Breast Cancer Research and Treatment (2020) 181:667–678 https://doi.org/10.1007/s10549-020-05600-x EPIDEMIOLOGY Tumour characteristics and survivorship in a cohort of breast cancer: the MCC‑Spain study Inés Gómez‑Acebo1,2,25 · Trinidad Dierssen‑Sotos1,2 · Camilo Palazuelos‑Calderón2 · Beatriz Pérez‑Gómez1,3 · Pilar Amiano5 · Marcela Guevara1,6,7 · Antonio J. Molina8 · Laia Domingo9,10 · María Fernández‑Ortiz2 · Victor Moreno1,11,12,13 · Juan Alguacil1,14 · Guillermo Fernández‑Tardón15 · Josefa Ibáñez16,17 · Rafael Marcos‑Gragera1,18,19 · Marian Diaz‑Santos20 · M. Henar Alonso1,11,12,13 · Jessica Alonso‑Molero2 · Gemma Castaño‑Vinyals1,21,22,23 · Andrés García Palomo24 · Eva Ardanaz1,6,7 · Amaia Molinuevo5 · Nuria Aragonés1,4 · Manolis Kogevinas1,21,22,23 · Marina Pollán1,3 · Javier Llorca1,2 Received: 13 December 2019 / Accepted: 18 March 2020 / Published online: 30 April 2020 © The Author(s) 2020 Abstract Purpose The objective of this study is to analyse the relative survival with breast cancer in women diagnosed after new treatments were generalised and to ascertain the current effect that tumour characteristics such as grade, stage or subtype have on survival as well as the new AJCC-pathological prognostic score. Methods The breast cancer MCC-Spain follow-up study is a prospective cohort study of 1685 incident breast cancer cases. Women between 20 and 85 years old were recruited between the years 2008 and 2013 in 18 hospitals located in 10 Spanish provinces and they have been followed until 2017/2018. Relative survival was estimated after 3, 5 and 8 years of follow-up using Ederer II method. In addition, Weibull regression adjusted by age, hospital, grade and stage was used to investigate prognosis factors. Results Among components of TNM staging system, tumour size greater than 50 mm (i.e. T3 or T4) more than doubled the risk of dying, while N3 nodal involvement and presence of metastasis had a huge effect on mortality. The AJCC pathologi- cal prognostic score strongly correlated with survival; thus, hazard ratios increased as the score rose, being 2.31, 4.00, 4.94, 7.92, 2.26, 14.9 and 58.9 for scores IB, IIA, IIB, IIIA, IIIB, IIIC and IV, respectively. Conclusion Both TNM staging and histological/molecular biomarkers are associated with overall survival in Spanish women with breast cancer; when both are combined in the AJCC pathological prognosis score, the prognostic value improved with risk indices that increased rapidly as the pathological prognosis score increased Keywords Cohort · Epidemiology · Breast cancer · Relative survival · MCC-spain Abbreviations Introduction BC Breast cancer BMI Body mass index Breast cancer (BC) is the most frequent cancer and the fourth MCC-Spain Multicase-control Spain cause of death by cancer in women in developed countries OR Odds ratios [1]. Rates of survival with breast cancer have increased RRR​ Relative risk ratio significantly all over the world in the past decades [2]. In CI Confidence intervals Spain, it has been estimated that 32,825 new BC cases were diagnosed in 2018, with an incidence rate (Age-standardised [World]) of 75.4 cases per 100,000 women-years [3]. Main Electronic supplementary material The online version of this factors associated with prognosis in breast cancer are stage article (https​://doi.org/10.1007/s1054​9-020-05600​-x) contains (enclosing tumour size and local infiltration, lymph nodes supplementary material, which is available to authorised users. and metastasis), tumour grade of differentiation, histologi- cal type and presence/absence of hormone and Her2/neu * Inés Gómez‑Acebo ines.gomez@unican.es receptors [2]. Extended author information available on the last page of the article 13 Vol.:(0123456789) 668 Breast Cancer Research and Treatment (2020) 181:667–678 Although paclitaxel has been available since 1993, breast general surgery, radiotherapy and pathology departments). cancer treatment has considerably improved from 1995 All the cases that were included had histological confirma- onwards as conservative surgery has become widely per- tion and included all the malignant BC (International Classi- formed. Systemic therapy has switched from CMF (cyclo- fication of Diseases 10th Revision (ICD-10) [15] code C50) phosphamide, methotrexate and 5-fluorouracil) to regimens and frequent breast cancers in situ (ICD-10: D05.1, D05.7), containing anthracycline [4], routine searches for oestrogen without a malignant history of BC, diagnosed between 2008 and progesterone receptors has allowed for the introduction and 2013 in the selected hospitals. Patients were residing in of tamoxifen, as well as identification of tumours positive the catchment areas of hospitals for at least 6 months before to Her2/neu has enabled targeted therapy with trastuzumab recruitment. A detailed description of methods and sam- [5], which was authorised in the European Union on August ple characteristics has been published elsewhere [16, 17] 28, 2000. In this regard, identification of intrinsic subtypes and then, 98% recruited patients have been followed until (i.e. luminal A, luminal B, Her2-positive and basal-like) has 2017/2018 (N = 1685). Only patients signing an informed allowed for agreement in general recommendations for first- consent were recruited; the informed consent also asked line therapy in early-stage breast cancer [6]. Survival has them for their permission for later consulting their medical thus enhanced and 5-year survival is now around 85% [7–9]. records during the follow-up. In spite of the improvement in systemic therapy, tumour stage remains a key prognosis factor in breast cancer patients Information obtained at recruitment [10], which reinforces the relevance of early diagnosis. The American Joint Committee for Cancer (AJCC) has recently When the patients were recruited, medical records were suggested a pathological prognostic score combining tumour reviewed in order to gather information on age, date of stage, grade and hormonal/Her2 receptors [11], which has diagnosis, tumour characteristics, including tumour size, been proved effective as survival predictor [12, 13]. lymph nodes, presence of metastases, grade of differentia- Studies carried out in population-based cancer registries tion. Furthermore, oestrogen (ER), progesterone (PgR) and covering 17% of the Spanish population reported that women human epidermal growth factor receptor 2 (HER2) status with BC diagnosed in Spain had 86.0% 5-year age-adjusted were assessed and then classified as positive (+), negative relative survival if diagnosed between 1995 and 1999 [7, 14] (−) and unknown. Specifically, ER and PgR were considered with no improvements for women diagnosed from 2000 to positive if 1% or more of neoplastic cells showed nuclear 2007 [7]. That study, however, included women diagnosed immunohistochemical (IHC) staining. HER2 was considered well before the extension of anthracycline-based chemother- positive (overexpressed) if graded 3 + on IHC performed, apy and the introduction of endocrine therapy with tamox- and all other grades (0 to 2 +) were considered negative ifen and anti-Her2 treatment with trastuzumab. The main unless fluorescence in situ hybridization of 2 + cases con- goal in this study is to analyse the relative survival with BC firmed increased gene copy number. Molecular subtypes of in women diagnosed in 2008–2013—after new treatments breast carcinoma is classified into luminal A (ER + and / were generalised—and to ascertain the current effect that or PR + , HER2-, Ki-67 < 14%), luminal B (ER + and / or tumour characteristics such as grade, stage or subtype have PR + , HER2-, or ER + and / or PR + , HER2-, Ki-67 ≥ 14%), on survival as well as the new AJCC-pathological prognostic HER2 + (ER-, PR-, HER2 +) and basal-like or triple negative score. is a tumour which does not meet any pathologic criteria for positivity of oestrogen receptor, progesterone receptor, or ERBB2 (ER-, PR- and HER2-) [6]. Tumour stage was based Methods on the AJCC standards [18]. Tumour size was classified as T1: ≤ 2 cm, T2: > 2 cm and ≤ 5 cm, T3: > 5 cm, T4: any size Setting and patients with direct extension to chest wall and/or skin; lymph node involvement was classified as N0: no pathologically proven The breast cancer MCC-Spain follow-up study is a prospec- positive lymph nodes, N1: 1–3 positive nodes, N2: 4–9 tive cohort study of incident breast cancer cases diagnosed positive nodes, N3: ≥ 10 positive nodes. Tumour grade was between 2008 and 2013 in 18 hospitals located in 10 Spanish categorised as I: well differentiated, II: moderately differen- provinces (Asturias, Barcelona, Cantabria, Girona, Gipuz- tiated, III: poorly differentiated, according to the modified koa, Huelva, León, Madrid, Navarra and Valencia). 1738 Bloom and Richardson grading system [19]; when Bloom women between 20 and 85 years old with newly diagnosed and Richardson grade was unavailable, other similar clas- primary breast cancer were recruited in the context of a sifications were used as it has been proved they are highly case–control study. The identification of incident cases was reliable [20]. First-line therapy was also recorded, includ- carried out by active search through periodic visits to the ing type of surgery (conservative or mastectomy), margins relevant hospital departments (i.e. gynaecology, oncology, (free or invaded), chemotherapy regime, radiotherapy, 13 Breast Cancer Research and Treatment (2020) 181:667–678 669 endocrine therapy, Her2-targeted therapy; for all systemic The mean age at diagnosis was 56.5 ± 12.6-year-old and therapies, whether they were adjuvant or neoadjuvant was 65% were postmenopausal. Other main characteristics of the also recorded. patients are shown in supplementary Table 1. Tumour characteristics Follow‑up Regarding the characteristics of the tumours, the most usual In 2017 and 2018, medical records were reviewed to obtain histological type was ductal (75.7%), followed by lobular the last date the patient was attended to in t the health system (6.5%). About half of cancers were T1 (52%) and lymph and to ascertain current patient’s vital status. For patients node negatives (52%). Only 42 women (2.5%) had metasta- that had not have been at hospital in the 3 months before sis at the time of diagnosis. their medical record review, the National Death Index Concerning intrinsic subtypes, 997 (59.2%) could be clas- (Índice Nacional de Defunciones) was consulted [21]. This sified as luminal A-like, 331 (19.6%) as luminal B-like, 81 index is a database recording date of death for all people (4.8%) as Her2 (non-luminal)-like and 130 (7.7%) as basal- who have died in Spain; it is hosted on the Ministry of like. Grade of differentiation could not be obtained from Health web page and can only be accessed with specific medical records in 481 patients (28.5%). Almost 31% of the authorisation for research purposes. Last date of search in cancers were moderately differentiated while poorly differ- the National Death Index was in December 2018. entiated accounted for about 21% cancers. Hormone recep- tor status was available for most cases; 83% were oestrogen Statistical analysis receptor positive, 73% progesterone positive and 17.4% were Her2 positive (Table 1). Deaths by any cause were considered events and patients alive in their last contact with hospital and not included in Relative survival the National Death Index were considered censored. Time of follow-up was considered as the time from diagnosis to 5-year relative survival with breast cancer was 93% (95% death or to the last contact in hospital in alive patients. Rela- CI 92 – 94) (Fig. 1a). Table 1 shows 3-, 5- and 8-year rela- tive survival, defined as the ratio of the observed survival tive survival according to tumour characteristics. Women in our study population and the expected survival in general diagnosed in stage I had the same survival probability than population Spanish women with the same age as cases in women with the same age without breast cancer (i.e. 100% the year of diagnosis, was estimated using Ederer II method relative survival) even after 8 years of follow-up. At the [22]; for this purpose, age-specific mortality rates from same time, relative survival decreased with follow-up time 2007 to 2018 obtained from the Spanish National Institute in women diagnosed in more advanced stages: 3-, 5- and for Statistics (INE) were used as reference. 3-, 5- and 8-year 8-year relative survival were 98%, 95% and 93% for breast survival Kaplan–Meier estimates were obtained. Weibull cancer diagnosed in stage II, 94%, 88% and 81% in women regression adjusting by age, hospital, grade and stage was diagnosed in staged III and 63%, 40% and 24% in those used to investigate prognosis factors; its results are displayed diagnosed in staged IV (Table 1 and Fig. 1b). Relative sur- as hazard ratios with their 95% confidence intervals. Groups vival also decreased as grading got less differentiated [8-year that included less than 25 people were excluded from this relative survival: 99% in well differentiated tumours, 93% in analysis. Stata 14/SE package was used in the analyses. moderately differentiated and 88% in poorly differentiated (Table 1 and Fig. 1c)]. Relative survival after 8 years of follow-up was 94% for luminal A-like breast cancers, 88% for luminal B-like, 82% for Her2 non-luminal) and 74% for Results basal-like cancers (Table 1 and Fig. 1d). Characteristics of patients Prognostic factors on overall mortality We followed 1685 women out of 1738 breast cancer patients Age, hospital, stage and grade-adjusted hazard ratios on (98%), who had primary breast cancer between 2007 and association between survival and tumour characteristics 2013. Patient-years of follow-up were 10,931 person-years. and first-line treatment are displayed in Table 2. Age, The maximum period for breast cancer follow-up was nine premenopausal status, tumour size, nodal infiltration and and a half years (from 13th July 2007 to 22nd March 2018) presence of metastasis were significantly associated with and 206 patients died in the follow-up; 5-year survival was overall mortality. Hazard ratios increased with tumour 90.7%. stage, being 1 (reference) for T1 and 1.62 (1.04 – 2.52) 13 670 Breast Cancer Research and Treatment (2020) 181:667–678 Table 1  Relative survival probability 3, 5 and 8 years after breast cancer diagnosis according to tumour characteristics Variable Category N (%) 3-year 5-year 8-year Survival (95% CI) Survival (95% CI) Survival (95% CI) Histology Ductal 1276 (75.73) 97 (95–98) 93 (92–95) 91 (88–93) Lobular 110 (6.53) 95 (88–98) 89 (80–94) 87 (76–94) Other 299 (17.74) 100 (98–101) 97 (94–99) 97 (92–100) Tumour size T1 876 (51.99) 100 (99–101) 98 (97–99) 97 (95–99) T2 424 (25.16) 96 (93–98) 91 (87–94) 87 (81–91) T3 73 (4.33) 88 (77–94) 75 (63–85) 71 (58–81) T4 39 (2.31) 74 (56–86) 59 (41–74) 59 (40–75) Tis 109 (6.47) 101 (101–101) 102 (102–102) 100 (91–103) Missing 141 (8.37) 93 (86–97) 87 (80–93) 81 (69–90) Node infiltration N0 877 (52.05) 100 (99–100) 97 (95–99) 98 (96–100) N1 441 (26.17) 97 (95–99) 94 (91–96) 90 (85–94) N2 142 (8.43) 93 (86–97) 85 (77–91) 78 (68–86) N3 49 (2.91) 81 (67–91) 77 (61–87) 61 (40–78) Missing 176 (10.45) 92 (86–96) 88 (81–93) 79 (69–87) Tumour stage 0 112 (6.65) 100 (95–101) 101 (96–102) 98 (89–102) I 594 (35.25) 101 (99–101) 99 (97–101) 100 (97–102) II 498 (29.55) 98 (96–100) 95 (92–97) 93 (88–96) III 187 (11.10) 94 (89–97) 88 (82–93) 81 (72–88) IV 42 (2.49) 63 (46–77) 40 (24–55) 24 (11–40) Oestrogen receptor Positive 1398 (82.97) 99 (98–100) 96 (94–97) 93 (91–95) Negative 244 (14.48) 86 (81–90) 81 (75–86) 80 (73–85) Progesterone receptor Positive 1237 (73.41) 100 (98–100) 97 (95–98) 95 (92–97) Negative 401 (23.80) 90 (86–93) 85 (80–88) 81 (76–85) Her2 Negative 1250 (74.18) 97 (96–98) 93 (92–95) 91 (88–93) Positive 294 (17.45) 97 (94–99) 93 (89–96) 89 (84–93) Intrinsic subtype Luminal A 997 (59.17) 100 (98–100) 96 (94–98) 94 (92–97) Luminal B 331 (19.64) 97 (95–99) 94 (90–97) 88 (83–93) Her2 81 (4.81) 94 (86–99) 87 (76–93) 82 (70–91) Basal-like 130 (7.72) 78 (70–85) 74 (65–81) 74 (64–82) Luminal ONI 91 (5.40) 100 (94–101) 101 (95–102) 101 (92–104) Grade I: Well differentiated 329 (19.53) 100 (98–101) 99 (96–101) 99 (95–102) II: Moderately differentiated 520 (30.86) 100 (98–101) 95 (92–97) 93 (89–96) III: Poorly differentiated 355 (21.07) 93 (89–95) 90 (86–94) 88 (83–92) Missing 481 (28.55) 96 (93–98) 91 (88–94) 88 (83–91) ONI Otherwise non-identified, Luminal ONI hormonal receptors positive, Her2 missing, Non-luminal ONI hormonal receptors negative, Her2 missing and 2.04 (1.13 – 3.68) for T2, T3, respectively. Breast and 2.42 (1.41 – 4.17) for poorly differentiated. To further cancers negative for oestrogen or progesterone receptors explore whether intrinsic subtype adds prognosis value to behaved worse than their opposite. Luminal A- like and tumour stage, we studied the effect of stage by stratifying luminal B-like tumours had similar prognosis; hazard for for intrinsic subtype; results in Fig. 2a show that higher Her2 (non-luminal) cancers were 50% higher than for stages had higher hazard ratios, whatever the intrinsic luminal A-like (hazard ratio = 1.50; 95% CI 0.87 – 2.59), subtype. However, when stratifying the effect of intrinsic and basal-like tumour’s hazard was three times that of subtype for tumour stage, we found that intrinsic subtype luminal A-like (hazard ratio = 3.50; 95% CI 2.31 – 5.30). had no effect at all in patients in stage I, while Her2 (non- Grading showed a dose–response association with mortal- luminal) and basal-like tumours had worse prognosis than ity, with hazard ratios 1 (reference) for well differentiated luminal-like tumours in patients with stage II and, espe- cancers, 1.48 (0.86 – 2.54) for moderately differentiated cially, stages III and IV (Fig. 2b). 13 Breast Cancer Research and Treatment (2020) 181:667–678 671 Fig. 1  Relative survival in Spanish women with breast cancer: a Overall survival, b survival according to TNM staging, c survival according to grading, d survival according to intrinsic subtype Pathologic prognostic score in women were negative in 66% patients. Radiotherapy was used in 1158 without neoadjuvant therapy women (68.7%), while adjuvant or neoadjuvant chemotherapy was administered to 50.13% patients. Adjuvant or neoadjuvant Figure 3 displays the association between the AJCC- Patho- hormone therapy was used in 1023 women (60.7%), immu- logic Prognostic Score and overall survival in women with- notherapy in 11.1% and Her2-targeted therapy in 152 patients out neoadjuvant therapy. Hazard ratios stepped up as the (9.0%). score rose, being 2.31, 4.00, 4.94, 7.92, 2.26, 14.9 and 58.9 Hazard ratios on association between survival and first-line for scores IB, IIA, IIB, IIIA, IIIB, IIIC and IV, respectively. treatment are displayed in Table 4. Type of surgery (mastec- These results are somewhat limited because of their wide tomy/conservative surgery) and surgical margins were not confidence intervals. This information appears stratified by associated with survival. Among systemic treatments, adju- menopausal status in supplementary material, where hazard vant therapy (whether chemo-, radio- or hormone-therapy) ratios increased faster in premenopausal women (Supple- was associated with lower mortality and both neoadjuvant mentary Table 2). and palliative treatment were associated with worse progno- sis (Table 4). First‑line treatment for patients Table 3 shows the characteristics of the treatment. Conserva- tive surgery was performed in 1231 (73.1%) patients and mastectomy in the remaining 454 (26.9%). Surgical margins 13 672 Breast Cancer Research and Treatment (2020) 181:667–678 Table 2  Characteristics of the Variable Category N (%) Hazard ratio (95% CI)* p tumour and their association with survival Histology Ductal 1276 (75.73) 1 (ref) Lobular 110 (6.53) 1.12 (0.68 to 1.83) 0.666 Other 299 (17.74) 0.44 (0.26 to 0.72) 0.001 Tumour size T1 876 (51.99) 1 (ref) T2 424 (25.16) 1.62 (1.04 to 2.52) 0.031 T3 73 (4.33) 2.04 (1.13 to 3.68) 0.018 T4 39 (2.31) 1.96 (0.98 to 3.91) 0.056 Tis 109 (6.47) 0.14 (0.02 to 0.91) 0.04 Missing 141 (8.37) 1.12 (0.61 to 2.06) 0.718 Node infiltration N0 877 (52.05) 1 (ref) N1 441 (26.17) 1.16 (0.76 to 1.78) 0.481 N2 142 (8.43) 1.17 (0.58 to 2.36) 0.651 N3 49 (2.91) 1.73 (0.81 to 3.72) 0.158 Missing 176 (10.45) 1.15 (0.65 to 2.03) 0.641 Tumour stage 0 112 (6.65) 0.68 (0.26 to 1.81) 0.444 I 594 (35.25) 1 (ref) II 498 (29.55) 2.30 (1.46 to 3.61) > 0.001 III 187 (11.10) 4.23 (2.60 to 6.88) > 0.001 IV 42 (2.49) 29.5 (16.9 to 51.6) > 0.001 Oestrogen receptor Positive 1398 (82.97) 1 (ref) Negative 244 (14.48) 2.31 (1.65 to 3.23) > 0.001 Progesterone receptor Positive 1237 (73.41) 1 (ref) Negative 401 (23.80) 2.13 (1.58 to 2.87) > 0.001 Her2 Negative 1250 (74.18) 1 (ref) Positive 294 (17.45) 0.79 (0.55 to 1.14) 0.209 Intrinsic subtype Luminal A 997 (59.17) 1 (ref) Luminal B 331 (19.64) 1.16 (0.81 to 1.66) 0.429 Her2 81 (4.81) 1.50 (0.87 to 2.59) 0.143 Basal-like 130 (7.72) 3.50 (2.31 to 5.30) > 0.001 Luminal ONI 91 (5.40) 0.20 (0.05 to 0.82) 0.026 Grade I: Well differentiated 329 (19.53) 1 (ref) II: Moderately differentiated 520 (30.86) 1.48 (0.86 to 2.54) 0.157 III: Bad differentiated 355 (21.07) 2.42 (1.41 to 4.17) 0.001 missing 481 (28.55) 2.31 (1.27 to 4.18) 0.006 Hazard ratios and 95% confidence intervals estimated via Weibull regression *Hazard ratios adjusted for age, hospital, stage and grade ONI otherwise not identified Discussion grading and hormonal/Her2 receptors) strongly correlated with survival. In spite of recent improvements in breast cancer systemic TNM staging has largely been the main factor regarding therapy, our results confirm that tumour stage and grad- survival with breast cancer, which led the AJCC to suggest ing are major prognosis factors regarding overall survival its Anatomic Stage Groups [11]. Our results on TNM com- with breast cancer. Among components of TNM stag- ponents are consistent with recent studies still proving that ing system, tumour size greater than 50 mm (i.e. T3 or TNM classification is a main contributor to survival [10, T4) more than doubled the risk of dying, while N3 nodal 23–25]. For instance, a large study with data from Nether- involvement and presence of metastasis had a huge effect lands Cancer Registry reported hazard ratios over 2.5 for T3 on mortality. The pathological prognostic score suggested or T4 tumours and 4.0 for N3 tumours in women diagnosed by the AJCC in 2017 combining anatomic characteristics between 2006 and 2012. These hazard ratios were similar (i.e. TNM staging) with biological characteristics (i.e. to those obtained in the same registry for women diagnosed 13 Breast Cancer Research and Treatment (2020) 181:667–678 673 (a) (b) Fig. 2  Hazard ratios for overall survival showing the interaction between TNM stage and intrinsic subtype. a Hazard ratios for TNM stage strati- fied by intrinsic subtype, b hazard ratios for intrinsic subtype stratified by TNM stage Fig. 3  Kaplan–Meier survival estimates according to the AJCC pathological prognostic score 13 674 Breast Cancer Research and Treatment (2020) 181:667–678 Table 3  First-line treatment and relative survival probability 3, 5 and 8 years after diagnosis Variable Category N (%) 3-year 5-year 8-year Survival (95% CI) Survival (95% CI) Survival (95% CI) Surgery Conservative 1231 (73.06) 98 (97–99) 95 (94–97) 93 (91–95) Mastectomy 454 (26.94) 94 (91–97) 89 (86–92) 87 (82–91) Surgical margins Negative 1104 (65.52) 98 (97–99) 95 (93–97) 94 (91–96) Positive 244 (14.48) 100 (97–101) 98 (94–100) 95 (89–98) missing 337 (20.00) 92 (88–95) 87 (82–90) 82 (76–87) Chemotherapy No 817 (48.49) 98 (97–100) 96 (94–98) 95 (91–97) Neoadjuvant 180 (10.68) 95 (90–97) 87 (81–92) 83 (76–88) Adjuvant 663 (39.35) 99 (97–100) 96 (94–98) 93 (90–96) Palliative 25 (1.48) 40 (21–59) 16 (5–33) Radiotherapy No 334 (22.45) 95 (92–98) 91 (87–95) 88 (81–93) Neoadjuvant 5 (0.34) 61 (13–89) 21 (1–60) Adjuvant 1128 (75.81) 100 (99–100) 97 (96–98) 95 (93–97) Palliative 21 (1.41) 31 (13–52) 16 (4–35) Immunotherapy No 1498 (88.90) 98 (97–99) 95 (93–96) 93 (90–95) Neoadjuvant 27 (1.60) 90 (70–97) 75 (54–88) 67 (45–83) Adjuvant 144 (8.55) 98 (94–100) 95 (89–98) 92 (85–97) Palliative 16 (0.95) 41 (17–64) 28 (9–51) Endocrine therapy No 662 (39.29) 94 (91–96) 90 (87–92) 86 (82–90) Neoadjuvant 19 (1.13) 103 (103–103) 76 (48–93) 72 (40–94) Adjuvant 985 (58.46) 100 (99–101) 98 (97–99) 97 (94–99) Palliative 19 (1.13) 61 (35–80) 26 (9–49) Table 4  First-line treatment and Variable Category N (%) Hazard ratio (95% CI) * p their association with survival Surgery Conservative 1231 (73.06) 1 (ref) Mastectomy 454 (26.94) 0.83 (0.56 to 1.22) 0.339 Surgical margins Negative 1104 (65.52) 1 (ref) Positive 244 (14.48) 0.84 (0.50 to 1.42) 0.518 Missing 337 (20.00) 1.03 (0.67 to 1.56) 0.905 Chemotherapy No 817 (48.49) 1 (ref) Neoadjuvant 180 (10.68) 1.18 (0.72 to 1.91) 0.514 Adjuvant 663 (39.35) 0.60 (0.40 to 0.89) 0.010 Palliative 25 (1.48) 8.44 (4.61 to 15.4) < 0.001 Radiotherapy No 334 (22.45) 1 (ref) Neoadjuvant 5 (0.34) 6.05 (1.88 to 19.5) 0.003 Adjuvant 1128 (75.81) 0.44 (0.30 to 0.65) < 0.001 Palliative 21 (1.41) 5.31 (2.57 to 11.0) < 0.001 Immunotherapy No 1498 (88.90) 1 (ref) Neoadjuvant 27 (1.60) 1.79 (0.87 to 3.67) 0.113 Adjuvant 144 (8.55) 0.83 (0.48 to 1.44) 0.518 Palliative 16 (0.95) 5.22 (2.57 to 10.6) < 0.001 Endocrine therapy No 662 (39.29) 1 (ref) Neoadjuvant 19 (1.13) 1.17 (0.50 to 2.76) 0.719 Adjuvant 985 (58.46) 0.51 (0.37 to 0.71) < 0.001 Palliative 19 (1.13) 2.18 (1.05 to 4.56) 0.037 Hazard ratios and 95% confidence intervals estimated via Weibull regression *Hazard ratios adjusted for age, hospital, stage and grade 13 Breast Cancer Research and Treatment (2020) 181:667–678 675 between 1999 and 2005 [10], which reinforces TNM staging that missing categories did not follow a clear pattern in our relevance for survival has not changed with recent treatment results as missing data in TNM components behaved almost improvements, thus strengthening the importance of early as the reference category (i.e. good prognostic), while diagnosis. missing data in grading performed as poorly differentiated Recent developments, however, have focused on tumour tumours. Therefore, it is not possible for us to attribute any biology, making the identification of breast cancer subtypes data omission in medical records to its irrelevance. Thirdly, via differentiation grade or presence of hormonal and Her2/ our sample could be biased towards cancers with better prog- neu receptors possible, heading to specific therapies accord- nostic as women with advanced disease could have refused ing to tumour histological and molecular characteristics to participate, thus underrepresenting women in stage IV; [6, 26–28]. Our results also confirmed higher survival for had this occurred, it could have overestimated relative sur- women with breast cancer positive to either oestrogen or vival; however, this problem should have not biased hazard progesterone receptors and negative to HER2 receptors, as ratios estimates. Finally, the multicentric design of our study well as worsening prognostic as grading increased from well is both a limitation and a strength. A limitation as routine to poorly differentiated tumour. In this regard, the new AJCC clinical practice could vary from hospital to hospital, eventu- Pathological Prognostic Score [11] has been suggested by ally increasing random variation; we dealt with this problem combining both anatomic staging and biological markers, by adjusting all our analyses for hospital. Nevertheless, it is thus refining the previous AJCC Anatomic Stage Groups also a strength as it obtained a fair representation of usual [12], which is now intended only for regions were biological clinical practice in Spain about 2008–2013. An additional markers are not routinely available [11]. In our cohort, the strength is the active prospective follow-up, carried out via AJCC Pathological Prognostic Score had large prognostic three complementary sources: medical records, phone calls value with hazard ratios swiftly rising as the Pathological to the patient and National Death Index consultation. Prognostic Score increased. In addition, this score presents a higher predictive ability in premenopausal women. This can be explained by (i) the rate at which the cancer devel- Conclusions ops—tumours could be faster and receive poorer prognosis in premenopausal women—and (ii) the fact that a younger In conclusion, both TNM staging and histological/molecular woman, i.e. premenopausal, with breast cancer is more likely biomarkers are associated with overall survival in Spanish to die because of the tumour itself, as opposed to postmeno- women with breast cancer; when both are combined in the pausal women, who have more chances to die due to other AJCC Pathological Prognostic Score, its prognostic value causes. Of note, the analysis of Pathological Prognostic vastly improved in women without neoadjuvant therapy. Score should have been restricted to women without neo- adjuvant therapy as this score is based on the pathological Acknowledgements The authors thank all patients and families for their participation in the MCC-Spain study, as well as all study centres TNM staging, which is obtained after surgery. and collaborators. Adjuvant chemotherapy, radiotherapy or endocrine therapy was associated with better prognosis in our cohort, Author contributions Study conception and design: IG-A, TD-S and which agrees with previous reports [4, 29–31]. Neoadju- JL. Statistical analysis: IG-A, TD-S and JL. Coordination of substudy vant therapy, however, was associated with lower survival. sites, recruitment and acquisition of data: IG-A, TD-S, CP, BP-G, PA, MG, AJM, LD, MF-O, VM, JA, GF-T, JI, RM-G, MD-S, MHA, JA-M, Results on systemic therapy—survival relationship can- GC-V, AGP, EA, AM, NA, MK, MPM, JL. Drafting of the manuscript: not be straight forwardly interpreted as, even after adjust- IG-A, TD-S and JL. Contributions to the final version of the manuscript ing for stage and grade, there is still room for confounding were made by all authors. by indication as women with more severe cancer are more likely treated with more aggressive therapies. For instance, Funding Biological samples were stored at the biobanks supported by Instituto de Salud Carlos III- FEDER: Parc de Salut MAR Biobank neoadjuvant therapy is not usually recommended for low (MARBiobanc) (RD09/0076/00036), ‘Biobanco La Fe’ (RD 09 proliferating cancers and hormone receptor-positive, HER2- 0076/00021) and FISABIO Biobank (RD09 0076/00058), as well as negative, grading I–II cancers [26]; therefore, its high hazard at the Public Health Laboratory of Gipuzkoa, the Basque Biobank, the ratios could be due to residual confounding as it is frequently ICOBIOBANC (sponsored by the Catalan Institute of Oncology), the IUOPA Biobank of the University of Oviedo and the ISCIII Biobank. indicated in cancers with worse prognostic. SNP genotyping services were provided by the Spanish ‘Centro Our study has some limitations. Firstly, the sample size— Nacional de Genotipado’ (CEGEN-ISCIII). We thank all the subjects although enough for general results—is small for studying who participated in the study and all MCC-Spain collaborators. This specific subgroups or interactions. Secondly, basal infor- work was supported by the ‘Acción Transversal del Cancer’, approved by the Spanish Ministry Council on the 11th October 2007, by the mation was obtained from medical records, which could Instituto de Salud Carlos III, co-founded by FEDER funds – ‘a way to have partial information on some variables, leading to some build Europe’ (grants PI08/1770, PI08/0533, PI08/1359, PI09/00773, amount of missing data. On this subject, it is noteworthy PI09/01286, PI09/01903, PI09/02078, PI09/01662, PI11/01403, 13 676 Breast Cancer Research and Treatment (2020) 181:667–678 PI11/01889, PI11/00226, PI11/01810, PI11/02213, PI12/00488, survival in modern times: population based study in 173 797 PI12/00265, PI12/01270, PI12/00715, PI12/00150, PI14/01219, patients. BMJ 351:h4901. https​://doi.org/10.1136/bmj.h4901​ PI14/00613 and PI15/00069). Support was also provided by the Fun- 3. Cancer today. https​://gco.iarc.fr/today​/home. Accessed 15 Apr dación Marqués de Valdecilla (grant API 10/09); the Junta de Castilla 2019 y León (grant LE22A10-2); the Consejería de Salud of the Junta de 4. Clarke M, Collins R, Darby S, et al (2005) Effects of chemother- Andalucía (2009-S0143); the Conselleria de Sanitat of the Generali- apy and hormonal therapy for early breast cancer on recurrence tat Valenciana (grant AP 061/10); the Recercaixa (grant 2010ACUP and 15-year survival: an overview of the randomised trials. Centre 00310); the Regional Government of the Basque Country; the Con- for Reviews and Dissemination (UK) sejería de Sanidad de la Región de Murcia; European Commission 5. Guevara M, Burgui R, Díaz-González J et al (2017) Care pat- grants FOOD-CT-2006–036224-HIWATE; the Spanish Association terns and changes in treatment for nonmetastatic breast cancer in Against Cancer (AECC) Scientific Foundation; the Catalan Govern- 2013–2014 versus 2005: a population-based high-resolution study. ment DURSI (grant 2014SGR647); the Fundación Caja de Ahorros de Eur J Cancer Prev. https:​ //doi.org/10.1097/CEJ.000000​ 00000​ 0037​ Asturias; the University of Oviedo; Societat Catalana de Digestologia; 9 and COST action BM1206 Eucolongene. 6. Goldhirsch A, Winer EP, Coates AS et al (2013) Personalizing the treatment of women with early breast cancer: highlights of the St gallen international expert consensus on the primary Compliance with ethical standards therapy of early breast cancer 2013. Ann Oncol Off J Eur Soc Med Oncol 24:2206–2223. https​: //doi.org/10.1093/annon​c / Conflict of interest The authors declare that they have no conflict of mdt30​3 interest. 7. Chirlaque MD, Salmerón D, Galceran J et al (2018) Cancer survival in adult patients in Spain. results from nine population- Ethical approval All procedures performed in studies involving human based cancer registries. Clin Transl Oncol 20:201–211. https​:// participants were in accordance with the ethical standards of the insti- doi.org/10.1007/s1209​4-017-1710-6 tutional and/or national research committee and with the 1964 Helsinki 8. Sant M, Chirlaque Lopez MD, Agresti R et al (1990) (2015) declaration and its later amendments or comparable ethical standards. Survival of women with cancers of breast and genital organs in Europe 1999–2007: results of the EUROCARE-5 study. Eur Research involving human and animal rights This article does not con- J Cancer Oxf Engl 51:2191–2205. https​: //doi.org/10.1016/j. tain any studies with animals performed by any of the authors. ejca.2015.07.022 9. Allemani C, Weir HK, Carreira H et al (2015) Global surveil- Informed consent The protocol of MCC-Spain was approved by the lance of cancer survival 1995–2009: analysis of individual data Ethics committees of the participating institutions [16] At recruitment, for 25,676,887 patients from 279 population-based registries in all participants were informed about the study objectives and signed an 67 countries (CONCORD-2). Lancet Lond Engl 385:977–1010. informed consent, which also included the authorization for following https​://doi.org/10.1016/S0140​-6736(14)62038​-9 up the patient via medical records or phone calls; only participants 10. Saadatmand S, Bretveld R, Siesling S, Tilanus-Linthorst MMA agreeing in being followed up were included in the inception cohorts. (2015) Influence of tumour stage at breast cancer detection on Confidentiality of data is secured by removing personal identifiers in survival in modern times: population based study in 173,797 the datasets. The database was registered in the Spanish Agency for patients. BMJ 351:h4901. https​://doi.org/10.1136/bmj.h4901​ Data Protection, number 2102672171. Permission to use the study 11. Giuliano AE, Edge SB, Hortobagyi GN (2018) Eighth edition of database will be granted to researchers outside the study group after the AJCC cancer staging manual: breast cancer. Ann Surg Oncol revision and approval of each request by the Steering Committee. 25:1783–1785. https​://doi.org/10.1245/s1043​4-018-6486-6 12. Lee SB, Sohn G, Kim J et al (2018) A retrospective prognostic evaluation analysis using the 8th edition of the American joint Open Access This article is licensed under a Creative Commons Attri- committee on cancer staging system for breast cancer. Breast bution 4.0 International License, which permits use, sharing, adapta- Cancer Res Treat 169:257–266. https​://doi.org/10.1007/s1054​ tion, distribution and reproduction in any medium or format, as long 9-018-4682-5 as you give appropriate credit to the original author(s) and the source, 13. Ibis K, Ozkurt S, Kucucuk S et al (2018) Comparison of Patho- provide a link to the Creative Commons licence, and indicate if changes logical Prognostic Stage and Anatomic Stage Groups Accord- were made. The images or other third party material in this article are ing to the Updated Version of the American Joint Committee included in the article’s Creative Commons licence, unless indicated on Cancer (AJCC) breast cancer staging 8th edition. Med Sci otherwise in a credit line to the material. If material is not included in Monit Int Med J Exp Clin Res 24:3637–3643. https​: //doi. the article’s Creative Commons licence and your intended use is not org/10.12659​/MSM.91102​2 permitted by statutory regulation or exceeds the permitted use, you will 14. Chirlaque MD, Salmerón D, Ardanaz E et al (2010) Cancer need to obtain permission directly from the copyright holder. To view a survival in Spain: estimate for nine major cancers. Ann Oncol copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. 3:21–29. https​://doi.org/10.1093/annon​c/mdq08​2 15. WHO | ICD-10 online versions. In: WHO. https​://www.who.int/ class​ifica​tions​/icd/icdon​linev​ersio​ns/en/. Accessed 9 Sep 2019 16. Castaño-Vinyals G, Aragonés N, Pérez-Gómez B et al (2015) Population-based multicase-control study in common tumors References in Spain (MCC-Spain): rationale and study design. Gac Sanit 29:308–315. https​://doi.org/10.1016/j.gacet​a.2014.12.003 1. Bray F, Ferlay J, Soerjomataram I et al (2018) Global cancer sta- 17. Alonso-Molero J, Molina AJ, Jiménez-Moleón JJ et al (2019) tistics 2018: GLOBOCAN estimates of incidence and mortality Cohort profile: the MCC-Spain follow-up on colorectal, breast worldwide for 36 cancers in 185 countries. CA Cancer J Clin. and prostate cancers: study design and initial results. BMJ https​://doi.org/10.3322/caac.21492​ Open. https​://doi.org/10.1136/bmjop​en-2019-03190​4 2. Saadatmand S, Bretveld R, Siesling S, Tilanus-Linthorst MMA 18. Edge SB, Compton CC (2010) The American Joint Committee (2015) Influence of tumour stage at breast cancer detection on on Cancer: the 7th Edition of the AJCC Cancer Staging Manual 13 Breast Cancer Research and Treatment (2020) 181:667–678 677 and the Future of TNM. Ann Surg Oncol 17:1471–1474. https​ 26. Gnant M, Harbeck N, Thomssen C (2011) St. Gallen 2011: sum- ://doi.org/10.1245/s1043​4-010-0985-4 mary of the consensus discussion. Breast Care 6:136–141. https​ 19. Rakha EA, El-Sayed ME, Lee AHS et al (2008) Prognostic sig- ://doi.org/10.1159/00032​8054 nificance of Nottingham histologic grade in invasive breast car- 27. Coates AS, Winer EP, Goldhirsch A et al (2015) Tailoring thera- cinoma. J Clin Oncol Off J Am Soc Clin Oncol 26:3153–3158. pies–improving the management of early breast cancer: St. gal- https​://doi.org/10.1200/JCO.2007.15.5986 len international expert consensus on the primary therapy of 20. Davis BW, Gelber RD, Goldhirsch A et al (1986) Prognostic early breast cancer 2015. Ann Oncol Off J Eur Soc Med Oncol significance of tumor grade in clinical trials of adjuvant therapy 26:1533–1546. https​://doi.org/10.1093/annon​c/mdv22​1 for breast cancer with axillary lymph node metastasis. Cancer 28. Curigliano G, Burstein HJ, Winer EP et al (2017) De-escalating 58:2662–2670 and escalating treatments for early-stage breast cancer: the St. 21. Navarro C (2006) The national death index: a largely expected gallen international expert consensus on the primary therapy of advance in the access to mortality data. Gac Sanit 20:421–423 early breast cancer 2017. Ann Oncol 28:1700–1712. https​://doi. 22. Dickman PW, Coviello E (2015) Estimating and modeling relative org/10.1093/annon​c/mdx30​8 survival. Stata J 15:186–215. https:​ //doi.org/10.1177/153686​ 7X15​ 29. Martinez FJ, Calverley PM, Goehring UM, Brose M, Fabbri LM, 01500​112 Rabe KF (2011) Effect of radiotherapy after breast-conserving 23. Colzani E, Liljegren A, Johansson ALV et al (2011) Prognosis surgery on 10-year recurrence and 15-year breast cancer death: of patients with breast cancer: causes of death and effects of time meta-analysis of individual patient data for 10 801 women in 17 since diagnosis, age, and tumor characteristics. J Clin Oncol randomised trials. Lancet 378:1707–1716. https:​ //doi.org/10.1016/ 29:4014–4021. https​://doi.org/10.1200/JCO.2010.32.6462 S0140​-6736(11)61629​-2 24. Pestalozzi BC, Zahrieh D, Mallon E et al (2008) Distinct clinical 30. Bergh J, Jönsson PE, Glimelius B et al (2001) A systematic over- and prognostic features of infiltrating lobular carcinoma of the view of chemotherapy effects in breast cancer. Acta Oncol Stockh breast: combined results of 15 international breast cancer study Swed 40:253–281 group clinical trials. J Clin Oncol 26:3006–3014. https​://doi. 31. Corona SP, Roviello G, Strina C et al (2019) Efficacy of extended org/10.1200/JCO.2007.14.9336 aromatase inhibitors for hormone-receptor-positive breast cancer: 25. Dunnwald LK, Rossing MA, Li CI (2007) Hormone receptor sta- a literature-based meta-analysis of randomized trials. Breast Edinb tus, tumor characteristics, and prognosis: a prospective cohort Scotl 46:19–24. https​://doi.org/10.1016/j.breas​t.2019.04.004 of breast cancer patients. Breast Cancer Res 9:R6. https​://doi. org/10.1186/bcr16​39 Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Affiliations Inés Gómez‑Acebo1,2,25 · Trinidad Dierssen‑Sotos1,2 · Camilo Palazuelos‑Calderón2 · Beatriz Pérez‑Gómez1,3 · Pilar Amiano5 · Marcela Guevara1,6,7 · Antonio J. Molina8 · Laia Domingo9,10 · María Fernández‑Ortiz2 · Victor Moreno1,11,12,13 · Juan Alguacil1,14 · Guillermo Fernández‑Tardón15 · Josefa Ibáñez16,17 · Rafael Marcos‑Gragera1,18,19 · Marian Diaz‑Santos20 · M. Henar Alonso1,11,12,13 · Jessica Alonso‑Molero2 · Gemma Castaño‑Vinyals1,21,22,23 · Andrés García Palomo24 · Eva Ardanaz1,6,7 · Amaia Molinuevo5 · Nuria Aragonés1,4 · Manolis Kogevinas1,21,22,23 · Marina Pollán1,3 · Javier Llorca1,2 1 9 Consortium for Biomedical Research in Epidemiology Department of Epidemiology and Evaluation, IMIM and Public Health (CIBER Epidemiología y Salud Pública, (Hospital del Mar Medical Research Institute), Passeig CIBERESP), Madrid, Spain Marítim, 25‑29, 08003 Barcelona, Spain 2 10 University of Cantabria – IDIVAL, Santander, Spain Research Network On Health Services in Chronic Diseases 3 (REDISSEC), Barcelona, Spain Cancer and Environmental Epidemiology Unit, National 11 Center for Epidemiology, Carlos III Institute of Health, Oncology Data Analytics Program, Hospitalet de Llobregat, Madrid, Spain Catalan Institute of Oncology (ICO), Barcelona, Spain 4 12 Epidemiology Section, Department of Health, Public Health Colorectal Cancer Group, ONCOBELL Program, Bellvitge Division, Madrid, Spain Biomedical Research Institute (IDIBELL), Hospitalet de 5 Llobregat, Barcelona, Spain Public Health Division of Gipuzkoa, BioDonostia Research 13 Institute, San Sebastian, Spain Department of Clinical Sciences, Faculty of Medicine, 6 University of Barcelona, Barcelona, Spain Navarra Public Health Institute, Pamplona, Spain 14 7 Centro de Investigación en Recursos Naturales, Salud y IdiSNA, Navarra Institute for Health Research, Pamplona, Medio Ambiente (RENSMA), Universidad de Huelva, Spain Huelva, Spain 8 Instituto de Biomedicina (IBIOMED), Universidad de León, 15 Instituto de Investigación Sanitaria del Principado de León, Spain Asturias-ISPA, UNIOVI and CIBERESP, Oviedo, Spain 13 678 Breast Cancer Research and Treatment (2020) 181:667–678 16 20 Cancer and Public Health Area, FISABIO - Public Health, Centro de Investigación en Recursos Naturales, Salud y Valencia, Spain Medio Ambiente (RENSMA), Universidad de Huelva, 17 Huelva, Spain General Directorate of Public Health, Valencia, 21 Valencian Community, Spain ISGlobal, Barcelona, Spain 18 22 Epidemiology Unit and Girona Cancer Registry. Oncology IMIM (Hospital del Mar Medical Research Institute), Coordination Plan, Department of Health, Autonomous Barcelona, Spain Government of Catalonia, Catalan Institute of Oncology, 23 Universitat Pompeu Fabra (UPF), Barcelona, Spain Av. França, s/n, 17007 Girona, Spain 24 19 Servicio de Oncología, Complejo Asistencial Universitario Descriptive Epidemiology, Genetics and Cancer Prevention de León, León, Spain Group, Biomedical Research Institute (IDIBGI), 25 C/ Dr. Castany, s/n, 17190 Salt, Spain Facultad de Medicina, Universidad de Cantabria, Avda. Herrera Oria s/n, 39011 Santander, Spain 13