Clinicopathological features and prognostic analysis of Her-2-negative breast mucinous adenocarcinoma: A SEER database analysis

doi:10.21203/rs.3.rs-2723855/v1

Download PDF

Research Article

Clinicopathological features and prognostic analysis of Her-2-negative breast mucinous adenocarcinoma: A SEER database analysis

https://doi.org/10.21203/rs.3.rs-2723855/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background

Breast mucinous adenocarcinoma is a group of malignant tumors of epithelial origin of the breast with a good prognosis; however, whether its treatment regimen should be consistent with that of non-specific types of breast cancer remains to be investigated.

Methods

In this study, the clinicopathological information, treatment modalities, and prognosis of 10,466 patients with mucinous adenocarcinoma of the breast from the Surveillance, Epidemiology, and End Results (SEER) database were analyzed.

Results

After adjusting for other factors, radiotherapy showed a favorable benefit in most subgroups; although this may be associated with a higher proportion of breast-conserving surgery in early cases, chemotherapy did not show any benefit. In particular, estrogen receptor-negative subgroup patients who received chemotherapy had a poorer prognosis.

Conclusions

Although derived from retrospective studies, the results of these analyses revealed trends that may warrant further study.

breast cancer

mucinous adenocarcinoma

chemotherapy

radiotherapy

As the most common type of malignancy worldwide, the main pathological type of breast cancer is non-specifically invasive [1,2]. Relatively well-established treatment guidelines have been developed for non-specifically invasive breast cancer [3,4]. However, guidelines for the treatment of some specific types of breast cancer, such as mucinous adenocarcinoma, have not yet been developed, and there is still a great deal of controversy regarding how to standardize the treatment of patients with mucinous adenocarcinoma because their prognosis is often very good [5,6]. Mucinous adenocarcinoma has a low incidence and is uncommon in clinical practice. Even when multicenter clinical trials are conducted to study specific treatment options, they require longer enrollment and follow-up periods, which limits similar efforts. High-quality screening and retrospective analysis of previous mucinous adenocarcinoma cases are effective alternatives to address this problem. Retrospective analyses, while not providing the same level of evidence as prospective analyses, are often more reflective of real-world treatment and prognosis and provide ideas for treatment improvement and clinical trials.

The sample size significantly impacts the credibility of the statistics; however, the veracity and accuracy of large datasets are often overlooked. After a certain level of data volume is reached, further increases do not provide richer information but rather highlight non-significant differences. If there is some bias in the big dataset itself, as data volume increases, these differences will likely be magnified, thus misleading clinical decisions.

All the cases we included here were Herceptin 2 (Her-2)-negative because the biological process involved in Her-2-positive breast cancer is largely dominated by the Her-2 protein and is managed using targeted therapy [7-10]. In contrast, we paid more attention to the therapeutic characteristics of other types of breast cancer that still lack protein therapies targeting tumor cell membranes. At the same time, owing to the early stages of breast mucinous adenocarcinoma, a considerable proportion of patients undergo breast-conserving surgery. Radiotherapy after breast-conserving surgery is a standard process [11-14], so the proportion of patients receiving it in this study was high.

Tumors are highly heterogeneous, and hence, it is difficult to specify the disease progression pattern that should be followed during the disease for a specific individual. Therefore, the data cleaning work performed often only removed those incomplete entries, but there was no way to verify the correctness of the remaining data entries. This uncertainty also had to be compensated for by increasing the sample size. However, this limitation cannot be avoided in retrospective studies.

To compensate for such shortcomings, we aimed to observe whether the data distribution is balanced among different groups. However, grouping conditions heavily depend on clinicopathological factors, and achieving a balanced group in retrospective studies is difficult. To eliminate the overall imbalance, we also enriched the subgroup data by increasing the sample size for a more detailed statistical analysis. We also performed the univariate and multifactor regression analyses of the individual clinicopathological factors included in the analysis to determine whether the data are consistent with the objective situation, especially in the multifactor regression analysis. We expect the results of this study to reveal the real situation in clinical practice and provide ideas and references for the design and conduct of subsequent related clinical trials and guideline development.

Patients

We searched the SEER database for clinicopathological and prognostic information on patients with mucinous adenocarcinoma of the breast. The inclusion criteria were: 1. patients diagnosed with breast mucinous adenocarcinoma and underwent breast cancer surgery; 2. 18–75 years old; 3. no distant metastasis before the operation; and 4. complete clinicopathological information, including whether chemotherapy or radiotherapy was administered. The exclusion criteria were: 1. patients younger than 18 years or older than 75 years old; 2. presence of distant metastatic lesions before the operation; and 3. incomplete clinicopathological information. Clinical and pathological findings and follow-up data, including age, tumor grade, tumor stage, nodal stage, estrogen receptor (ER) status, progesterone receptor (PR) status, radiation, and chemotherapy, were collected prospectively and recorded in a database. Eventually, the number of patients we included in the statistical analysis was 10,466.

Statistical analysis

Data from all eligible patients were analyzed for overall survival. Survival curves were estimated using the Kaplan–Meier method, and treatment comparisons were made using the log-rank test. Potential prognostic factors were entered into a Cox regression model, including age, tumor grade, stage, tumor stage, nodal stage, ER status, PR status, radiation, and chemotherapy. In the multivariate analysis, the prognostic factors detected in the univariate analysis and treatment group were included as covariates in the Cox regression model. Two-sided P-values were calculated for all tests and are reported here. P-values less than 0.05 were considered to indicate statistical significance. Analyses were performed using SPSS software, version 16.0.

Patient characteristics

Nine clinicopathological factors were identified and included in the analysis. The distribution of various clinicopathological factors was subsequently compared between the two groups (chemotherapy or not and radiation or not). We determined that only ER and PR were evenly distributed in the grouping of whether or not to administer radiotherapy, which stems from the fact that their expression status has no significant influence on the application of radiotherapy in clinical practice (Table 1). The other items were unevenly distributed in the middle of whether or not to administer radiotherapy, thereby indicating these clinicopathological factors profoundly influence the decision to administer radiotherapy and the strong correlation between them and the decision to administer chemotherapy. Moreover, in both groups with or without chemotherapy, all clinicopathological factors were not equally distributed because even the ER and PR expression status, which reached an equilibrium distribution in both groups with or without radiotherapy, could profoundly influence the decision to treat with chemotherapy.

Univariable analyses and multivariable analyses
In the univariate analyses, we found that each factor included in the statistical analysis presented significant internal prognostic differences. Patients older than 60 years showed a worse prognosis; the prognosis progressively worsened with increasing histological grade; positive expression of ER and PR were the signs of a good prognosis; and T-stage, N-stage, and TNM-stage also showed a distinct trend of prognostic changes. Patients who received radiotherapy showed a better prognosis, which was normal; then, patients who received chemotherapy showed a significant and unexpected trend of poor prognosis (Figure 1).
All factors, excluding stage, were included in the multivariate regression analysis. The reasons for this are explained in detail in the Discussion. After multifactorial regression, the independent prognostic function of ER status and chemotherapy was eliminated, and the prognostic importance of histological grade diminished; however, histological grade remained an independent prognostic factor. In contrast, radiotherapy remained an independent prognostic factor (Table 2).

Subgroup analysis of the benefit of chemotherapy

Subgroup analysis in the two groups of patients who received or did not receive chemotherapy was performed using multifactorial regression to ensure the comprehensiveness of the statistical results. The specific distribution of patients and comparative values of the prognostic differences are presented in Table 3. For ease of observation, we visualized the data in a forest plot (Figure 2). Chemotherapy was found to cause significant prognostic differences in almost all subgroups. Only three subgroups showed significant prognostic changes in response to chemotherapy: ER-negative, stage I, and stage IIA. In stages I and IIA, chemotherapy brought a relatively poor prognosis. However, contrary to our conventional perception, ER-negative patients, instead of benefiting from chemotherapy, have a significantly poorer prognosis due to chemotherapy. The survival curves for this group of patients are shown in Figure 3, which shows a steep decline in the survival curves of patients who received chemotherapy compared to those of patients who did not receive chemotherapy at approximately 5 years postoperatively.

Subgroup analysis of the benefit of radiotherapy

In contrast to the uncertain benefits of chemotherapy, almost all subgroups benefited from radiotherapy. As shown in the forest plot, the HR and 95% confidence interval bands in many subgroups were not only relatively leftward but also short in length, indicating that the benefit of radiotherapy was clear and consistent in most patients. All subgroups benefitted from radiotherapy, except for the late-stage and late T-stage subgroups, where the benefit was not significant.

We performed a retrospective study by retrieving data from the SEER database of over 10,000 cases of breast mucinous adenocarcinoma for analysis. Univariate analysis of each clinicopathological factor revealed that the dataset was consistent with the distribution of characteristics and prognostic trends in general breast cancer cases. Among factors such as T-stage and N-stage, the HR ratio increased significantly as the stage increased, indicating that these data are of high quality and can provide a credible database for the prognostic impact of treatment modalities. Meanwhile, the large data volume also provided statistical credibility for subgroup analysis.

In the two groups of patients with or without radiotherapy, all factors except ER and PR status showed an unbalanced distribution, whereas in the two groups with or without chemotherapy, all factors showed an unbalanced distribution. This is consistent with the treatment guidelines and clinicians' decisions that molecular typing has little influence on radiotherapy decisions, whereas molecular typing and TNM staging directly determine the need for postoperative chemotherapy. This uneven distribution is thought to affect the correctness of the subsequent statistical results; that is, if there is a higher proportion of patients with poor prognosis in a certain group, then the survival curve of this group tends to be more toward poor survival status. We conclude that in the present study, although the distribution of patients between the groups was uneven, the results of the subsequent subgroup analysis, especially under multifactorial regression, provided credible statistical conclusions.

Table 2 shows that the data characteristics of this study were consistent with the clinical and prognostic profiles of breast cancer. In the univariate regression analysis, age >60 years, high histological grade, negative ER and PR expression, high TNM-stage, not receiving radiotherapy, and having received chemotherapy showed significantly poor prognostic characteristics. In particular, the HR increased with higher stages of the T-stage, N-stage, and TNM-stage, and this trend remained significant after multifactorial regression analysis. We did not include the stage in the multifactorial regression analysis because there was a large causal relationship between it (T-stage and N-stage) and the joint inclusion, which could easily cause bias in the statistical results. However, the causal relationship between each other factor was not very strong to analyze the patients as a whole more comprehensively, and also because of the other factors in the single-factor regression. All factors except stage were included in the multifactor regression to provide a more comprehensive analysis of the patients as a whole and also because other factors showed independence in the univariate regression. We determined from the results of the multifactorial regression analysis that the presence or absence of chemotherapy did not have a strong effect on the prognosis of the patients in general, whereas receiving radiotherapy did have a significant impact on the overall survival of the patients. This is inconsistent with the results of some previous analyses; therefore, we needed to perform further subgroup analyses to stratify the patients carefully to determine which patients could benefit significantly from radiation versus chemotherapy.
Thus, we performed a subgroup analysis using multifactorial regression to identify subgroups that could significantly benefit from chemotherapy and radiotherapy in terms of survival differences. During the analysis, factors other than those in which the subgroup was located were included in the analysis to obtain comprehensive results. For example, when analyzing the differences in benefits with or without chemotherapy in the ER-negative subgroup, the ER factor was excluded from the multifactorial analysis, and all factors except ER status were included in the multifactorial analysis. Data on the benefits of chemotherapy and radiotherapy in the subgroups for each factor are presented in Tables 3 and 4, respectively. For ease of reading, we present the data as a forest plot. First, chemotherapy did not provide a significant benefit in most subgroups. Even in most subgroups, patients not treated with chemotherapy showed some survival advantage, although this was not statistically significant. However, impaired survival after chemotherapy was significant in patients with stage I and IIA disease. This can be because earlier patients, to a large extent, did not need chemotherapy, regardless of whether they underwent breast-conserving surgery or total mastectomy, whereas, for patients with more advanced disease, chemotherapy barely resulted in a significant survival benefit and showed a significant trend towards impaired survival. Surprisingly, chemotherapy significantly impaired survival in ER-negative patients.
In contrast to the impaired survival observed with chemotherapy in almost all subgroups, radiotherapy showed a diametrically opposite trend. Almost all subgroups of patients benefited from radiotherapy. In the forest plot, the benefit of radiotherapy was not significant in the subgroups with late disease, such as T3, T4, N1, and N2 stage disease. Our analysis suggests that this may be related to the fact that patients with mucinous adenocarcinoma generally have an early stage of disease, and a high proportion of them undergo breast-conserving surgery. Patients with early-stage disease obtained a significant survival benefit from radiotherapy. However, in the late stage of the disease, when radiotherapy was required regardless of surgery, the survival benefit of radiotherapy was small after combining several factors. Of course, due to the lack of information on the surgical modality in the raw data, it is not possible to stratify the data by surgical modality. Therefore, it is not possible to determine the exact benefit of radiotherapy for patients with advanced disease who underwent a total mastectomy. From only the data available to us, it appears that the benefit of radiotherapy diminishes with progressively later disease stages, suggesting that patients with early-stage disease are generally sensitive to radiotherapy, and in patients with more advanced disease stages, radiotherapy benefit may not account for as much of the comprehensive treatment.

A considerable proportion (at least 80%) of patients in this study were hormone receptor-positive. Currently, endocrine and targeted treatment of these patients is increasingly tending towards multi-drug combinations and long-term medication [15-20]. Since the proportion of patients with mucinous adenocarcinoma is slightly higher than that of patients with non-specific breast cancer, the synergy between endocrine therapy, chemotherapy, radiotherapy, and other treatment methods will also be the focus of the research. In addition, the Her-2 low expression subgroup may have some independent pathological and clinical features in non-specific breast cancer [21], and whether such patients with mucinous adenocarcinoma should be treated independently is also a question to be studied.

SEER: The Surveillance, Epidemiology, and End Results

Her-2: Herceptin 2

ER: estrogen receptor

HR: hormone receptor

PR: progesterone receptor

TNM: tumor, node, metastasis

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Availability of data and materials

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Competing interests

The authors declare that they have no competing interests.

Funding

This study was not supported by any funding,

Authors’ contributions

XL and HZ participated in study design and manuscript drafting. QX and ZW participated in statistical analysis and manuscript drafting. All authors have read and approved the final version of the manuscript.

Acknowledgements

Not applicable.

Loibl Sibylle, Poortmans Philip, Morrow Monica et al. Breast cancer [J]. Lancet, 2021, 397: 1750-1769.
Casasent Anna K, Almekinders Mathilde M, Mulder Charlotta et al. Learning to distinguish progressive and non-progressive ductal carcinoma in situ [J]. Nat Rev Cancer, 2022, 22: 663-678.
Biganzoli Laura, Battisti Nicolò Matteo Luca, Wildiers Hans et al. Updated recommendations regarding the management of older patients with breast cancer: a joint paper from the European Society of Breast Cancer Specialists (EUSOMA) and the International Society of Geriatric Oncology (SIOG).[J] .Lancet Oncol, 2021, 22: e327-e340.
Tung Nadine M, Zakalik Dana, Somerfield Mark R et al. BRCA Adjuvant PARP Inhibitors in Patients With High-Risk Early-Stage HER2-Negative Breast Cancer and Germline Mutations: ASCO Hereditary Breast Cancer Guideline Rapid Recommendation Update.[J] .J Clin Oncol, 2021, 39: 2959-2961.
Marrazzo E, Frusone F, Milana F, Sagona A, Gatzemeier W, Barbieri E, Bottini A, Canavese G, Rubino AO, Eboli MG, Rossetti CM, Testori A, Errico V, De Luca A, Tinterri C. Mucinous breast cancer: A narrative review of the literature and a retrospective tertiary single-centre analysis. Breast. 2020 Feb;49:87-92.
Ginter PS, Tang X, Shin SJ. A review of mucinous lesions of the breast. Breast J. 2020 Jun;26(6):1168-1178.
Swain Sandra M, Shastry Mythili, Hamilton Erika, Targeting HER2-positive breast cancer: advances and future directions.[J] .Nat Rev Drug Discov, 2023, 22: 101-126.
Goutsouliak Kristina, Veeraraghavan Jamunarani, Sethunath Vidyalakshmi et al. Towards personalized treatment for early stage HER2-positive breast cancer [J]. Nat Rev Clin Oncol, 2020, 17: 233-250.
Choong Grace M, Cullen Grace D, O'Sullivan Ciara C, Evolving standards of care and new challenges in the management of HER2-positive breast cancer.[J] .CA Cancer J Clin, 2020, 70: 355-374.
Oh Do-Youn, Bang Yung-Jue, HER2-targeted therapies - a role beyond breast cancer [J]. Nat Rev Clin Oncol, 2020, 17: 33-48.
Recht Abram, Whole-Breast Irradiation Is the Preferred Standard of Care for the Majority of Patients With Early-Stage Breast Cancer [J]. J Clin Oncol, 2020, 38: 2263-2267.
Hepel Jaroslaw T, Wazer David E, Partial Breast Irradiation Is the Preferred Standard of Care for a Majority of Women With Early-Stage Breast Cancer [J]. J Clin Oncol, 2020, 38: 2268-2272.
Moran Meena S, Ho Alice Y, Radiation Therapy for Low-Risk Breast Cancer: Whole, Partial, or None?[J]. J Clin Oncol, 2022, 40: 4166-4172.
Torres Mylin A, Horst Kathleen C, Freedman Gary M, Postmastectomy and Regional Nodal Radiation for Breast Cancer [J]. J Clin Oncol, 2020, 38: 2299-2309.
Agostinetto Elisa, Gligorov Joseph, Piccart Martine, Systemic therapy for early-stage breast cancer: learning from the past to build the future [J]. Nat Rev Clin Oncol, 2022, 19: 763-774.
Hicks David, O'Regan Ruth M, Improving Outcomes for High-Risk Hormone Receptor-Positive Breast Cancer With CDK Inhibition [J]. J Clin Oncol, 2022, 40: 1142-1146.
Borges Virginia F, Options for Endocrine-Refractory, Hormone Receptor-Positive Breast Cancer: Which Target and When?[J]. J Clin Oncol, 2021, 39: 3890-3896.
Burstein Harold J, Systemic Therapy for Estrogen Receptor-Positive, HER2-Negative Breast Cancer [J]. N Engl J Med, 2020, 383: 2557-2570.
Pinkerton JoAnn V, Hormone Therapy for Postmenopausal Women [J]. N Engl J Med, 2020, 382: 446-455.
Lambertini Matteo, Blondeaux Eva, Perrone Francesco et al. Improving Adjuvant Endocrine Treatment Tailoring in Premenopausal Women With Hormone Receptor-Positive Breast Cancer [J]. J Clin Oncol, 2020, 38: 1258-1267.
Tarantino Paolo, Hamilton Erika, Tolaney Sara M et al. HER2-Low Breast Cancer: Pathological and Clinical Landscape [J]. J Clin Oncol, 2020, 38: 1951-1962.

Table 1: Characteristics of radiation and chemotherapy population (n = 10466)

	Radiation (%)			Chemotherapy (%)
	No	Yes	p Value	No	Yes	p Value
Characteristics
Age group			< 0.001			< 0.001
≤ 60	1320 (24)	1845 (37)		2118 (24)	1047 (72)
> 60	4188 (76)	3113 (63)		6892 (76)	409 (28)
Grade			< 0.001			< 0.001
Non-known	1079 (20)	807 (16)		1663 (19)	223 (15)
I	2664 (48)	2583 (52)		4708 (52)	539 (37)
II	1546 (28)	1380 (28)		2385 (26)	541 (37)
III	219 (4)	188 (4)		254 (3)	153 (11)
ER			0.524			< 0.001
Negative	151 (3)	126 (3)		194 (2)	83 (6)
Positive	5357 (97)	4832 (97)		8816 (98)	1373 (94)
PR			0.039			< 0.001
Negative	829 (15)	676 (14)		1195 (13)	310 (21)
Positive	4679 (85)	4282 (86)		7815 (87)	1146 (79)
Stage			< 0.001			< 0.001
I	3499 (64)	3613 (73)		6631 (73)	481 (33)
IIA	1492 (27)	965 (20)		1909 (21)	548 (38)
IIB	345 (6)	214 (4)		325 (4)	234 (16)
IIIA	95 (2)	118 (2)		73 (1)	140 (10)
IIIB	77(1)	48 (1)		72 (1)	53 (3)
Pathological tumor stage			< 0.001			< 0.001
T1	3707 (67)	3799 (77)		6852 (76)	654 (45)
T2	1511 (27)	964 (19)		1871 (21)	604 (41)
T3	213 (4)	147 (3)		215 (2)	145 (10)
T4	77 (2)	48 (1)		72 (1)	53 (4)
Pathological nodal stage			0.001			< 0.001
N0	4987 (91)	4506 (91)		8535 (94)	958 (66)
N1	464 (8)	363 (7)		432 (5)	395 (27)
N2	57 (1)	89 (2)		43 (1)	103 (7)
Radiation						< 0.001
No				4916 (55)	592 (41)
Yes				4094 (45)	864 (59)
Chemotherapy			< 0.001
No	4916 (89)	4094 (83)
Yes	592 (11)	864 (17)

Table 2: HR for death in intention-to-treat population (n = 10466) —univariable and multivariable analyses

	Univariable analyses		Multivariable analyses
	HR (95% CI)	p Value*	HR (95% CI)	p Value†
Age group		< 0.001		< 0.001
≤ 60	1 (Ref)		1 (Ref)
> 60	1.643 (1.369 - 1.972)	< 0.001	1.946 (1.594 - 2.376)	< 0.001
Grade		< 0.001		0.035
Non-known	1 (Ref)		1 (Ref)
I	0.792 (0.638 - 0.983)	0.034	0.854 (0.688 - 1.061)	0.153
II	1.099 (0.877 - 1.378)	0.413	1.000 (0.796 - 1.256)	0.998
III	1.975 (1.424 - 2.738)	< 0.001	1.610 (1.153 - 2.247)	0.005
ER		0.001		0.277
Negative	1 (Ref)		1 (Ref)
Positive	0.536 (0.373 - 0.769)	0.001	0.822 (0.556 - 1.215)	0.325
PR		< 0.001		< 0.001
Negative	1 (Ref)		1 (Ref)
Positive	0.589 (0.488 - 0.711)	< 0.001	0.675 (0.550 - 0.826)	< 0.001
Stage		< 0.001
I	1 (Ref)
IIA	2.158 (1.794 - 2.595)	< 0.001
IIB	3.836 (2.964 - 4.964)	< 0.001
IIIA	6.174 (4.533 - 8.410)	< 0.001
IIIB	10.117 (7.117 - 14.382)	< 0.001
Pathological tumor stage		< 0.001		< 0.001
T1	1 (Ref)		1 (Ref)
T2	2.400 (2.018 - 2.855)	< 0.001	2.117 (1.769 - 2.533)	< 0.001
T3	5.071 (3.853 - 6.675)	< 0.001	4.051 (3.018 - 5.437)	< 0.001
T4	9.658 (6.807 - 13.702)	< 0.001	6.072 (4.153 - 8.877)	< 0.001
Pathological nodal stage		< 0.001		< 0.001
N0	1 (Ref)		1 (Ref)
N1	2.404 (1.944 - 2.973)	< 0.001	1.599 (1.265 - 2.022)	< 0.001
N2	4.574 (3.215 - 6.507)	< 0.001	2.221 (1.499 - 3.292)	< 0.001
Radiation		< 0.001		< 0.001
No	1 (Ref)		1 (Ref)
Yes	0.521 (0.442 - 0.613)	< 0.001	0.590 (0.499 - 0.699)	< 0.001
Chemotherapy		< 0.001		0.261
No	1 (Ref)		1 (Ref)
Yes	1.509 (1.248 - 1.825)	< 0.001	1.095 (0.866 - 1.384)	0.448

Ref=reference category.

*Derived from tests of HR for prognostic factors in univariate model adjusted for treatment group in Cox proportional-hazards model.

†Cox-regression analysis, controlling for prognostic factors listed in table.

Table 3. Tests for Heterogeneity of Treatment Effect According to the Clinicopathological Characteristics of the Patients with and without Chemotherapy.

Subgroup	Without chemotherapy	With chemotherapy	HR (95% CI)	p Value
	No. of deaths/no. of patients
Total	488/9010	136/1456	1.143 (0.905 – 1.442)	0.261
Age group
≤ 60	68/2118	86/1047	1.360 (0.948 – 1.951)	0.095
> 60	420/6892	50/409	0.950 (0.684 – 1.319)	0.759
Grade
Non-known	101/1663	21/223	1.327 (0.758 – 2.326)	0.322
I	224/4708	31/539	0.870 (0.569 – 1.331)	0.521
II	138/2385	58/541	1.167 (0.790 – 1.723)	0.438
III	25/254	26/153	1.368 (0.682 – 2.744)	0.378
ER
Negative	13/194	18/83	2.627 (1.085 – 6.357)	0.032
Positive	475/8816	118/1373	1.049 (0.822 – 1.340)	0.701
PR
Negative	89/1195	51/310	1.398 (0.900 – 2.171)	0.136
Positive	399/7815	85/1146	0.996 (0.752 – 1.318)	0.976
Stage
I	257/6631	22/481	2.462 (1.766 – 3.431)	< 0.001
IIA	148/1909	42/548	1.796 (1.241 – 2.597)	0.002
IIB	44/325	29/234	1.473 (0.856 – 2.532)	0.162
IIIA	19/73	28/140	1.471 (0.798 – 2.712)	0.217
IIIB	20/72	15/53	1.692 (0.706 – 4.055)	0.238
Pathological tumor stage
T1	276/6852	34/654	1.373 (0.912 – 2.064)	0.128
T2	153/1871	65/604	1.154 (0.811 – 1.641)	0.427
T3	39/215	22/145	0.569 (0.299 – 1.083)	0.086
T4	20/72	15/53	0.814 (0.361 – 1.834)	0.619
Pathological nodal stage
N0	422/8535	66/958	1.238 (0.928 – 1.653)	0.146
N1	55/432	48/395	0.894 (0.566 – 1.414)	0.633
N2	11/42	22/103	1.013 (0.396 – 2.591)	0.978
Radiation
No	327/4916	71/592	1.297 (0.972 – 1.731)	0.078
Yes	161/4094	65/864	0.893 (0.594 – 1.343)	0.586

The P values are for hazard ratios for death in the group with and without chemotherapy, with 95% confidence intervals.

The treatment is better for without chemotherapy when the HR is > 1; and is better for with chemotherapy when the HR is < 1.

Table 4. Tests for Heterogeneity of Treatment Effect According to the Clinicopathological Characteristics of the Patients with and without Radiation.

Subgroup	Without Radiation	With Radiation	HR (95% CI)	p Value
	No. of deaths/no. of patients
Total	398/5508	226/4958	0.580 (0.490 – 0.685)	< 0.001
Age group
≤ 60	80/1320	74/1845	0.574 (0.416 – 0.791)	0.001
> 60	318/4188	152/3113	0.585 (0.481 – 0.712)	< 0.001
Grade
Non-known	78/1079	44/807	0.668 (0.458 – 0.973)	0.036
I	168/2664	87/2583	0.519 (0.399 – 0.677)	< 0.001
II	120/1546	76/1380	0.654 (0.487 – 0.878)	0.005
III	32/219	19/188	0.380 (0.195 – 0.740)	0.004
ER
Negative	20/151	11/126	0.241 (0.104 – 0.556)	0.001
Positive	378/5357	215/4832	0.589 (0.496 – 0.699)	< 0.001
PR
Negative	83/829	57/676	0.694 (0.488 – 0.989)	0.043
Positive	315/4679	169/4282	0.536 (0.443 – 0.648)	< 0.001
Stage
I	171/3499	108/3613	0.536 (0.420 – 0.682)	< 0.001
IIA	133/1492	57/965	0.591 (0.431 – 0.811)	0.001
IIB	45/345	28/214	0.988 (0.604 – 1.617)	0.963
IIIA	26/95	21/118	0.476 (0.236 – 0.958)	0.038
IIIB	23/77	12/48	0.680 (0.301 – 1.538)	0.354
Pathological tumor stage
T1	193/3707	117/3799	0.522 (0.414 – 0.657)	< 0.001
T2	143/1511	75/964	0.671 (0.503 – 0.896)	0.007
T3	39/213	22/147	0.723 (0.405 – 1.289)	0.272
T4	23/77	12/48	0.680 (0.301 – 1.538)	0.354
Pathological nodal stage
N0	316/4987	172/4506	0.579 (0.480 – 0.699)	< 0.001
N1	66/464	37/363	0.692 (0.452 – 1.060)	0.090
N2	16/57	17/89	0.437 (0.185 – 1.036)	0.060
Chemotherapy
No	327/4916	161/4094	0.622 (0.512 – 0.754)	< 0.001
Yes	71/592	65/864	0.477 (0.336 – 0.676)	< 0.001

The P values are for hazard ratios for death in the group with and without chemotherapy, with 95% confidence intervals.

The treatment is better for without radiation when the HR is > 1; and is better for with radiation when the HR is < 1.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Clinicopathological features and prognostic analysis of Her-2-negative breast mucinous adenocarcinoma: A SEER database analysis

Status:

Version 1

Abstract

Figures

Background

Methods

Results

Discussion

Conclusions

Abbreviations

Declarations

References

Tables

Additional Declarations

Status:

Version 1