) PABC (Additional file 1: Appendix 1). The form also asked “Did the patient stop breastfeeding while re- Median (IQR) 212 (58.5, 362) 145 (59, 328) 0.907 ceiving breast cancer treatment? If yes, why?” Treat- Stage at diagnosis ment and outcomes data were collected at the time Non-metastatic 9 (75.0) 47 (75.8) > 0.999 of chart review - March 31, 2015. Outcomes were Metastatic 3 (25.0) 14 (22.6) based on treatment notes documenting patients’ sta- Unknown 0 1 (1.6) tus as of this date and categorized as alive and in Hormone receptor status care, lost to follow-up and deceased. A patient was considered lost to follow-up if her last recorded Positive 8 (66.7) 33 (53.2) 0.798 clinic visit was six months or more prior to March Negative 3 (25.0) 23 (37.1) 31, 2015 and there was no indication in her chart that she had finished treatment. Dusengimana et al. BMC Cancer (2018) 18:634 Page 4 of 8 For detailed information on demographics, diagnostic for women with PABC and 145 days (IQR: 59, 328) for delays and staging, we consulted a second database de- women with non-PABC (p = 0.907). Three (25.0%) of the veloped from surveys conducted as part of a separate women with PABC had metastatic disease and 8 (66.7%) study of diagnostic delays [21]. All women ages 21 or had tumors that were hormone receptor status positive. older with a breast complaint who were available at the We did not identify any significant differences between time of survey administration were included in this women with PABC and women with non-PABC in terms study. This database was linked to the clinical database of diagnostic delays or stage at presentation (Table 1). using unique patient identification numbers. For the current analysis, we used survey responses from women PABC treatment and outcomes with pathologically confirmed cancer. Patient delay was Among the 12 women identified as having PABC, 3 were calculated as the time between breast symptom onset pregnant at the time of diagnosis, and 9 were breastfeed- and first presentation to a doctor or nurse and system ing. Eleven patients (91.7%) underwent at least one treat- delay was calculated as the time between the first visit to ment modality and one woman (who was pregnant at a doctor or nurse and the date of the first pathology re- diagnosis) was lost to follow-up before starting treat- port confirming breast cancer. ment (Table 2). As first course of treatment, four (36.4%) women received endocrine therapy, five (45.5%) Analysis women received neoadjuvant IV chemotherapy, one We described demographic characteristics, diagnostic (9.1%) woman received endocrine therapy and neoadju- delays and stage at diagnosis for the women ages 21 and vant IV chemotherapy and one (9.1%) woman underwent older contained in the secondary dataset. We used Wil- a mastectomy. Of the 11 patients who received any coxon rank sum and Fisher’s exact tests to compare treatment, 3 women (27.2%) had indication of a delay or characteristics of women with PABC to women with modification in treatment due to pregnancy or breast- non-PABC at the alpha = 0.05 significance level. We de- feeding. Of these, two women delayed the start of treat- scribed treatment and outcomes among women with ment because of pregnancy and one woman delayed the PABC only, indicating any modifications made to treat- start of treatment because she was breastfeeding. Four ment due to pregnancy or breastfeeding. Statistical ana- women (36.3%) were documented to have stopped lyses were performed using Stata v.13 (College Station, breastfeeding in order to start treatment. One other TX: Stata Corp LP). woman stopped breastfeeding during her treatment, but the reasons were not specified. For the other 3 women Results who were breastfeeding at diagnosis and received treat- Characteristics of women with PABC ment, cessation or continuation of breastfeeding was not There were 252 women that presented to BCCOE with documented in the record. confirmed breast cancer within the study window, and As of March 31, 2015, six (50.0%) of the women with 117 (46.4%) were between 18 and 50 years old (Fig. 1). PABC were alive, of which four (66.7%) were still receiv- Of these 117 patients, 12 (10.3%) met our criteria for ing chemotherapy or endocrine therapy, one (16.7%) was having PABC. All 12 (100%) patients with PABC and 62 on palliative care and one (16.7%) completed treatment. (59.0%) patients with non-PABC were contained in the Three (25.0%) patients were deceased, all of which had secondary dataset (Table 1). The median age for women metastatic disease at diagnosis and three (25.0%) patients with PABC was 37 years (interquartile range [IQR]: 35, were lost to follow-up. 38). Most women with PABC were married or in a rela- tionship (n = 11, 91.7%) and had a primary school level Discussion of education or less (n = 6, 50.0%). Nine women (75.0%) We identified 12 women with PABC who enrolled at had Mutuelle de Sante, the national health insurance, BCCOE between July 1, 2012 and February 28, 2014, and 7 (58.3%) lived within one hour of the nearest health representing 10.3% of women ages 18–50 years diag- center. The only significant demographic differences be- nosed with breast cancer. The proportion of women tween women with PABC and women with non-PABC with PABC is somewhat higher than rates documented were that women with PABC were significantly younger in Europe (for example, 7.1% of women under 45 with (p = 0.006) and more likely to be married (p = 0.035). breast cancer in Sweden [40]) but lower than rates docu- The median time from first onset of symptoms to first mented in other sub-Saharan sites (for example, 21.2% visit to a health facility (patient delay) was 109 days of premenopausal women with breast cancer at a Niger- (IQR: 6.5, 325.5) for women with PABC and 139.5 days ian hospital were diagnosed during or within two years (IQR: 28, 402) for women with non-PABC (p = 0.367). after pregnancy [6]). However, we believe that our study The median time from first visit to a health facility and underestimates the true rate of PABC prevalence at diagnosis (system delay) was 212 days (IQR: 58.5, 362) BCCOE since pregnancy and breastfeeding status in our Dusengimana et al. BMC Cancer (2018) 18:634 Page 5 of 8 Fig. 1 A flowchart describing sample included in PABC analysis cancer patients is not routinely documented. Women Table 2 Treatment of women with PABC (N = 12) with PABC did experience considerable delays prior to n% diagnosis, with a median time of 109 days from symp- Type of PABC (n = 12) tom onset to first visit at a health facility, and 212 days from first visit to a health facility to initial diagnosis. Pregnant at diagnosis 3 25.0 Based on our review of the literature, we hypothesized Breastfeeding at diagnosis 9 75.0 that women with PABC would experience longer diag- Patient received any treatment nostic delays and as a result have more advanced disease Yes 11 91.7 at diagnosis [5, 25, 30, 41]. We did not find this in our No (lost to follow-up prior to treatment) 1 8.3 population, which may be attributable to our small sam- Limited to individuals who received treatment (n = 11) ple size which limited our power to detect differences between women with and without PABC. However, since Type of treatment initially received long diagnostic delays have been observed in patients Endocrine therapy 4 36.4 with PABC in other settings [3, 9, 27, 28, 42–44] and be- Neoadjuvant IV chemotherapy 5 45.5 cause most patients with breast cancer at BCCOE ex- Endocrine therapy and neoadjuvant IV chemotherapy 1 9.1 perience long diagnostic delays regardless of pregnancy Mastectomy 1 9.1 and lactation status, [21] educational interventions for Any indication of treatment delays or modifications due to pregnancy patients and providers in Rwanda are important for pro- or breastfeeding? moting earlier breast cancer detection, and should in- No 9 81.8 clude the fact that breast cancer can occur during pregnancy and lactation. Leveraging existing health care Yes 3 27.2 interactions that occur for pregnant and breastfeeding Delayed start of treatment because of pregnancy 2 18.2 women could make pregnancy an opportunity for early Delayed start of treatment because of breastfeeding 1 9.1 diagnosis instead of a risk factor for delay and late stage Stopped breastfeeding to start treatment 4 36.4 presentation [20, 21, 45, 46]. Dusengimana et al. BMC Cancer (2018) 18:634 Page 6 of 8 In our study, three women with PABC had docu- settings with more barriers to cancer care, the challenges mented modifications made to their treatment plan be- facing women with PABC could be even more acute. cause of pregnancy or breastfeeding. Treatment Further research both in our settings and in others will considerations in PABC are complex and often require help identify the range of issues that need to be ad- an individualized approach, though key principles (such dressed across the region. as avoiding chemotherapy in the first trimester) should be disseminated in breast cancer treatment protocols Conclusion [31]. Effective and safe treatment for patients with PABC This study highlights that PABC is an important clinical also requires multidisciplinary collaboration [31]. At challenge among patients diagnosed with breast cancer in BCCOE, this often entailed email or phone consultations Rwanda. As the burden of breast cancer rises in LMICs, with surgical or medical oncologists and obstetricians further research is needed to illuminate the scope of this based in Rwanda’s capital city or in the United States. issue, understand its medical, cultural and psychosocial im- Further investigations into how prognosis is associated plications for patients, and help develop context-specific with delayed or modified treatment because of preg- management protocols in sub-Saharan Africa. nancy or breastfeeding in our setting are needed. In addition to the psychosocial challenges faced by any Additional file young woman with a cancer diagnosis, patients in SSA and other low-income regions face particular challenges Additional files 1: Appendix 1, Portion of data collection form designed by researchers to record delays, deviations and modifications in treatment when diagnosed with PABC. Four patients stopped for PABC cohort. This form was used during the medical record abstraction breastfeeding because of cancer treatment. While this process to identify treatment delays and modifications due to pregnancy or can be important for the health of the baby, stopping breastfeeding for patients with PABC. (DOCX 17 kb) breastfeeding is a particular challenge in Rwanda be- cause of the expense of appropriate alternatives (e.g. in- Abbreviations AJCC: American Joint Committee on Cancer; BCCOE: Butaro Cancer Center of fant formula) with resulting consequences for babies’ Excellence; IQR: Interquartile range; LMICs: Low- and middle-income coun- health. These challenges must be addressed by oncology tries; PABC: Pregnancy-associated breast cancer; RAMA: Rwanda Assurance programs in Rwanda and similar settings, through finan- Maladie (Employed National Medical Insurance); SSA: Sub-Saharan Africa; USA: United States of America cial support for procurement of formula. Additionally, for women who are pregnant at diagnosis, therapeutic Acknowledgements abortion may be the preferred choice, and in some set- We acknowledge the Partners In Health/Inshuti Mu Buzima Intermediate Operational Research Training Program under which this study was tings therapeutic abortion may be legally, logistically, developed and funded.. We acknowledge the leadership and clinicians for and culturally difficult to pursue. Research focused spe- facilitating data collection. We also acknowledge Michel Abewe, who cifically on managing PABC in SSA and similar reviewed the patient charts for PABC indications. low-resource settings that have limited diagnostic and Funding treatment capacity will be vital to understand the com- The Intermediate Operational Research Training Program, including some of plex needs of these patients and address the increasing the data collection costs, was supported by Partners In Health/Inshuti Mu Buzima-Rwanda and the Breast Cancer Research Foundation, grant BCRF-17- burden of the disease [40]. 147. The sub-cohort data collection, analysis and manuscript writing for the This study has several limitations. First, the small delays study was funded by Dana-Farber Cancer Institute and the Global number of women with PABC limited our ability to de- Women’s Health Fellowship at Brigham and Women’s Hospital. tect differences in delays and stage at presentation. Sec- Availability of data and materials ond, PABC status is likely underreported because The data that supports the findings of this study is available from the pregnancy or breastfeeding status is not systematically Rwanda Ministry of Health and Partners In Health/Inshuti Mu Buzima, but restrictions apply to the availability of the data, which was used with documented in our medical record system. Our findings permission for the current study and therefore not publicly available. Data is underscore the importance of comprehensive and sys- however available from the corresponding author upon reasonable request tematic data collection and management systems in and with permission of Partners In Health/Inshuti Mu Buzima and the Ministry of Health. LMICs in order to accurately assess patient diagnoses, treatment and outcomes. Third, due to a lack of suffi- Authors’ contributions cient follow-up time to have key outcomes data, we did Conception/Design: LP, VH, JMVD, TM, LS and ST; Data collection: VH, JMVD, TM, IN, TF, and LP; Data analysis: VH, JMVD, LP, RB and BHG; Results not compare outcomes of patients with PABC patients interpretation: All co-authors Manuscript writing: JMVD, VH, LP, RB, BHG to outcomes of patients with non-PABC. We plan to and NG; Final approval of manuscript: All co-authors. study this in the future, when we have more individuals and longer follow-up. Our findings are not necessarily Ethics approval and consent to participate There were two data sources included in this study; both received ethics generalizable to other settings in sub-Saharan Africa, approval from the Rwanda National Ethics Committee (Rwanda) and Partners particularly facilities that are not rural, or hospitals that Institutional Review Board (Boston, USA). The first data source was a chart are not cancer referral facilities. We suspect that in extraction of routinely collected clinical data. Because there was no direct Dusengimana et al. BMC Cancer (2018) 18:634 Page 7 of 8 contact with patients and minimal risks, both ethics committees waived the 17. DeSantis CE, Bray F, Ferlay J, Lortet-Tieulent J, Anderson BO, Jemal A. need for consent for that aspect of the project. The second data source was International variation in female breast cancer incidence and mortality rates. collected as part of an ongoing study [21]; women were consented prior to Cancer Epidemiol Biomark Prev. 2015;24(10):1495–506. https://doi.org/10. data collection as part of the study. The consent obtained was written. 1158/1055-9965. 18. Kantelhardt EJ, Cubasch H, Hanson C. Taking on breast cancer in East Africa: Competing interests global challenges in breast cancer. Curr Opin Obstet Gynecol. 2015;27(1): The authors declare that they have no competing interests 108–14. https://doi.org/10.1097/GCO.0000000000000139. 19. Brinton LA, Figueroa JD, Awuah B, Yarney J, Wiafe S, Woon SN, et al. Breast cancer in sub-Saharan Africa: opportunities for prevention. Breast Cancer Publisher’sNote Res and Treat. 2014;144(3):467–78. https://doi.org/10.1007/s10549-014-2868-z. Springer Nature remains neutral with regard to jurisdictional claims in 20. Anderson BO, Ilbawi AM, El Saghir NS. Breast cancer in low and middle published maps and institutional affiliations. income countries (LMICs): a shifting tide in global health. Breast J. 2015; 21(1):111–8. https://doi.org/10.1111/tbj.12357. Author details 21. Pace LE, Mpunga T, Hategekimana V, Dusengimana JM, Habineza H, 1 2 Partners In Health/Inshuti Mu Buzima, P.O.Box 3432, Kigali, Rwanda. Ministry Bigirimana JB, et al. Delays in breast cancer presentation and diagnosis at of Health, Butaro District Hospital, Butaro, Rwanda. Harvard Medical School, two rural cancer referral centers in Rwanda. Oncologist. 2015;207(7):780–8. Boston, MA, USA. Brigham and Women’s Hospital, Boston, MA, USA. https://doi.org/10.1634/theoncologist.2014-0493. 5 6 University of Pennsylvania, Philadelphia, PA, USA. Dana Farber Cancer 22. Donkor A, Lathlean J, Wiafe S, Vanderpuye V, Fenlon D, Yarney J, et al. Institute, Boston, MA, USA. Factors contributing to late presentation of breast cancer in Africa: a systematic literature review. Arch Med. 2015;8:2. Received: 15 September 2017 Accepted: 21 May 2018 23. Gewefel H, Sarhia B. Breast cancer in adolescent and young adult women. Clinical Breast Cancer. 2014;14(6):390–5. https://doi.org/10.1016/j.clbc.2014.06.002. 24. Mair LP. General characteristics of African family and marriage in African References marriage and social change. New York: Routledge; 2013. p. 1–8. 1. American Cancer Society. Treating Breast Cancer During Pregnancy. Last 25. Miranda JJ, Kinra S, Casas JP, Davey Smith G, Ebrahim S. Non-communicable revised 12/01/2014. http://www.cancer.org/cancer/breastcancer/ diseases in low- and middle-income countries: context, determinants and moreinformation/pregnancy-and-breast-cancer. Accessed 20 July 2016. health policy. Trop Med Int Health. 2008;13(10):1225–34. https://doi.org/10. 2. Van Calsteren K, Heyns L, De Smet F, Van Eycken L, Gziri MM, Van Gemert 1111/j.1365-3156. W, et al. Cancer during pregnancy: An analysis of 215 patients emphasizing 26. Akarolo-Anthony SN, Ogundiran TO, Adebamowo CA. Emerging breast the obstetrical and the neonatal outcomes. J Clin Oncol. 2009;28(4):683–9. cancer epidemic: evidence from Africa. Breast Cancer Res. 2010;12(Suppl 4): https://doi.org/10.1200/JCO.2009.23.2801. S8. https://doi.org/10.1186/bcr2737. 3. Litton JK, Theriault RP, Gonzalez-Anjulo AM. Breast cancer diagnosis during 27. Chiedozi LC, Iweze FI, Aboh IF, Ajabor LN. Breast cancer in pregnancy and pregnancy. Women’s Health (Lond Engl). 2009;5(3):243–9. https://doi.org/10. lactation. Trop Geogr Med. 1988;40(1):26–30. 2217/whe.09.2. 28. Al-Amri AM. Clinical presentation and causes of the delayed diagnosis of 4. Loibl S, von Minckwitz G, Gwyn K, Ellis P, Blohmer JU, Schlegelberger B, et breast cancer in patients with pregnancy associated breast cancer. J Family al. Breast carcinoma during pregnancy: international recommendations from Community Med. 2015;22(2):96–100. https://doi.org/10.4103/2230-8229. an expert meeting. Cancer. 2006;106(2):237–46. 5. Anderson JM. Mammary cancers and pregnancy. Br Med J. 1979;1(6171): 29. Sule EA, Ewemade F. Management of pregnancy associated breast cancer 1124–7. with chemotherapy in a developing country. Int J Surg Case Rep. 2015;17: 6. Hou N, Ogundiran T, Ojengbede O, Morhason-Bello I, Zheng Y, Fackenthal J, 117–20. https://doi.org/10.1016/j.ijscr.2015.10.008. et al. Risk factors for pregnancy-associated breast cancer: a report from the 30. Ibrahim NA, Oludara MA. Socio-demographic factors and reasons associated Nigerian breast Cancer study. Ann Epidemiol. 2013;23(9):551–7. with delay in breast cancer presentation: a study in Nigerian women. Breast. 7. Amant F, von Minckwitz G, Han SN, Bontenbal M, Ring AE, Giermek J, et al. 2012;21(3):416–8. https://doi.org/10.1016/j.breast.2012.02.006. Prognosis of women with primary breast cancer diagnosed during 31. Shachar SS, Gallagher K, McGuire K, Zagar TM, Faso A, Muss HB, et al. pregnancy: results from an international collaborative study. Journal Clin Multidisciplinary Management of Breast Cancer during Pregnancy. Oncol. 2013;31(20):2532–9. https://doi.org/10.1200/JCO.2012.45.6335. Oncologist. 2017;22:324–34. 8. Theriault RL, Litton JK. Pregnancy during or after breast cancer diagnosis: 32. Murthy SS, Tapela N, Muhimpundu MA, Balinda JP, Musabyemariya F, Kirby what do we know and what do we need to know? J Clin Onc. 2013;31: K. A national framework for breast cancer control: a report on Rwanda’s 2521–2. https://doi.org/10.1200/JCO.2013.49.7347. inaugural symposium on the management of breast cancer. J Cancer Policy. 9. Woo JC, Yu T, Hurd TC. Breast cancer in pregnancy: a literature review. Arch 2015;6:3–7. Surg. 2003;138:91–8. discussion 99 33. Tapela NM, Mpunga T, Hedt-Gauthier B, Moore M, Mpanumusingo E, Xu MJ, 10. Bonnier P, Romain S, Dilhuydy JM, Bonichon F, Julien JP, Charpin C, et al. et al. Pursuing equity in cancer care: implementation, challenges and Influence of pregnancy on the outcome of breast cancer: a case-control preliminary findings of a public cancer referral center in rural Rwanda. BMC study. Int J Cancer. 1997;72:720–7. Cancer. 2016;16:237. https://doi.org/10.1186/s12885-016-2256-77. 11. Hassan I, Muhammed I, Attah MM, Maboqunje O. Breast cancer during 34. Shulman LN, Mpunga T, Tapela N, Wagner CM, Fadelu T, Binagwago A. pregnancy and lactation in Zaria, Nigeria. East Afr Med J. 1995;72(5):280–2. Bringing cancer care to the poor: experiences from Rwanda. Nat Rev 12. Polivka J Jr, Altun I, Golubnitschaja O, et al. Pregnancy associated breast Cancer. 2014;14(12):815–21. https://doi.org/10.1038/nrc3848. Cancer: the risky status quo and new concepts of predictive medicine. 35. Mubiligi JM, Hedt-Gauthier B, Mpunga T, Tapela N, Okao P, Harries AD, et al. EPMA J. 2018;9(1):1–13. https://doi.org/10.1007/s13167-018-0129-7. Caring for patients with surgically resectable cancers: experience from a 13. Amant F, Vandenbroucke T, Verheecke M, Fumagalli M, Halaska MJ, Boere I, specialized center in rural Rwanda. Public Health Action. 2014;4(2):128–32. et al. Pediatric outcome after maternal cancer diagnosed during pregnancy. https://doi.org/10.5588/pha.14.0002. N Engl J Med. 2015;373:1824–34. https://doi.org/10.1056/ NEJMoa1508913. 36. O’Neil DS, Keating NL, Dusengimana JMV, Hategekimana V, Umwizera A, 14. Meisel JL, Economy KE, Calvillo KZ, Schapira L, Tung NM, Gelber S, et al. Mpunga T, et al. Quality of breast cancer treatment at a rural cancer center Contemporary multidisciplinary treatment of pregnancy-associated breast in Rwanda. J Global Onc Published online before print May 12, 2017. doi: cancer. Springerplus. 2013;2(1):297. https://doi.org/10.1186/2193-1801-2-297. https://doi.org/10.1200/JGO.2016.008672. 15. Assi HA, Khoury KE, Dbouk H, Khalil LE, Mouhieddine TH, Epidemiology ESNS. Prognosis of breast cancer in young women. J Thorac Dis. 2013; 37. Rubagumya F, Greenbergy L, Manirakiza A, DeBoer R, Park PH, Mpunga T, 5(Suppl 1):S2–8. https://doi.org/10.3978/j.issn.2072-1439.2013.05.24. Shulman LN. Increasing global access to cancer care: models of care with 16. Murphy CG, Mallam D, Stein S, Patil S, Howard J, Sklarin N, et al. Current or non-oncologists as primary providers. Lancet Oncol. 2017;18:1000–2. th recent pregnancy is associated with adverse pathologic features but not 38. Edge SB, Compton CC. The American joint committee on Cancer: the 7 impaired survival in early breast cancer. Cancer. 2012;118(13):3254–9. edition of the AJCC Cancer staging manual and the future of TNM. Ann https://doi.org/10.1002/cncr.26654. Surg Oncol. 2010;17(6):1471–4. Dusengimana et al. BMC Cancer (2018) 18:634 Page 8 of 8 39. National Institute of Statistics of Rwanda, Ministry of Health Rwanda, and ICF International. Rwanda Demographic and Health Survey 2010. Calverton: National Institute of Statistics of Rwanda, Ministry of Health Rwanda, and ICF International; 2012. 40. Andersson TM, Johansson AL, Hsieh CC, Cnattingius S, Lambe M. Increasing incidence of pregnancy-associated breast cancer in Sweden. Obstet Gynecol. 2009;114(3):568–72. https://doi.org/10.1097/AOG.0b013e3181b19154. 41. Yang YL, Chan KA, Hsieh FJ, Chang LY, Wang MY. Pregnancy-associated breast cancer in Taiwanese women: potential treatment delay and impact on survival. PLoS One. 2014;9(11):e111934. https://doi.org/10.1371/journal. pone.0111934. 42. Anderson BO, Petrek JA, Byrd DR, Senie RT, Borgen PI. Pregnancy influences breast cancer stage at diagnosis in women 30 years of age and younger. Ann Surg Oncol. 1996;3:204–11. 43. Ishida T, Yokoe T, Kasumi F, Sakamoto G, Makita M, Tominaga T, Simozuma K, Enomoto K, Fujiwara K, Nanasawa T, et al. Clinicopathologic characteristics and prognosis of breast cancer patients associated with pregnancy and lactation: analysis of case–control study in Japan. Jpn J Cancer Res. 1992;83: 1143–9. https://doi.org/10.1111/j.1349-7006.1992.tb02737.x. 44. Basaran D, Turgal M, Beksac K, Ozyuncu O, Aran O, Beksac MS. Pregnancy- associated breast cancer: Clinicopathological characteristics of 20 cases with focus on identifiable causes of diagnostic delay. Breast Care. 2014;9:355–9. https://doi.org/10.1159/000366436. 45. Galukande M, Kiguli-Malwadde E. Rethinking breast cancer screening strategies in resource-limited settings. Afr Health Sci. 2010;10(1):89–92. 46. Adebamowo CA, Akarolo-Anthony S. Cancer in Africa: opportunities for collaborative research and training. Afr J Med Med Sci. 2009;38(Suppl 2):5–13.

Journal

BMC CancerSpringer Journals

Published: Jun 5, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off