Clarification attempt of the mechanism of late recurrence by micro- and macro-analyses in estrogen receptor-positive breast cancer

Clarification attempt of the mechanism of late recurrence by micro- and macro-analyses in estrogen receptor-positive breast cancer | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Clarification attempt of the mechanism of late recurrence by micro- and macro-analyses in estrogen receptor-positive breast cancer Sae Kitano, Ryo Tsunashima, Chikage Kato, Akira Watanabe, Yoshiaki Sota, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-3389190/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 6 You are reading this latest preprint version Abstract Purpose The mechanism of late recurrence (LR) of estrogen receptor (ER)-positive breast cancer remains unclear. As prediction models for LR of ER-positive breast cancer, 42-gene classifier (42GC), which analyzes “micro-factors (gene expression patterns)” and the Clinical Treatment Score post-5 years (CTS5), which analyzes “macro-factors (clinicopathological factors)”, were developed; however, improving the accuracy of these models is desirable. We aimed to clarify the mechanism and develop a new prediction model by combining 42GC and CTS5. Methods We selected 2,454 patients with ER-positive breast cancer from public microarray databases. We performed recurrence prognostic analysis using 42GC and CTS5. Results In “the basic research” for recurrent patients (n = 347), the 42GC LR and CTS5 low-risk groups tended to have LR. In “the clinical research” for recurrence-free patients 5 years after surgery (n = 671), the 42GC LR and CTS5 high-risk group had a significantly higher LR rate after 5 years (16.9%) than the 42GC non-LR and CTS5 low-risk group (5.41%) ( p = 0.037). Conclusion In “the basic research,” we found that both micro-and macro-factors were associated with the mechanisms of early recurrence and LR. Meanwhile, in “the clinical research,” we found that the mechanistic tendency toward LR (the CTS5 low-risk group) differed from the high rate of LR (the CTS5 high-risk group). Therefore, differentiating between the biological mechanisms elucidated in “the basic research” and the decision-making process concerning extended hormonal therapy in “the clinical research” is necessary. These findings propose the development of a novel prediction model for LR. ER-positive breast cancer Late recurrence Extended hormonal therapy 42-gene classifier Clinical Treatment Score post-5 years Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 1. Introduction Although some patients with estrogen receptor (ER)-positive breast cancer develop late recurrence (LR) after completing 5 years of adjuvant endocrine therapy, the mechanism of LR has not yet been clarified, and no highly accurate prediction model for LR has been developed. Recently, the efficacy of extended hormonal therapy has been reported in several studies (ATLAS, aTTom, etc.) [ 1 – 7 ]. However, no clear criteria are available for determining which patients are suitable for extended hormonal therapy [ 8 – 11 ]. The Early Breast Cancer Trialists Collaborative Group conducted a meta-analysis of 88 trials, involving 62,923 patients with ER-positive breast cancer who were disease-free 5 years after scheduled hormonal therapy [ 12 ]. They reported that distant recurrence occurred at a steady rate throughout the study period from 5 to 20 years; however, the annual rate of distant recurrence remained low throughout that period. Therefore, extended hormonal therapy for several patients leads to wasteful and excessive medication. Therefore, identifying patients who benefit from extended hormonal therapy and those who can omit extended hormonal therapy is clinically significant. In this study, our hypothesis of early and LR mechanism is shown in Fig. 1. Our hypothesis is that if primary breast cancer cannot be completely surgically resected, only a small amount of cancer cells remains in the body just after surgery. These are gradually cultured in the body, and when they grow over the “detection line,” which can be detected by imaging tests, they become clinically recurrence. The X-axis is the time after surgery; the Y-axis is the amount of residual cancer cells after surgery in the body, and the slope angle of the red・blue lines corresponds to the cell proliferation potency. We believe that the growth rate of cancer cells during the disease is relatively constant. It is conceivable that the early recurrence group (the red line) has fast cell proliferation (large slope angle), whereas the LR group (the blue line) has slow cell proliferation (small slope angle). Additionally, it is conceivable that patients with a large amount of residual cancer cells after surgery, which is the high starting point on the Y-axis for cell proliferation, tend to develop early recurrence (the solid red–blue lines), whereas those with a small amount of residual cancer cells after surgery, which is the low starting point on the Y-axis, tend to develop LR (the dotted red–blue lines). Therefore, the timing of early or LR is determined by micro-factors (cell proliferation potency [slope angle of the line]) and macro-factors (residual cancer cells after surgery [Y-axis]). Thus, micro-factors can be detected by gene expression analysis using 42-gene classifier (42GC), and macro-factors are correlated with clinicopathological factors (tumor size ( T ), nodal status ( N ), etc.) and can be examined using the Clinical Treatment Score post-5 years (CTS5). Based on this hypothesis, we first started analyzing micro-factors in 2018. The original 42 gene probes related to the differences between the early and LR groups (differences in the slope angle of the red–blue lines in Fig. 1) were extracted by comprehensive gene expression analysis, as the micro-factors, and then, we have succeeded in developing a prediction model, 42GC, specific for LR [ 13 ]. However, 42GC does not include macro-factors (clinicopathological factors [ T , N , etc.] [Y-axis]). As a prediction model for LR by macro-factors, the usefulness of the CTS5 calculated from clinicopathological factors in the ATAC and BIG 1–98 trials has been reported [ 14 ]. The CTS5 was calculated from the following macro-factors: tumor size, nodal status, tumor grade, and age, to classify the risk of LR into three groups: low risk, intermediate risk, or high risk. It is a simple and convenient tool. Therefore, in this study, we aimed to prove the hypothesis of early and LR mechanisms in “the basic research” and to combine micro-factors (42GC) and macro-factors (CTS5) in “the clinical research.” Then, we attempted to develop a highly accurate prediction model for LR. In fact, using several public databases, we have shown that a highly accurate prediction model for LR can be developed by combining 42GC and CTS5. We would like the readers to read with attention the antonyms of “micro-factors” vs. “macro-factors” and “basic research” vs. “clinical research.” 2. Materials and Methods 2.1. Patients The patients of this study were extracted from 28 breast cancer public database cohorts as shown in Table 1. Their microarray CEL files were downloaded and used for the analysis. Among them, four cohorts (i.e., GSE6532, GSE12093, GSE17705, and GSE26971), which are the training sets for developing 42GC, were excluded. Finally, we used 24 cohorts (3,814 patients). Among them, 2,668 patients with ER-positive breast cancer were selected, and 2,454 patients of whom prognostic analysis was available were included in the analysis (Figs. 2a and 5a). Of the aforementioned 24 cohorts, 14 (i.e., GSE61304, GSE25066, GSE45255, E-TABM-158, GSE6532, GSE7390, GSE9195, GSE16391, GSE19615, GSE42658, GSE69031, GSE2990, GSE4922, and GSE234114) had complete information on clinicopathological factors to calculate the CTS5, totaling 2,206 patients. All of these were accompanied by their respective CEL files, and then, 42GC scores were also obtained for analysis. Among them, 1,346 patients with ER-positive breast cancer were selected, and 1,263 patients of whom prognostic analysis was available were included in the analysis (Figs. 2b and 5b). 2.2. Pathological examination The patients’ ER, PR, and HER2 status was determined immunohistochemically according to a previously described method [ 15 ]. For patients with unknown HER2 status, they were divided into the HER2-positive or HER2-negative group using the cutoff value of 4,800 (HER2 mRNA expression level in MAS5) as previously reported [ 16 ]. 2.3. Calculation of 42GC scores The method for calculating 42GC scores, a specialized prediction model for LR by multigene assay, was similar to the original paper [ 13 ]. 2.4. Calculation of the CTS5 The CTS5 was calculated using the original formula [ 14 ]. 2.5. Statistics Distant recurrence-free survival (DRFS) for the risk groups was analyzed using the Kaplan–Meier method and evaluated using log-rank tests. Associations between the prediction model risk groups and their clinicopathological parameters were evaluated using Fisher’s exact test. All statistical analyses were two-sided, and p -values < 0.05 were used to denote statistical significance. All statistical analyses were performed using R (version 4.1.3). 3. Results 3.1. “The basic research”: Proof of our hypothesis in Fig. 1 for early and LR mechanisms To classify recurrent patients according to their characteristics, we first examined recurrent patients only. 3.1.1. 42GC (micro-factors) Of the 2,454 patients with ER-positive breast cancer of whom prognostic analysis was available, 625 were included (Fig. 2a) and classified into the NLR and LR groups by 42GC. The NLR group comprised 299 patients (47.8%), whereas the LR group comprised 326 patients (52.2%), and the time of recurrence was later in the LR group than in the NLR group with a strongly significant difference ( p = 8.04e–06) (Fig. 3). 3.1.2. CTS5 (macro-factors) Of the 1,263 patients with ER-positive breast cancer, of whom the CTS5 was calculable and prognostic analysis was available, 347 were included (Fig. 2b), and the CTS5 was calculated (Fig. 4). According to the CTS5, patients were classified into the low-risk group (n = 95, 27.3%), the intermediate-risk group (n = 135, 38.9%), and the high-risk group (n = 117, 33.7%). The time of recurrence was in the following order: low-, intermediate-, and high-risk groups, with a strongly significant difference ( p < 2.2e–16) (Fig. 4). 3.2. “The clinical research”: Validation of our hypothesis in Fig. 1 in real-world data from several patients We examined the time of recurrence in real-world data from several patients in clinical practice with and without recurrence. 3.2.1. 42GC (micro-factors) The 2,454 patients with ER-positive breast cancer, with and without recurrence, were classified into the NLR and LR groups according to 42GC. The NLR group comprised 970 patients (39.5%), whereas the LR group comprised 1,484 patients (60.5%) (Fig. 5a). The patients in the LR group showed a significantly better prognosis than those in the NLR group ( p = 2.38e–09), and the time of recurrence in the LR group tended to be later, reflecting the mechanism by mediated the micro-factors in Fig. 1 (Fig. 5a). 3.2.2. CTS5 (macro-factors) We calculated the CTS5 of the 1,263 patients with ER-positive breast cancer with and without recurrence whose CTS5 was calculable (Figs. 2b and 5b). According to the CTS5, the patients were classified into the low-risk group (n = 411, 32.5%), intermediate-risk group (n = 541, 42.8%), and high-risk group (n = 311, 24.6%). The prognosis was in the following order: the low-, intermediate-, and high-risk groups, with a statistically significant difference ( p = 5.13e–10), and the time of recurrence in the low-risk group tended to be later, reflecting the mechanism mediated by the macro-factors in Fig. 1 (Fig. 5b). However, important findings in this study were that the rate of LR was the highest in the CTS5 high-risk group. Specifically, the findings indicated that the percentage of patients with recurrence in the CTS5 high-risk group (B in Fig. 5b) was higher than that in the CTS5 low-risk group (A in Fig. 5b) during the 5–10-year period of extended hormonal therapy (pink line in Fig. 5b). The findings in Fig. 5b are shown below in detail. A: In the CTS5 low-risk group, some patients had recurrence even after 10 years; therefore, the time of recurrence in the CTS5 low-risk group tended to be later in terms of the mechanism mediated by the macro-factors (Y-axis in Fig. 1); however, the 10-year DRFS rate (75.2%) decreased by 6.0% only compared with the 5-year DRFS rate (81.2%). B: In the CTS5 high-risk group, no patients had recurrence after 10 years; therefore, the time of recurrence in the CTS5 high-risk group tended to be earlier in terms of the mechanism mediated by the macro-factors (Y-axis in Fig. 1); however, the 10-year DRFS rate (51.1%) decreased by 13.2% compared with the 5-year DRFS rate (64.3%). Therefore, it is considered that the CTS5 high-risk group should be recommended for the extended hormonal therapy than the CTS5 low-risk group. 3.3. Analysis of distant recurrence-free patients 5 years after surgery Of the 1,263 patients with ER-positive breast cancer with and without recurrence whose 42GC and CTS5 were calculable, 671 were recurrence-free 5 years after surgery (Fig. 2b). Then, the 671 patients were classified according to 42GC and CTS5 (Table 2a). The CTS5 high-risk group and the 42GC LR group had a high recurrence rate of 16.9% 5 years or more after surgery, whereas the CTS5 low-risk group and the 42GC NLR group had a low recurrence rate of 5.41% 5 years or more after surgery; then, the former group had significantly more patients with LR than the latter group ( p = 0.037, hazard ratio [HR] = 3.58). The four groups in the middle of Table 2a—the CTS5 high-risk and 42GC NLR groups, the CTS5 intermediate-risk and 42GC LR groups, the CTS5 intermediate-risk and 42GC NLR groups, and the CTS5 low-risk and 42GC LR groups, had a recurrence rate of approximately 10% 5 years or more after surgery (blue letters in Table 2a). 3.4. Additional analysis according to 95-gene classifier (95GC) We attempted to classify these four middle groups according to 95GC [ 17 – 23 ]. We applied 95GC to the patients of these four groups and classified them into the high- and low-risk groups (Table 2b). As a result, by attaching the 95GC high-risk group and the top row (the CTS5 high-risk and 42GC LR groups) and attaching the 95GC low-risk group and the bottom row (the CTS5 low-risk and 42GC NLR groups), we finally classified the patients into two risk groups with late recurrence rates of 17.7% (n = 164) and 8.28% (n = 507) ( p = 0.0042; HR = 2.13) (Table 2b). 4. Discussion 4.1. The micro-factors: The development of 42GC and the new validation We first hypothesized the mechanisms of early and LR (Fig. 1) and proposed that both micro- and macro-factors are involved in recurrence. Then, in our previous report in 2018, we only analyzed micro-factors (gene expression patterns of primary tumors), corresponding to the slope angle of the lines (cell proliferation potency) in Fig. 1, and focused on the differences in gene expression between early and LR. Then, we developed, as a multigene assay, a prediction model specific for LR: 42GC [ 13 ]. 42GC classifies the high-risk group for late recurrence as the LR group and the low-risk group as the NLR group. Thus, in the validation set of 221 patients, those classified into the LR group by 42GC were likely to develop recurrence significantly later than those classified into the NLR group ( p = 0.020) [ 13 ]. In this study, as a new independent validation of 42GC, we analyzed the 625 patients with recurrence, including their respective CEL files in public databases that were apart from the training sets, and again showed that the LR group was likely to develop recurrence significantly later than the NLR group ( p = 8.04e–06) (Fig. 3). However, in Fig. 3, although the difference was strongly significant, some patients in the LR group had early recurrence and some patients in the NLR group had late recurrence. This suggests the need for a more accurate prediction model for LR by adding the analysis of macro-factors (clinicopathological factors)—CTS5. Additionally, because the 42GC LR group accounted for approximately half (52.2%) of the patients (Fig. 3), reducing the proportion of patients in the 42GC LR group was required to reduce the number of patients receiving extended hormonal therapy. 4.2. “The basic research”: Proof of our hypothesis in Fig. 1 for early and LR mechanisms Therefore, we first attempted to prove our hypothesis of early and LR mechanisms using not only micro-factors but also macro-factors in our hypothesis (Fig. 1). As mentioned above, Fig. 3 shows that the LR group with a smaller slope angle was more likely to develop recurrence significantly later than the NLR group according to 42GC (micro-factors). Figure 4 shows that the CTS5 low-risk group, which was assumed to have a little amount of residual cancer cells after surgery, was more likely to develop recurrence significantly later according to the CTS5 (macro-factors). These results confirmed our hypothesis of early and LR mechanisms and demonstrated that both micro- and macro-factors affected the time of recurrence. 4.3. “The clinical research”: Development of a prediction model specific for LR Next, to develop a prediction model specific for LR, applying the aforementioned findings to clinical practice, we conducted a study involving patients with and without recurrence. In this study, we analyzed 671 patients without recurrence 5 years after surgery (approximately 10% of them have recurred afterward, and approximately 90% have not recurred until the end) by combining 42GC (micro-factors) and CTS5 (macro-factors) (Table 2a). Surprisingly, in this situation where most patients did not have recurrence, the LR rate was found to be the highest in the CTS5 high-risk group, opposing the results of the analysis based on our mechanism (Fig. 4), which showed that the CTS5 low-risk group was more likely to develop recurrence later. Therefore, the CTS5 low-risk group is likely to develop LR, based on the biological mechanism; however, the CTS5 low-risk group has a much lower T , N factors during surgery, and then, the recurrence rate itself in clinical practice is quite low and clinically not worthy of attention. Therefore, distinguishing between the biological mechanism in “the basic research” and the decision-making for extended hormonal therapy in “the clinical research” is necessary. Based on this, the discussion on recurrence rates in the real world, where many patients do not have recurrence, appears to be more complex than the discussion on the biological mechanism. In the investigation of the six-group classification (Table 2a) for the 671 patients without recurrence 5 years after surgery (Fig. 2b), the late recurrence rate was 16.9% in the CTS5 high-risk and 42GC LR group combination, whereas it was 5.41% in the CTS5 low-risk and 42GC NLR group combination. The risk difference between the two groups was more than threefold, indicating that the former combination was a better indication for extended hormonal therapy than the latter. This is the main content of this study. 4.4. Additional analysis according to 95GC The problem we confronted here was that the middle four groups (blue letters in Table 2a) had LR rates of approximately 10% (Table 2a). In the middle four groups, most (approximately 90%) had no LR and thus can be considered able to omit extended hormonal therapy. However, as an additional analysis, we attempted to further classify these four groups according to 95GC, a powerful recurrence risk classification method [ 17 – 23 ]. We have previously reported the results of several studies that have significantly classified the intermediate-risk group by OncotypeDX into the two groups by 95GC [ 17 – 23 ]. We newly classified the four groups into two groups according to 95GC: high and low. As a result, by attaching the 95GC high-risk group and the top row (the CTS5 high-risk and 42GC LR groups) and attaching the 95GC low-risk group and the bottom row (the CTS5 low-risk and 42GC NLR groups), we finally classified the four groups into the two risk groups with late recurrence rates of 17.7% (n = 164) and 8.28% (n = 507), respectively ( p = 0.0042; HR = 2.13) (Table 2b). Consequently, we could complete a new algorithm in which only approximately one-fourth of patients with ER-positive breast cancer received extended hormonal therapy and the remaining three-fourths did not receive extended hormonal therapy. Thus, the actual clinical practice is quite complex, and the results led to the conclusion that three methods of analysis—42GC, CTS5, and 95GC—were required to determine the candidates for extended hormonal therapy. Further validation with more patients is required. 5. Conclusion Herein, we have successfully validated our hypothesis regarding early recurrence and LR mechanisms using 42GC and CTS5. We attempted to apply these findings to clinical data and have developed a novel algorithm that recommends extended hormonal therapy for approximately one-fourth of ER-positive patients with breast cancer. This study demonstrated that both micro-and macro-factors affected the time of recurrence. Additionally, there was a difference between a tendency toward LR based on biological mechanisms (the CTS5 low-risk group) and a higher LR rate (the CTS5 high-risk group). It would be beneficial to confirm our findings with a larger patient cohort in the future. Abbreviations ER, estorogen receptor; PR, progesterone receptor; HER2, human epidermal growth factor receptor 2; 42GC, 42-gene classifier; LR, late recurrence; NLR, non-late recurrence; CTS5, Clinical Treatment Score post-5 years; 95GC, 95-gene classifier; T , tumor size; N , nodal status; 5-year DRFS, 5-year distant recurrence-free survival; HR, hazard ratio Declarations Authors’ contributions All authors contributed to the study conception and design. Conceptualization: Ryo Tsunashima and Yasuto Naoi ; Methodology: Ryo Tsunashima ; Formal analysis: Sae Kitano and Saya Matsumoto ; Writing - original draft: Sae Kitano ; Data curation: Ryo Tsunashima , Yoshiaki Sota , and Akira Watanabe ; Visualization: Sae Kitano ; Supervision: Yasuto Naoi , Koichi Sakaguchi , Midori Morita , and Chikage Kato ; Project administration and Writing - review & editing: Yasuto Naoi . All authors have read and approved the final manuscript. Data availability The datasets analyzed during the current study are not publicly available due to the study protocol but are available from the corresponding author on reasonable request. Ethics approval This study was performed in accordance with the principles of the Declaration of Helsinki and has been approved by the ethics committee of our institution. Consent to participate Informed consent was provided by the patients to participate in the study. Consent to publish Not applicable. Declaration of competing interest Yasuto Naoi has received research funding from Sysmex, ONO, Daiichi-Sankyo and AstraZeneca, and honoraria from AstraZeneca, Pfizer, Eli Lilly, ONO, Daiichi-Sankyo and Chugai outside the submitted work; he holds joint patents with Sysmex including Curebest™ 95GC Breast (JP.5725274.B2) and has received patent royalties outside the submitted work. Funding The authors declare that no funds, grants, or other support were received during the preparation of this manuscript. Acknowledgements We thank Professor Kenzo Shimazu from the Department of Breast and Endocrine Surgery, Osaka University Graduate School of Medicine, Osaka, Japan for contributing to this study. References Davies C, Pan H, Godwin J, Gray R, Arriagada R, Raina V, et al (2013) Long-term effects of continuing adjuvant tamoxifen to 10 years versus stopping at 5 years after diagnosis of oestrogen receptor-positive breast cancer: ATLAS, a randomised trial. Lancet 381(9869):805-816. https://doi.org/10.1016/S0140-6736(12)61963-1 Gray RG, Rea D, Handley K, Bowden SJ, Perry P, Earl HM, et al (2013) aTTom：long-term effects of continuing adjuvant tamoxifen to 10 years versus stopping at 5 years in 6,953 women with early breast cancer. J Clin Oncol 31(18):5. Schiavon G, Smith IE (2014) Status of adjuvant endocrine therapy for breast cancer. Breast Cancer Res 16(2):206. https://doi.org/10.1186/bcr3636 Goss PE, Ingle JN, Martino S, Robert NJ, Muss HB, Piccart MJ, et al (2005) Randomized trial of letrozole following tamoxifen as extended adjuvant therapy in receptor-positive breast cancer: Updated findings from NCIC CTG MA.17. J Natl Cancer Inst 97(17):1262-1271. https://doi.org/10.1093/jnci/dji250 Goss PE, Ingle JN, Pritchard KI, Robert NJ, Muss H, Gralow J, et al (2016) Extending aromatase-inhibitor adjuvant therapy to 10 years. N Engl J Med 375(3):209-219. https://doi.org/10.1056/NEJMoa1604700 The 2022 edition of the Clinical Practice Guidelines for the Japanese Breast Cancer Society, Treatment ed, Drug therapy in Clinical Question 4 Iwase T, Saji S, Iijima K, Higaki K, Ohtani S, Sato Y, et al (2023) Postoperative adjuvant anastrozole for 10 or 5 years in patients with hormone receptor-positive breast cancer: AERAS, a randomized multicenter open-label phase III trial. J Clin Oncol 41(18):3329-3338. https://doi.org/10.1200/JCO.22.00577. Yamashita H, Ogiya A, Shien T, Horimoto Y, Masuda N, Inao T, et al (2016) Clinicopathological factors predicting early and late distant recurrence in estrogen receptor-positive, HER2-negative breast cancer. Breast Cancer 23(6):830-843. https://doi.org/10.1007/s12282-015-0649-0 Kennecke HF, Olivotto IA, Speers C, Norris B, Chia SK, Bryce C, Gelmon KA (2007) Late risk of relapse and mortality among postmenopausal women with estrogen responsive early breast cancer after 5 years of tamoxifen. Ann Oncol 18(1):45-51. https://doi.org/10.1093/annonc/mdl334 Sestak I, Dowsett M, Zabaglo L, Lopez-Knowles E, Ferree S, Cowens JW, Cuzick J (2013) Factors predicting late recurrence for estrogen receptor-positive breast cancer. J Natl Cancer Inst 105(19):1504-1511. https://doi.org/10.1093/jnci/djt244 Sestak I, Cuzick J (2015) Markers for the identification of late breast cancer recurrence. Breast Cancer Res 17:10. https://doi.org/10.1186/s13058-015-0516-0 Pan H, Gray R, Braybrooke J, Davies C, Taylor C, McGale P, et al (2017) 20-year risks of breast-cancer recurrence after stopping endocrine therapy at 5 years. N Engl J Med 377(19):1836-1846. https://doi.org/10.1056/nejmc1715968 Tsunashima R, Naoi Y, Shimazu K, Kagara N, Shimoda M, Tanei T, Miyake T, Kim SJ, Noguchi S, et al (2018) Construction of a novel multi-gene assay (42-gene classifier) for prediction of late recurrence in ER-positive breast cancer patients. Breast Cancer Res Treat 171(1):33-41. https://doi.org/10.1007/s10549-018-4812-0 Dowsett M, Sestak I, Regan MM, Dodson A, Viale G, Thürlimann B, Colleoni M, Cuzick J (2018) Integration of clinical variables for the prediction of late distant recurrence in patients with estrogen receptor–positive breast cancer treated with 5 years of endocrine therapy: CTS5. J Clin Oncol 36(19): 1941-1948. https://doi.org/10.1200/JCO.2017.76.4258 Naoi Y, Kishi K, Tanei T, Tsunashima R, Tominaga N, Baba Y, Kim SJ, Taguchi T, Tamaki Y, Noguchi S (2011) Development of 95-gene classifier as a powerful predictor of recurrences in node-negative and ER-positive breast cancer patients. Breast Cancer Res Treat 128(3):633-641. https://doi.org/10.1007/s10549-010-1145-z Győrffy B, Benke Z, Lánczky A, Balázs B, Szállási Z, Timár J, Schäfer R (2012) RecurrenceOnline: An online analysis tool to determine breast cancer recurrence and hormone receptor status using microarray data. Breast Cancer Res Treat 132(3):1025-1034. https://doi.org/10.1007/s10549-011-1676-y Naoi Y, Noguchi S (2016) Multi-gene classifiers for prediction of recurrence in breast cancer patients. Breast Cancer 23(1):12-18. https://doi.org/10.1007/s12282-015-0596-9. Naoi Y, Saito Y, Kishi K, Shimoda M, Kagara N, Miyake T, Tanei T, Shimazu K, Kim SJ, Noguchi S (2019) Development of recurrence risk score using 95‑gene classifier and its application to formalin‑fixed paraffin‑embedded tissues in ER‑positive, HER2‑negative and node‑negative breast cancer. Oncol Rep 42(6):2680-2685. https://doi.org/10.3892/or.2019.7358. Naoi Y, Tsunashima R, Shimazu K, Noguchi S (2021) The multigene classifiers 95GC/42GC/155GC for precision medicine in ER-positive HER2-negative early breast cancer. Cancer Sci 112(4):1369-1375. https://doi.org/10.1111/cas.14838. Fujii T, Masuda H, Cheng YC, Yang F, Sahin AA, Naoi Y, et al (2021) A 95-gene signature stratifies recurrence risk of invasive disease in ER-positive, HER2-negative, node-negative breast cancer with intermediate 21-gene signature recurrence scores. Breast Cancer Res Treat 189(2):455-461. https://doi.org/10.1007/s10549-021-06276-7. Tsukamoto F, Arihiro K, Takahashi M, Ito KI, Ohsumi S, Takashima S, Oba T, Yoshida M, Kishi K, Yamagishi K, Kinoshita T (2021) Multicenter retrospective study on the use of Curebest™ 95GC breast for estrogen receptor-positive and node-negative early breast cancer. BMC Cancer 21(1):1077. https://doi.org/10.1186/s12885-021-08778-5. Yamashita H, Hatanaka KC, Yamagishi K, Saito Y, Hamasaki K, Taniguchi M, Okumura A, Nange A, Matsuno Y, Hatanaka Y (2023) Evaluation of 95-gene classifier of formalin-fixed paraffin-embedded tissues in ER-positive, HER2-negative, and node-negative breast cancer. Anticancer Res 43(2):707-711. https://doi.org/0.21873/anticanres.16209. Naoi Y, Tsunashima R, Shimazu K, Oikawa M, Imanishi S, Koyama H, et al (2023) Validation of the prognosis of patients with ER‑positive, HER2‑negative and node‑negative invasive breast cancer classified as low risk by Curebest™ 95GC breast in a multi‑institutional registry study. Oncol Lett 25(5):209. https://doi.org/10.3892/ol.2023.13794. Tables Table 1 and 2 are available in the Supplementary Files section. Supplementary Files Table1.xlsx Table 1. Patient characteristics Table2.pptx Table 2a. Risk classification Table for late recurrence by combining 42GC and CTS5 Table 2b. Risk classification Table for late recurrence by combining, 42GC, CTS5, and 95GC Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Reject, reconsider with recommended revisions 06 Nov, 2023 Reviewers agreed at journal 22 Oct, 2023 Reviewers invited by journal 22 Oct, 2023 Editor invited by journal 20 Oct, 2023 Editor assigned by journal 13 Oct, 2023 First submitted to journal 12 Oct, 2023 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-3389190","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":242044443,"identity":"a0eb4efa-681e-4531-89aa-c38549cf0963","order_by":0,"name":"Sae Kitano","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABAklEQVRIiWNgGAWjYDACHsYGECUnAeYdgIuz4dPSCNJjLMHATLQWBrA1iTPQtOAG8j2H2x/83GGTPrO9/+ADhjN3EhvEDjB++MHAl4dLi8HZxsbG3jNpubN5DjMbMNx4ltggncAs2cPAVoxTCz/QL7xth3PnSSSzSTB8OJy4/3YCgzTQL4kNuBzWz9jY+LftcLocTAvIlt/4tDAAHdYMtCVBGqzlBlgLG15bDM4cbJwt25ZmOLPnsLFBwplnxg3SiW2WPQa4/SLfk/7g49s2G3mJ440PH3w4dke2QTr58I0fFcdwhhgqSABHDCiiDI4lEKcFKS5riNYyCkbBKBgFwx4AAH74Wqmgu6K0AAAAAElFTkSuQmCC","orcid":"https://orcid.org/0009-0008-6896-3088","institution":"Kyoto Prefectural University of Medicine: Kyoto Furitsu Ika Daigaku","correspondingAuthor":true,"submittingAuthor":false,"prefix":"","firstName":"Sae","middleName":"","lastName":"Kitano","suffix":""},{"id":242044444,"identity":"527ea988-95af-451f-a3f7-6ff21ab8b784","order_by":1,"name":"Ryo Tsunashima","email":"","orcid":"https://orcid.org/0000-0003-2663-0142","institution":"Rinku General Medical Center: Rinku Sogo Iryo Center","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Ryo","middleName":"","lastName":"Tsunashima","suffix":""},{"id":242044445,"identity":"e6722ba2-3bad-4e9a-8c85-20c10b598505","order_by":2,"name":"Chikage Kato","email":"","orcid":"","institution":"Kyoto Prefectural University of Medicine: Kyoto Furitsu Ika Daigaku","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Chikage","middleName":"","lastName":"Kato","suffix":""},{"id":242044446,"identity":"770e7278-80fb-4029-aa46-1d9c67365f41","order_by":3,"name":"Akira Watanabe","email":"","orcid":"","institution":"Kyoto Prefectural University of Medicine: Kyoto Furitsu Ika Daigaku","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Akira","middleName":"","lastName":"Watanabe","suffix":""},{"id":242044447,"identity":"eb8b75d5-8ee5-4fe6-a7fe-7bb1ecf34a69","order_by":4,"name":"Yoshiaki Sota","email":"","orcid":"","institution":"Osaka University School of Medicine Graduate School of Medicine: Osaka Daigaku Daigakuin Igakukei Kenkyuka Igakubu","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Yoshiaki","middleName":"","lastName":"Sota","suffix":""},{"id":242044448,"identity":"d9856af5-b983-450c-9b72-5c220266c8d2","order_by":5,"name":"Saya Matsumoto","email":"","orcid":"","institution":"Kyoto Prefectural University of Medicine: Kyoto Furitsu Ika Daigaku","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Saya","middleName":"","lastName":"Matsumoto","suffix":""},{"id":242044449,"identity":"e03fcc50-d9e4-4231-9c36-f255262f427f","order_by":6,"name":"Midori Morita","email":"","orcid":"","institution":"Kyoto Prefectural University of Medicine: Kyoto Furitsu Ika Daigaku","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Midori","middleName":"","lastName":"Morita","suffix":""},{"id":242044450,"identity":"f20e2be5-c086-4b5e-a6c8-9a6d34740eea","order_by":7,"name":"Koichi Sakaguchi","email":"","orcid":"","institution":"Kyoto Prefectural University of Medicine: Kyoto Furitsu Ika Daigaku","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Koichi","middleName":"","lastName":"Sakaguchi","suffix":""},{"id":242044451,"identity":"d0a84bde-a445-47c3-8b11-a7f6e89e1ac1","order_by":8,"name":"Yasuto Naoi","email":"","orcid":"","institution":"Kyoto Prefectural University of Medicine: Kyoto Furitsu Ika Daigaku","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Yasuto","middleName":"","lastName":"Naoi","suffix":""}],"badges":[],"createdAt":"2023-09-26 14:38:25","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-3389190/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-3389190/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":45339925,"identity":"24a2f44f-902a-4609-adfc-6a0543d76fe8","added_by":"auto","created_at":"2023-10-27 20:12:52","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":37719,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eOur hypothesis of early and late recurrence mechanisms.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe X-axis is the time after surgery, the slope angle of the red–blue lines corresponds to the cell proliferation potency (micro-factors), and the Y-axis is the amount of residual cancer cells after surgery in the body (macro-factors).\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-3389190/v1/d6cff0dddf0131ae27d97fab.png"},{"id":45339928,"identity":"471cc44b-7a52-4b26-8595-0bc268d35e9d","added_by":"auto","created_at":"2023-10-27 20:12:52","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":92409,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eDiagram of patient selection. 42GC (a) and CTS5 (b).\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-3389190/v1/b1c3624cad1b683d6f4f9bb7.png"},{"id":45339922,"identity":"c9ae9a1a-9a24-47a1-a4ae-62ce56a393cf","added_by":"auto","created_at":"2023-10-27 20:12:51","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":68183,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eClassification of the time of recurrence according to 42GC only for patients with recurrence.\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-3389190/v1/4c85b49616459f56fa8ba876.png"},{"id":45339924,"identity":"06f207d5-accd-49ed-a027-c93a9288713c","added_by":"auto","created_at":"2023-10-27 20:12:52","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":70096,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eClassification of the time of recurrence according to the CTS5 only for patients with recurrence.\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-3389190/v1/0a76cf2a82d80097596f17d5.png"},{"id":45339927,"identity":"1909a285-c7e1-4d07-bd8a-ef3fb508664d","added_by":"auto","created_at":"2023-10-27 20:12:52","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":80298,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eKaplan–Meier survival curves of distant recurrence-free survival (DRFS) rates for patients with and without recurrence according to 42GC (a) and the CTS5 (b).\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-3389190/v1/87c1f539332c8e0b4eeb642b.png"},{"id":45340586,"identity":"9d9cec36-8e23-4fb4-9b0f-baf5db624995","added_by":"auto","created_at":"2023-10-27 20:20:52","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1020195,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3389190/v1/8cb39798-9604-49f2-83a1-eb012ea5e224.pdf"},{"id":45339921,"identity":"5c54af82-fff7-4d9f-b5d6-0c03d23be281","added_by":"auto","created_at":"2023-10-27 20:12:51","extension":"xlsx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":16446,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eTable 1. Patient characteristics\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"Table1.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-3389190/v1/8e7f9c11b78e70baa300e38d.xlsx"},{"id":45339923,"identity":"b5e6327d-5eeb-4b3d-9e70-1624e8c9d4a3","added_by":"auto","created_at":"2023-10-27 20:12:51","extension":"pptx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":56403,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eTable 2a. Risk classification Table for late recurrence by combining 42GC and CTS5\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 2b. Risk classification Table for late recurrence by combining, 42GC, CTS5, and 95GC\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"Table2.pptx","url":"https://assets-eu.researchsquare.com/files/rs-3389190/v1/93ca362e41a417a1da1228bd.pptx"}],"financialInterests":"","formattedTitle":"Clarification attempt of the mechanism of late recurrence by micro- and macro-analyses in estrogen receptor-positive breast cancer","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eAlthough some patients with estrogen receptor (ER)-positive breast cancer develop late recurrence (LR) after completing 5 years of adjuvant endocrine therapy, the mechanism of LR has not yet been clarified, and no highly accurate prediction model for LR has been developed.\u003c/p\u003e \u003cp\u003eRecently, the efficacy of extended hormonal therapy has been reported in several studies (ATLAS, aTTom, etc.) [\u003cspan additionalcitationids=\"CR2 CR3 CR4 CR5 CR6\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. However, no clear criteria are available for determining which patients are suitable for extended hormonal therapy [\u003cspan additionalcitationids=\"CR9 CR10\" citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThe Early Breast Cancer Trialists Collaborative Group conducted a meta-analysis of 88 trials, involving 62,923 patients with ER-positive breast cancer who were disease-free 5 years after scheduled hormonal therapy [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. They reported that distant recurrence occurred at a steady rate throughout the study period from 5 to 20 years; however, the annual rate of distant recurrence remained low throughout that period. Therefore, extended hormonal therapy for several patients leads to wasteful and excessive medication. Therefore, identifying patients who benefit from extended hormonal therapy and those who can omit extended hormonal therapy is clinically significant.\u003c/p\u003e \u003cp\u003eIn this study, our hypothesis of early and LR mechanism is shown in Fig.\u0026nbsp;1.\u003c/p\u003e \u003cp\u003eOur hypothesis is that if primary breast cancer cannot be completely surgically resected, only a small amount of cancer cells remains in the body just after surgery. These are gradually cultured in the body, and when they grow over the \u0026ldquo;detection line,\u0026rdquo; which can be detected by imaging tests, they become clinically recurrence. The X-axis is the time after surgery; the Y-axis is the amount of residual cancer cells after surgery in the body, and the slope angle of the red・blue lines corresponds to the cell proliferation potency. We believe that the growth rate of cancer cells during the disease is relatively constant.\u003c/p\u003e \u003cp\u003eIt is conceivable that the early recurrence group (the red line) has fast cell proliferation (large slope angle), whereas the LR group (the blue line) has slow cell proliferation (small slope angle). Additionally, it is conceivable that patients with a large amount of residual cancer cells after surgery, which is the high starting point on the Y-axis for cell proliferation, tend to develop early recurrence (the solid red\u0026ndash;blue lines), whereas those with a small amount of residual cancer cells after surgery, which is the low starting point on the Y-axis, tend to develop LR (the dotted red\u0026ndash;blue lines). Therefore, the timing of early or LR is determined by micro-factors (cell proliferation potency [slope angle of the line]) and macro-factors (residual cancer cells after surgery [Y-axis]). Thus, micro-factors can be detected by gene expression analysis using 42-gene classifier (42GC), and macro-factors are correlated with clinicopathological factors (tumor size (\u003cem\u003eT\u003c/em\u003e), nodal status (\u003cem\u003eN\u003c/em\u003e), etc.) and can be examined using the Clinical Treatment Score post-5 years (CTS5).\u003c/p\u003e \u003cp\u003eBased on this hypothesis, we first started analyzing micro-factors in 2018. The original 42 gene probes related to the differences between the early and LR groups (differences in the slope angle of the red\u0026ndash;blue lines in Fig.\u0026nbsp;1) were extracted by comprehensive gene expression analysis, as the micro-factors, and then, we have succeeded in developing a prediction model, 42GC, specific for LR [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. However, 42GC does not include macro-factors (clinicopathological factors [\u003cem\u003eT\u003c/em\u003e, \u003cem\u003eN\u003c/em\u003e, etc.] [Y-axis]).\u003c/p\u003e \u003cp\u003eAs a prediction model for LR by macro-factors, the usefulness of the CTS5 calculated from clinicopathological factors in the ATAC and BIG 1\u0026ndash;98 trials has been reported [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThe CTS5 was calculated from the following macro-factors: tumor size, nodal status, tumor grade, and age, to classify the risk of LR into three groups: low risk, intermediate risk, or high risk. It is a simple and convenient tool.\u003c/p\u003e \u003cp\u003eTherefore, in this study, we aimed to prove the hypothesis of early and LR mechanisms in \u0026ldquo;the basic research\u0026rdquo; and to combine micro-factors (42GC) and macro-factors (CTS5) in \u0026ldquo;the clinical research.\u0026rdquo; Then, we attempted to develop a highly accurate prediction model for LR. In fact, using several public databases, we have shown that a highly accurate prediction model for LR can be developed by combining 42GC and CTS5.\u003c/p\u003e \u003cp\u003eWe would like the readers to read with attention the antonyms of \u0026ldquo;micro-factors\u0026rdquo; \u003cem\u003evs.\u003c/em\u003e \u0026ldquo;macro-factors\u0026rdquo; and \u0026ldquo;basic research\u0026rdquo; \u003cem\u003evs.\u003c/em\u003e \u0026ldquo;clinical research.\u0026rdquo;\u003c/p\u003e"},{"header":"2. Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1. Patients\u003c/h2\u003e \u003cp\u003eThe patients of this study were extracted from 28 breast cancer public database cohorts as shown in Table\u0026nbsp;1.\u003c/p\u003e \u003cp\u003eTheir microarray CEL files were downloaded and used for the analysis. Among them, four cohorts (i.e., GSE6532, GSE12093, GSE17705, and GSE26971), which are the training sets for developing 42GC, were excluded. Finally, we used 24 cohorts (3,814 patients). Among them, 2,668 patients with ER-positive breast cancer were selected, and 2,454 patients of whom prognostic analysis was available were included in the analysis (Figs.\u0026nbsp;2a and 5a).\u003c/p\u003e \u003cp\u003eOf the aforementioned 24 cohorts, 14 (i.e., GSE61304, GSE25066, GSE45255, E-TABM-158, GSE6532, GSE7390, GSE9195, GSE16391, GSE19615, GSE42658, GSE69031, GSE2990, GSE4922, and GSE234114) had complete information on clinicopathological factors to calculate the CTS5, totaling 2,206 patients. All of these were accompanied by their respective CEL files, and then, 42GC scores were also obtained for analysis. Among them, 1,346 patients with ER-positive breast cancer were selected, and 1,263 patients of whom prognostic analysis was available were included in the analysis (Figs.\u0026nbsp;2b and 5b).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2. Pathological examination\u003c/h2\u003e \u003cp\u003eThe patients\u0026rsquo; ER, PR, and HER2 status was determined immunohistochemically according to a previously described method [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. For patients with unknown HER2 status, they were divided into the HER2-positive or HER2-negative group using the cutoff value of 4,800 (HER2 mRNA expression level in MAS5) as previously reported [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e].\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3. Calculation of 42GC scores\u003c/h2\u003e \u003cp\u003eThe method for calculating 42GC scores, a specialized prediction model for LR by multigene assay, was similar to the original paper [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e].\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e\u003cb\u003e2.4. Calculation of the CTS5\u003c/b\u003e\u003c/h2\u003e \u003cp\u003eThe CTS5 was calculated using the original formula [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e].\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e2.5. Statistics\u003c/h2\u003e \u003cp\u003eDistant recurrence-free survival (DRFS) for the risk groups was analyzed using the Kaplan\u0026ndash;Meier method and evaluated using log-rank tests. Associations between the prediction model risk groups and their clinicopathological parameters were evaluated using Fisher\u0026rsquo;s exact test. All statistical analyses were two-sided, and \u003cem\u003ep\u003c/em\u003e-values\u0026thinsp;\u0026lt;\u0026thinsp;0.05 were used to denote statistical significance. All statistical analyses were performed using R (version 4.1.3).\u003c/p\u003e \u003c/div\u003e"},{"header":"3. Results","content":"\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003e3.1. \u0026ldquo;The basic research\u0026rdquo;: Proof of our hypothesis in Fig.\u0026nbsp;1 for early and LR mechanisms\u003c/h2\u003e \u003cp\u003eTo classify recurrent patients according to their characteristics, we first examined recurrent patients only.\u003c/p\u003e \u003cdiv id=\"Sec10\" class=\"Section3\"\u003e \u003ch2\u003e3.1.1. 42GC (micro-factors)\u003c/h2\u003e \u003cp\u003eOf the 2,454 patients with ER-positive breast cancer of whom prognostic analysis was available, 625 were included (Fig.\u0026nbsp;2a) and classified into the NLR and LR groups by 42GC.\u003c/p\u003e \u003cp\u003eThe NLR group comprised 299 patients (47.8%), whereas the LR group comprised 326 patients (52.2%), and the time of recurrence was later in the LR group than in the NLR group with a strongly significant difference (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;8.04e\u0026ndash;06) (Fig.\u0026nbsp;3).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section3\"\u003e \u003ch2\u003e3.1.2. CTS5 (macro-factors)\u003c/h2\u003e \u003cp\u003eOf the 1,263 patients with ER-positive breast cancer, of whom the CTS5 was calculable and prognostic analysis was available, 347 were included (Fig.\u0026nbsp;2b), and the CTS5 was calculated (Fig.\u0026nbsp;4). According to the CTS5, patients were classified into the low-risk group (n\u0026thinsp;=\u0026thinsp;95, 27.3%), the intermediate-risk group (n\u0026thinsp;=\u0026thinsp;135, 38.9%), and the high-risk group (n\u0026thinsp;=\u0026thinsp;117, 33.7%).\u003c/p\u003e \u003cp\u003eThe time of recurrence was in the following order: low-, intermediate-, and high-risk groups, with a strongly significant difference (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;2.2e\u0026ndash;16) (Fig.\u0026nbsp;4).\u003c/p\u003e \u003cp\u003e \u003cb\u003e3.2. \u0026ldquo;The clinical research\u0026rdquo;: Validation of our hypothesis in Fig.\u0026nbsp;1 in real-world data from several patients\u003c/b\u003e \u003c/p\u003e \u003cp\u003eWe examined the time of recurrence in real-world data from several patients in clinical practice with and without recurrence.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section3\"\u003e \u003ch2\u003e3.2.1. 42GC (micro-factors)\u003c/h2\u003e \u003cp\u003eThe 2,454 patients with ER-positive breast cancer, with and without recurrence, were classified into the NLR and LR groups according to 42GC. The NLR group comprised 970 patients (39.5%), whereas the LR group comprised 1,484 patients (60.5%) (Fig.\u0026nbsp;5a). The patients in the LR group showed a significantly better prognosis than those in the NLR group (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;2.38e\u0026ndash;09), and the time of recurrence in the LR group tended to be later, reflecting the mechanism by mediated the micro-factors in Fig.\u0026nbsp;1 (Fig.\u0026nbsp;5a).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section3\"\u003e \u003ch2\u003e3.2.2. CTS5 (macro-factors)\u003c/h2\u003e \u003cp\u003eWe calculated the CTS5 of the 1,263 patients with ER-positive breast cancer with and without recurrence whose CTS5 was calculable (Figs.\u0026nbsp;2b and 5b).\u003c/p\u003e \u003cp\u003eAccording to the CTS5, the patients were classified into the low-risk group (n\u0026thinsp;=\u0026thinsp;411, 32.5%), intermediate-risk group (n\u0026thinsp;=\u0026thinsp;541, 42.8%), and high-risk group (n\u0026thinsp;=\u0026thinsp;311, 24.6%).\u003c/p\u003e \u003cp\u003eThe prognosis was in the following order: the low-, intermediate-, and high-risk groups, with a statistically significant difference (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;5.13e\u0026ndash;10), and the time of recurrence in the low-risk group tended to be later, reflecting the mechanism mediated by the macro-factors in Fig.\u0026nbsp;1 (Fig.\u0026nbsp;5b).\u003c/p\u003e \u003cp\u003eHowever, important findings in this study were that the rate of LR was the highest in the CTS5 high-risk group. Specifically, the findings indicated that the percentage of patients with recurrence in the CTS5 high-risk group (B in Fig.\u0026nbsp;5b) was higher than that in the CTS5 low-risk group (A in Fig.\u0026nbsp;5b) during the 5\u0026ndash;10-year period of extended hormonal therapy (pink line in Fig.\u0026nbsp;5b).\u003c/p\u003e \u003cp\u003eThe findings in Fig.\u0026nbsp;5b are shown below in detail.\u003c/p\u003e \u003cp\u003eA: In the CTS5 low-risk group, some patients had recurrence even after 10 years; therefore, the time of recurrence in the CTS5 low-risk group tended to be later in terms of the mechanism mediated by the macro-factors (Y-axis in Fig.\u0026nbsp;1); however, the 10-year DRFS rate (75.2%) decreased by 6.0% only compared with the 5-year DRFS rate (81.2%).\u003c/p\u003e \u003cp\u003eB: In the CTS5 high-risk group, no patients had recurrence after 10 years; therefore, the time of recurrence in the CTS5 high-risk group tended to be earlier in terms of the mechanism mediated by the macro-factors (Y-axis in Fig.\u0026nbsp;1); however, the 10-year DRFS rate (51.1%) decreased by 13.2% compared with the 5-year DRFS rate (64.3%).\u003c/p\u003e \u003cp\u003eTherefore, it is considered that the CTS5 high-risk group should be recommended for the extended hormonal therapy than the CTS5 low-risk group.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003e3.3. Analysis of distant recurrence-free patients 5 years after surgery\u003c/h2\u003e \u003cp\u003eOf the 1,263 patients with ER-positive breast cancer with and without recurrence whose 42GC and CTS5 were calculable, 671 were recurrence-free 5 years after surgery (Fig.\u0026nbsp;2b).\u003c/p\u003e \u003cp\u003eThen, the 671 patients were classified according to 42GC and CTS5 (Table\u0026nbsp;2a).\u003c/p\u003e \u003cp\u003eThe CTS5 high-risk group and the 42GC LR group had a high recurrence rate of 16.9% 5 years or more after surgery, whereas the CTS5 low-risk group and the 42GC NLR group had a low recurrence rate of 5.41% 5 years or more after surgery; then, the former group had significantly more patients with LR than the latter group (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.037, hazard ratio [HR]\u0026thinsp;=\u0026thinsp;3.58).\u003c/p\u003e \u003cp\u003eThe four groups in the middle of Table\u0026nbsp;2a\u0026mdash;the CTS5 high-risk and 42GC NLR groups, the CTS5 intermediate-risk and 42GC LR groups, the CTS5 intermediate-risk and 42GC NLR groups, and the CTS5 low-risk and 42GC LR groups, had a recurrence rate of approximately 10% 5 years or more after surgery (blue letters in Table\u0026nbsp;2a).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003e3.4. Additional analysis according to 95-gene classifier (95GC)\u003c/h2\u003e \u003cp\u003eWe attempted to classify these four middle groups according to 95GC [\u003cspan additionalcitationids=\"CR18 CR19 CR20 CR21 CR22\" citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. We applied 95GC to the patients of these four groups and classified them into the high- and low-risk groups (Table\u0026nbsp;2b). As a result, by attaching the 95GC high-risk group and the top row (the CTS5 high-risk and 42GC LR groups) and attaching the 95GC low-risk group and the bottom row (the CTS5 low-risk and 42GC NLR groups), we finally classified the patients into two risk groups with late recurrence rates of 17.7% (n\u0026thinsp;=\u0026thinsp;164) and 8.28% (n\u0026thinsp;=\u0026thinsp;507) (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0042; HR\u0026thinsp;=\u0026thinsp;2.13) (Table\u0026nbsp;2b).\u003c/p\u003e \u003c/div\u003e"},{"header":"4. Discussion","content":"\u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003e4.1. The micro-factors: The development of 42GC and the new validation\u003c/h2\u003e \u003cp\u003eWe first hypothesized the mechanisms of early and LR (Fig.\u0026nbsp;1) and proposed that both micro- and macro-factors are involved in recurrence. Then, in our previous report in 2018, we only analyzed micro-factors (gene expression patterns of primary tumors), corresponding to the slope angle of the lines (cell proliferation potency) in Fig.\u0026nbsp;1, and focused on the differences in gene expression between early and LR. Then, we developed, as a multigene assay, a prediction model specific for LR: 42GC [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. 42GC classifies the high-risk group for late recurrence as the LR group and the low-risk group as the NLR group. Thus, in the validation set of 221 patients, those classified into the LR group by 42GC were likely to develop recurrence significantly later than those classified into the NLR group (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.020) [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eIn this study, as a new independent validation of 42GC, we analyzed the 625 patients with recurrence, including their respective CEL files in public databases that were apart from the training sets, and again showed that the LR group was likely to develop recurrence significantly later than the NLR group (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;8.04e\u0026ndash;06) (Fig.\u0026nbsp;3).\u003c/p\u003e \u003cp\u003eHowever, in Fig.\u0026nbsp;3, although the difference was strongly significant, some patients in the LR group had early recurrence and some patients in the NLR group had late recurrence. This suggests the need for a more accurate prediction model for LR by adding the analysis of macro-factors (clinicopathological factors)\u0026mdash;CTS5. Additionally, because the 42GC LR group accounted for approximately half (52.2%) of the patients (Fig.\u0026nbsp;3), reducing the proportion of patients in the 42GC LR group was required to reduce the number of patients receiving extended hormonal therapy.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003e4.2. \u0026ldquo;The basic research\u0026rdquo;: Proof of our hypothesis in Fig.\u0026nbsp;1 for early and LR mechanisms\u003c/h2\u003e \u003cp\u003eTherefore, we first attempted to prove our hypothesis of early and LR mechanisms using not only micro-factors but also macro-factors in our hypothesis (Fig.\u0026nbsp;1).\u003c/p\u003e \u003cp\u003eAs mentioned above, Fig.\u0026nbsp;3 shows that the LR group with a smaller slope angle was more likely to develop recurrence significantly later than the NLR group according to 42GC (micro-factors).\u003c/p\u003e \u003cp\u003eFigure 4 shows that the CTS5 low-risk group, which was assumed to have a little amount of residual cancer cells after surgery, was more likely to develop recurrence significantly later according to the CTS5 (macro-factors).\u003c/p\u003e \u003cp\u003eThese results confirmed our hypothesis of early and LR mechanisms and demonstrated that both micro- and macro-factors affected the time of recurrence.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003e4.3. \u0026ldquo;The clinical research\u0026rdquo;: Development of a prediction model specific for LR\u003c/h2\u003e \u003cp\u003eNext, to develop a prediction model specific for LR, applying the aforementioned findings to clinical practice, we conducted a study involving patients with and without recurrence. In this study, we analyzed 671 patients without recurrence 5 years after surgery (approximately 10% of them have recurred afterward, and approximately 90% have not recurred until the end) by combining 42GC (micro-factors) and CTS5 (macro-factors) (Table\u0026nbsp;2a).\u003c/p\u003e \u003cp\u003eSurprisingly, in this situation where most patients did not have recurrence, the LR rate was found to be the highest in the CTS5 high-risk group, opposing the results of the analysis based on our mechanism (Fig.\u0026nbsp;4), which showed that the CTS5 low-risk group was more likely to develop recurrence later.\u003c/p\u003e \u003cp\u003eTherefore, the CTS5 low-risk group is likely to develop LR, based on the biological mechanism; however, the CTS5 low-risk group has a much lower \u003cem\u003eT\u003c/em\u003e,\u003cem\u003eN\u003c/em\u003e factors during surgery, and then, the recurrence rate itself in clinical practice is quite low and clinically not worthy of attention.\u003c/p\u003e \u003cp\u003eTherefore, distinguishing between the biological mechanism in \u0026ldquo;the basic research\u0026rdquo; and the decision-making for extended hormonal therapy in \u0026ldquo;the clinical research\u0026rdquo; is necessary. Based on this, the discussion on recurrence rates in the real world, where many patients do not have recurrence, appears to be more complex than the discussion on the biological mechanism.\u003c/p\u003e \u003cp\u003eIn the investigation of the six-group classification (Table\u0026nbsp;2a) for the 671 patients without recurrence 5 years after surgery (Fig.\u0026nbsp;2b), the late recurrence rate was 16.9% in the CTS5 high-risk and 42GC LR group combination, whereas it was 5.41% in the CTS5 low-risk and 42GC NLR group combination. The risk difference between the two groups was more than threefold, indicating that the former combination was a better indication for extended hormonal therapy than the latter. This is the main content of this study.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003e4.4. Additional analysis according to 95GC\u003c/h2\u003e \u003cp\u003eThe problem we confronted here was that the middle four groups (blue letters in Table\u0026nbsp;2a) had LR rates of approximately 10% (Table\u0026nbsp;2a). In the middle four groups, most (approximately 90%) had no LR and thus can be considered able to omit extended hormonal therapy. However, as an additional analysis, we attempted to further classify these four groups according to 95GC, a powerful recurrence risk classification method [\u003cspan additionalcitationids=\"CR18 CR19 CR20 CR21 CR22\" citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. We have previously reported the results of several studies that have significantly classified the intermediate-risk group by OncotypeDX into the two groups by 95GC [\u003cspan additionalcitationids=\"CR18 CR19 CR20 CR21 CR22\" citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. We newly classified the four groups into two groups according to 95GC: high and low. As a result, by attaching the 95GC high-risk group and the top row (the CTS5 high-risk and 42GC LR groups) and attaching the 95GC low-risk group and the bottom row (the CTS5 low-risk and 42GC NLR groups), we finally classified the four groups into the two risk groups with late recurrence rates of 17.7% (n\u0026thinsp;=\u0026thinsp;164) and 8.28% (n\u0026thinsp;=\u0026thinsp;507), respectively (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0042; HR\u0026thinsp;=\u0026thinsp;2.13) (Table\u0026nbsp;2b). Consequently, we could complete a new algorithm in which only approximately one-fourth of patients with ER-positive breast cancer received extended hormonal therapy and the remaining three-fourths did not receive extended hormonal therapy.\u003c/p\u003e \u003cp\u003eThus, the actual clinical practice is quite complex, and the results led to the conclusion that three methods of analysis\u0026mdash;42GC, CTS5, and 95GC\u0026mdash;were required to determine the candidates for extended hormonal therapy.\u003c/p\u003e \u003cp\u003eFurther validation with more patients is required.\u003c/p\u003e \u003c/div\u003e"},{"header":"5. Conclusion","content":"\u003cp\u003eHerein, we have successfully validated our hypothesis regarding early recurrence and LR mechanisms using 42GC and CTS5. We attempted to apply these findings to clinical data and have developed a novel algorithm that recommends extended hormonal therapy for approximately one-fourth of ER-positive patients with breast cancer.\u003c/p\u003e\n\u003cp\u003eThis study demonstrated that both micro-and macro-factors affected the time of recurrence. Additionally, there was a difference between a tendency toward LR based on biological mechanisms (the CTS5 low-risk group) and a higher LR rate (the CTS5 high-risk group). It would be beneficial to confirm our findings with a larger patient cohort in the future.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eER, estorogen receptor;\u0026nbsp;PR, progesterone receptor; HER2, human epidermal growth factor receptor 2;\u0026nbsp;42GC, 42-gene classifier; LR, late recurrence; NLR, non-late recurrence; CTS5, Clinical Treatment Score post-5 years; 95GC, 95-gene classifier; \u003cem\u003eT\u003c/em\u003e, tumor size; \u003cem\u003eN\u003c/em\u003e, nodal status; 5-year DRFS, 5-year distant recurrence-free survival; HR, hazard ratio\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAuthors\u0026rsquo; contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll authors contributed to the study conception and design.\u003c/p\u003e\n\u003cp\u003eConceptualization: \u003cstrong\u003eRyo Tsunashima\u003c/strong\u003e and \u003cstrong\u003eYasuto Naoi\u003c/strong\u003e; Methodology: \u003cstrong\u003eRyo Tsunashima\u003c/strong\u003e;\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eFormal analysis: \u003cstrong\u003eSae Kitano\u003c/strong\u003e and \u003cstrong\u003eSaya Matsumoto\u003c/strong\u003e;\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eWriting - original draft: \u003cstrong\u003eSae Kitano\u003c/strong\u003e;\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eData curation: \u003cstrong\u003eRyo Tsunashima\u003c/strong\u003e,\u003cstrong\u003e\u0026nbsp;Yoshiaki Sota\u003c/strong\u003e,\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eand\u003cstrong\u003e\u0026nbsp;Akira Watanabe\u003c/strong\u003e;\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eVisualization: \u003cstrong\u003eSae Kitano\u003c/strong\u003e;\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eSupervision: \u003cstrong\u003eYasuto Naoi\u003c/strong\u003e,\u003cstrong\u003e\u0026nbsp;Koichi Sakaguchi\u003c/strong\u003e,\u003cstrong\u003e\u0026nbsp;Midori Morita\u003c/strong\u003e,\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eand\u003cstrong\u003e\u0026nbsp;Chikage Kato\u003c/strong\u003e;\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eProject administration and Writing - review \u0026amp; editing: \u003cstrong\u003eYasuto Naoi\u003c/strong\u003e.\u003c/p\u003e\n\u003cp\u003eAll authors have read and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets analyzed during the current study are not publicly available due to the study protocol but are available from the corresponding author on reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics approval\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was performed in accordance with the principles of the Declaration of Helsinki and has been approved by the ethics committee of our institution.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eInformed consent was provided by the patients to participate in the study.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent to publish\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDeclaration of competing interest\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eYasuto Naoi\u0026nbsp;\u003c/strong\u003ehas received research funding from Sysmex, ONO, Daiichi-Sankyo and AstraZeneca, and honoraria from AstraZeneca, Pfizer, Eli Lilly, ONO, Daiichi-Sankyo and Chugai outside the submitted work; he holds joint patents with Sysmex including Curebest\u0026trade; 95GC Breast\u003csup\u003e\u0026nbsp;\u003c/sup\u003e(JP.5725274.B2) and has received patent royalties outside the submitted work.\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that no funds, grants, or other support were received during the preparation of this manuscript.\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe thank Professor Kenzo Shimazu from the Department of Breast and Endocrine Surgery, Osaka University Graduate School of Medicine, Osaka, Japan for contributing to this study.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eDavies C, Pan H, Godwin J, Gray R, Arriagada R, Raina V, et al (2013) Long-term effects of continuing adjuvant tamoxifen to 10 years versus stopping at 5 years after diagnosis of oestrogen receptor-positive breast cancer: ATLAS, a randomised trial. Lancet 381(9869):805-816. https://doi.org/10.1016/S0140-6736(12)61963-1\u003c/li\u003e\n\u003cli\u003eGray RG, Rea D, Handley K, Bowden SJ, Perry P, Earl HM, et al (2013) aTTom：long-term effects of continuing adjuvant tamoxifen to 10 years versus stopping at 5 years in 6,953 women with early breast cancer. J Clin Oncol 31(18):5.\u003c/li\u003e\n\u003cli\u003eSchiavon G, Smith IE (2014) Status of adjuvant endocrine therapy for breast cancer. Breast Cancer Res 16(2):206. https://doi.org/10.1186/bcr3636\u003c/li\u003e\n\u003cli\u003eGoss PE, Ingle JN, Martino S, Robert NJ, Muss HB, Piccart MJ, et al (2005) Randomized trial of letrozole following tamoxifen as extended adjuvant therapy in receptor-positive breast cancer: Updated findings from NCIC CTG MA.17. J Natl Cancer Inst 97(17):1262-1271. https://doi.org/10.1093/jnci/dji250\u003c/li\u003e\n\u003cli\u003eGoss PE, Ingle JN, Pritchard KI, Robert NJ, Muss H, Gralow J, et al (2016) Extending aromatase-inhibitor adjuvant therapy to 10 years. N Engl J Med 375(3):209-219. https://doi.org/10.1056/NEJMoa1604700\u003c/li\u003e\n\u003cli\u003eThe 2022 edition of the Clinical Practice Guidelines for the Japanese Breast Cancer Society, Treatment ed, Drug therapy in Clinical Question 4\u003c/li\u003e\n\u003cli\u003eIwase T, Saji S, Iijima K, Higaki K, Ohtani S, Sato Y, et al (2023) Postoperative adjuvant anastrozole for 10 or 5 years in patients with hormone receptor-positive breast cancer: AERAS, a randomized multicenter open-label phase III trial. J Clin Oncol 41(18):3329-3338. https://doi.org/10.1200/JCO.22.00577.\u003c/li\u003e\n\u003cli\u003eYamashita H, Ogiya A, Shien T, Horimoto Y, Masuda N, Inao T, et al (2016) Clinicopathological factors predicting early and late distant recurrence in estrogen receptor-positive, HER2-negative breast cancer. Breast Cancer 23(6):830-843. https://doi.org/10.1007/s12282-015-0649-0\u003c/li\u003e\n\u003cli\u003eKennecke HF, Olivotto IA, Speers C, Norris B, Chia SK, Bryce C, Gelmon KA (2007) Late risk of relapse and mortality among postmenopausal women with estrogen responsive early breast cancer after 5 years of tamoxifen. Ann Oncol 18(1):45-51. https://doi.org/10.1093/annonc/mdl334\u003c/li\u003e\n\u003cli\u003eSestak I, Dowsett M, Zabaglo L, Lopez-Knowles E, Ferree S, Cowens JW, Cuzick J (2013) Factors predicting late recurrence for estrogen receptor-positive breast cancer. J Natl Cancer Inst 105(19):1504-1511. https://doi.org/10.1093/jnci/djt244\u003c/li\u003e\n\u003cli\u003eSestak I, Cuzick J (2015) Markers for the identification of late breast cancer recurrence. Breast Cancer Res 17:10. https://doi.org/10.1186/s13058-015-0516-0\u003c/li\u003e\n\u003cli\u003ePan H, Gray R, Braybrooke J, Davies C, Taylor C, McGale P, et al (2017) 20-year risks of breast-cancer recurrence after stopping endocrine therapy at 5 years. N Engl J Med 377(19):1836-1846. https://doi.org/10.1056/nejmc1715968\u003c/li\u003e\n\u003cli\u003eTsunashima R, Naoi Y, Shimazu K, Kagara N, Shimoda M, Tanei T, Miyake T, Kim SJ, Noguchi S, et al (2018) Construction of a novel multi-gene assay (42-gene classifier) for prediction of late recurrence in ER-positive breast cancer patients. Breast Cancer Res Treat 171(1):33-41. https://doi.org/10.1007/s10549-018-4812-0\u003c/li\u003e\n\u003cli\u003eDowsett M, Sestak I, Regan MM, Dodson A, Viale G, Th\u0026uuml;rlimann B, Colleoni M, Cuzick J (2018) Integration of clinical variables for the prediction of late distant recurrence in patients with estrogen receptor\u0026ndash;positive breast cancer treated with 5 years of endocrine therapy: CTS5. J Clin Oncol 36(19): 1941-1948. https://doi.org/10.1200/JCO.2017.76.4258\u003c/li\u003e\n\u003cli\u003eNaoi Y, Kishi K, Tanei T, Tsunashima R, Tominaga N, Baba Y, Kim SJ, Taguchi T, Tamaki Y, Noguchi S (2011) Development of 95-gene classifier as a powerful predictor of recurrences in node-negative and ER-positive breast cancer patients. Breast Cancer Res Treat 128(3):633-641. https://doi.org/10.1007/s10549-010-1145-z\u003c/li\u003e\n\u003cli\u003eGyőrffy B, Benke Z, L\u0026aacute;nczky A, Bal\u0026aacute;zs B, Sz\u0026aacute;ll\u0026aacute;si Z, Tim\u0026aacute;r J, Sch\u0026auml;fer R (2012) RecurrenceOnline: An online analysis tool to determine breast cancer recurrence and hormone receptor status using microarray data. Breast Cancer Res Treat 132(3):1025-1034. https://doi.org/10.1007/s10549-011-1676-y\u003c/li\u003e\n\u003cli\u003eNaoi Y, Noguchi S (2016) Multi-gene classifiers for prediction of recurrence in breast cancer patients. Breast Cancer 23(1):12-18. https://doi.org/10.1007/s12282-015-0596-9.\u003c/li\u003e\n\u003cli\u003eNaoi Y, Saito Y, Kishi K, Shimoda M, Kagara N, Miyake T, Tanei T, Shimazu K, Kim SJ, Noguchi S (2019) Development of recurrence risk score using 95‑gene classifier and its application to formalin‑fixed paraffin‑embedded tissues in ER‑positive, HER2‑negative and node‑negative breast cancer. Oncol Rep 42(6):2680-2685. https://doi.org/10.3892/or.2019.7358.\u003c/li\u003e\n\u003cli\u003eNaoi Y, Tsunashima R, Shimazu K, Noguchi S (2021) The multigene classifiers 95GC/42GC/155GC for precision medicine in ER-positive HER2-negative early breast cancer. Cancer Sci 112(4):1369-1375. https://doi.org/10.1111/cas.14838.\u003c/li\u003e\n\u003cli\u003eFujii T, Masuda H, Cheng YC, Yang F, Sahin AA, Naoi Y, et al (2021) A 95-gene signature stratifies recurrence risk of invasive disease in ER-positive, HER2-negative, node-negative breast cancer with intermediate 21-gene signature recurrence scores. Breast Cancer Res Treat 189(2):455-461. https://doi.org/10.1007/s10549-021-06276-7.\u003c/li\u003e\n\u003cli\u003eTsukamoto F, Arihiro K, Takahashi M, Ito KI, Ohsumi S, Takashima S, Oba T, Yoshida M, Kishi K, Yamagishi K, Kinoshita T (2021) Multicenter retrospective study on the use of Curebest\u0026trade; 95GC breast for estrogen receptor-positive and node-negative early breast cancer. BMC Cancer 21(1):1077. https://doi.org/10.1186/s12885-021-08778-5.\u003c/li\u003e\n\u003cli\u003eYamashita H, Hatanaka KC, Yamagishi K, Saito Y, Hamasaki K, Taniguchi M, Okumura A, Nange A, Matsuno Y, Hatanaka Y (2023) Evaluation of 95-gene classifier of formalin-fixed paraffin-embedded tissues in ER-positive, HER2-negative, and node-negative breast cancer. Anticancer Res 43(2):707-711. https://doi.org/0.21873/anticanres.16209.\u003c/li\u003e\n\u003cli\u003eNaoi Y, Tsunashima R, Shimazu K, Oikawa M, Imanishi S, Koyama H, et al (2023) Validation of the prognosis of patients with ER‑positive, HER2‑negative and node‑negative invasive breast cancer classified as low risk by Curebest\u0026trade; 95GC breast in a multi‑institutional registry study. Oncol Lett 25(5):209. https://doi.org/10.3892/ol.2023.13794.\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003eTable 1 and 2 are available in the Supplementary Files section.\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"breast-cancer-research-and-treatment","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"brea","sideBox":"Learn more about [Breast Cancer Research and Treatment](https://www.springer.com/journal/10549)","snPcode":"10549","submissionUrl":"https://submission.nature.com/new-submission/10549/3","title":"Breast Cancer Research and Treatment","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"ER-positive breast cancer, Late recurrence, Extended hormonal therapy, 42-gene classifier, Clinical Treatment Score post-5 years","lastPublishedDoi":"10.21203/rs.3.rs-3389190/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-3389190/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003ePurpose\u003c/h2\u003e \u003cp\u003eThe mechanism of late recurrence (LR) of estrogen receptor (ER)-positive breast cancer remains unclear. As prediction models for LR of ER-positive breast cancer, 42-gene classifier (42GC), which analyzes \u0026ldquo;micro-factors (gene expression patterns)\u0026rdquo; and the Clinical Treatment Score post-5 years (CTS5), which analyzes \u0026ldquo;macro-factors (clinicopathological factors)\u0026rdquo;, were developed; however, improving the accuracy of these models is desirable. We aimed to clarify the mechanism and develop a new prediction model by combining 42GC and CTS5.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eWe selected 2,454 patients with ER-positive breast cancer from public microarray databases. We performed recurrence prognostic analysis using 42GC and CTS5.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eIn \u0026ldquo;the basic research\u0026rdquo; for recurrent patients (n\u0026thinsp;=\u0026thinsp;347), the 42GC LR and CTS5 low-risk groups tended to have LR. In \u0026ldquo;the clinical research\u0026rdquo; for recurrence-free patients 5 years after surgery (n\u0026thinsp;=\u0026thinsp;671), the 42GC LR and CTS5 high-risk group had a significantly higher LR rate after 5 years (16.9%) than the 42GC non-LR and CTS5 low-risk group (5.41%) (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.037).\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e \u003cp\u003eIn \u0026ldquo;the basic research,\u0026rdquo; we found that both micro-and macro-factors were associated with the mechanisms of early recurrence and LR. Meanwhile, in \u0026ldquo;the clinical research,\u0026rdquo; we found that the mechanistic tendency toward LR (the CTS5 low-risk group) differed from the high rate of LR (the CTS5 high-risk group). Therefore, differentiating between the biological mechanisms elucidated in \u0026ldquo;the basic research\u0026rdquo; and the decision-making process concerning extended hormonal therapy in \u0026ldquo;the clinical research\u0026rdquo; is necessary. These findings propose the development of a novel prediction model for LR.\u003c/p\u003e","manuscriptTitle":"Clarification attempt of the mechanism of late recurrence by micro- and macro-analyses in estrogen receptor-positive breast cancer","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2023-10-27 20:12:47","doi":"10.21203/rs.3.rs-3389190/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Reject, reconsider with recommended revisions","date":"2023-11-06T21:39:55+00:00","index":"","fulltext":""},{"type":"reviewerAgreed","content":"","date":"2023-10-22T12:30:49+00:00","index":0,"fulltext":""},{"type":"reviewersInvited","content":"","date":"2023-10-22T10:37:59+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"Breast Cancer Research and Treatment","date":"2023-10-20T15:50:13+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2023-10-13T07:10:44+00:00","index":"","fulltext":""},{"type":"submitted","content":"Breast Cancer Research and Treatment","date":"2023-10-13T01:55:19+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"breast-cancer-research-and-treatment","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"brea","sideBox":"Learn more about [Breast Cancer Research and Treatment](https://www.springer.com/journal/10549)","snPcode":"10549","submissionUrl":"https://submission.nature.com/new-submission/10549/3","title":"Breast Cancer Research and Treatment","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"41730b3a-f70e-434d-840c-cbf3969bc764","owner":[],"postedDate":"October 27th, 2023","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2023-10-27T20:12:47+00:00","versionOfRecord":[],"versionCreatedAt":"2023-10-27 20:12:47","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-3389190","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-3389190","identity":"rs-3389190","version":["v1"]},"buildId":"GqpaHPwrfC8PjnIFayRh5","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}