ORIGINAL ARTICLE

pISSN 2288-6575 • eISSN 2288-6796

[Link]
Annals of Surgical Treatment and Research

A simplified risk scoring system for predicting high-risk

groups in gene expression tests for patients with estrogen
receptor-positive, human epidermal growth factor
receptor 2-negative, and node-positive breast cancer
Kwang Hyun Yoon1,3,*, Suk Jun Lee2,*, Yujin Kim2, Jee Hyun Ahn2, Jee Ye Kim2, Hyung Seok Park2, Seung Il Kim2,
Seho Park2
1
Department of Surgery, Gangneung Asan Medical Center, Gangneung, Korea
2
Division of Breast Surgery, Department of Surgery, Yonsei University College of Medicine, Seoul, Korea
3
Yonsei University Graduate School of Medicine, Seoul, Korea

Purpose: The gene expression test (GET) was used to predict the response to chemotherapy and the recurrence risk.
Several randomized clinical trials have demonstrated that some patients with node-positive disease can achieve favorable
survival outcomes even without adjuvant chemotherapy. This study aimed to predict the results of Oncotype DX (Genomic
Health) and MammaPrint (Agendia) using traditional clinicopathological factors.
Methods: We reviewed the records of 311 patients who underwent GET for hormone receptor-positive/human epidermal
growth factor receptor 2 (HER2)-negative primary invasive breast cancer with node-positive disease between 2015 and
2022 at Severance Hospital and Gangneung Asan Medical Center. Univariate and multivariate logistic regression analyses
assessed the relationships between clinicopathological variables and risk stratification using the GET results.
Results: A simple scoring system was created by assigning integer values to each variable. A score of 3 was assigned
for histological grade 3, a score of 2 for pathologic T2 or above, and a score of 1 for a lower progesterone receptor (1–20
or Alled score 3–6), HER2 2-positive, and high Ki-67 (>20). In the validation cohort, overall accuracy was 0.798 (95%
confidence interval, 0.744–0.844).
Conclusion: The high GET risk results can be predicted using traditional clinicopathological factors: tumor size,
progesterone receptor, histological grade, HER2, and Ki-67. These results will be useful for treatment decision-making
among clinically high-risk patients with HR-positive/HER2-negative and node-positive disease, helping to identify patients
to whom the GET assay may not apply.
[Ann Surg Treat Res 2023;105(6):360-368]

Key Words: Breast neoplasms, Gene expression test, Lymphatic metastasis, MammaPrint, Oncotype DX

INTRODUCTION tumor burden, hormone receptor (HR) status (estrogen receptor

[ER] or progesterone receptor [PR]), human epidermal growth
Postoperative adjuvant systemic treatment for breast cancer is factor receptor 2 (HER2), and the Ki-67 index [1-5]. Axillary
conducted considering various oncological factors, including age, lymph node (ALN) metastases have served as indicators of

Received August 18, 2023, Revised September 9, 2023, *Kwang Hyun Yoon and Suk Jun Lee have contributed equally to this work
Accepted October 5, 2023 as co-first authors.
Corresponding Author: Seho Park Copyright ⓒ 2023, the Korean Surgical Society
Division of Breast Surgery, Department of Surgery, Yonsei University cc Annals of Surgical Treatment and Research is an Open Access Journal. All
College of Medicine, 50-1 Yonsei-ro, Seodaemun-gu, Seoul 03722, Korea articles are distributed under the terms of the Creative Commons Attribution Non-
Tel: +82-2-2228-2134, Fax: +82-2-313-8289 Commercial License ([Link] which
E-mail: PSH1025@[Link] permits unrestricted non-commercial use, distribution, and reproduction in any
ORCID: [Link] medium, provided the original work is properly cited.

360
 Kwang Hyun Yoon, et al: Traditional factors can predict gene expression result in ALN metastases

poor prognosis and criteria for adjuvant systemic treatment [6]. of ODX and MMP in patients with ER (+), HER2 (−), and ALN
Even with confirmed ALN metastases, some patients with HR metastases.
(+) and HER2 (−) breast cancer tend to show relatively good
oncological outcomes compared with other subtypes, even METHODS
without undergoing chemotherapy (CTx) [7,8]. Previously, CTx
was uniformly administered based on anatomical staging [2]. The project was reviewed and approved by the Institutional
However, due to the high survival rates in these groups, there Review Board (IRB) of the Severance Hospital (No. 4-2023-0725)
has been a shift in research focus toward minimizing the side and Gangneung Asan Medical Center (No. 2023-08-006). The
effects of treatments [9]. Current research aims to tailor these patient’s consent to participate was waived by the IRB owing to
treatments to individuals to reduce side effects and maintain the retrospective nature of the study.
similar or noninferior survival outcomes. Among such research,
the RxPONDER (Rx for Positive Node, Endocrine Responsive Patient selection and histopathologic
Breast Cancer) and MINDACT (Microarray In Node-Negative and characteristics
1 to 3 Positive Lymph Node Disease May Avoid Chemotherapy) Patients with primary breast cancer who underwent
trials are the most representative studies [7,8]. upfront surgery between January 2015 and December 2022
The Oncotype DX (ODX; Genomic Health) is a diagnostic tool were retrospectively selected from the medical databases of
that analyzes the expression of 16 breast cancer-related genes Severance Hospital (Seoul, Korea) and Gangneung Asan Medical
and 5 reference genes in tumor tissues, providing a recurrence Center (Gangneung, Korea). Among these subjects, HR (+), HER2
score (RS) ranging from 0 to 100. This score is used to predict (­–), and pathologically node-positive patients were included.
the future risk of recurrence and the potential benefits of CTx. Inclusion criteria were refined for patients who underwent
MammaPrint (MMP; Agendia) analyzes 70 genes to evaluate the genomic profiling tests such as ODX and MMP. A total of 311
risk of distant metastasis within 10 years in early-stage breast patients were included in this retrospective study to predict
cancer. This result is classified as low or high genetic risk. In GET results using traditional clinicopathological factors. The
the case of the low-risk group, they can expect a good prognosis medical database cataloged patient characteristics, including
without CTx, even if they are classified as clinically high risk age at diagnosis, body mass index (BMI), and menopausal
due to ALN metastasis. status. Additionally, it documented the type and results of GET.
Recently, randomized controlled trials have been published Furthermore, it incorporated postoperative pathological results,
using ODX and MMP in patients with breast cancer with encompassing elements such as histological type, histological
ALN metastases. As a result of these studies, it has been grade (HG), the size of invasive breast cancer, the number of
demonstrated that CTx can be omitted in certain patients with lymph node metastases, the ER expression, PR expression,
HR (+), HER2 (−), and ALN metastases who are classified as HER2 expression, and the Ki-67 index. HG was assessed using
clinically high risk. Based on the results of the randomized the modified Bloom-Richardson grading system. Based on the
controlled trial, these gene expression tests (GETs) have become policy of each institution, tumors were classified as positive for
standard tools in treatment decision-making and are included ER and PR if they demonstrated ≥1% of nuclear-stained cells or
in the clinical guidelines of both the American Society of had a score of 3 or more according to the Allred scoring method.
Clinical Oncology and the National Comprehensive Cancer ODX risk groups were stratified into low risk, with RSs ranging
Network [10-13]. from 0 to 25, and high risk, with RSs ranging from 26 to 100.
GET is an important part of decision-making in breast cancer The risk groups for MMP were determined based on the results
treatment; however, several disadvantages exist. In the Korean of the MMP test reports.
healthcare environment, national health insurance does not
cover the cost of these tests, which adds a financial burden to Data and statistical analysis
patients already bearing the cost of breast cancer treatment. GET results were classified as low or high risk. In cases
Additionally, the time-consuming process of these tests can where clinical pathological findings are continuous variables,
delay the initiation of systemic adjuvant CTx for patients until they are converted into binary or multinomial variables for
these results become available. GET is an essential tool to analysis based on medical evidence or distribution. For binary
progress toward tailored treatment by identifying patients who or multinomial variables, according to the results of GET, a chi-
could potentially forego CTx. However, the high cost and the square test or Fisher exact test was performed for comparison.
potential for delay of CTx are major barriers. For these reasons, We reviewed the patient count annually and divided the data
utilizing clinicopathological values to predict the outcomes of into 2 segments: from 2015 to the year when the cumulative
gene expression assays could prove beneficial. This study aimed number of patients reached 75% and the period after that. A
to develop and validate a model for predicting the outcomes predictive model was developed using 75% of the cumulative

Annals of Surgical Treatment and Research 361

 Annals of Surgical Treatment and Research 2023;105(6):360-368

patient data, and the remaining 25% was utilized for validation. exist between the 2 groups when categorized based on an
Within the development cohort, linear regression was age threshold of 50 years and menopausal status (52.3 ± 11.1
performed to investigate patient characteristics and traditional years vs. 51.73 ± 9.1 years, P = 0.658). A higher rate of subjects
clinical pathological indicators related to high-risk groups. To with a BMI of 25 kg/m2 or above was observed in the high-
make it simple and practical, points were assigned to each risk group. The histologic grade (17.6% vs. 2.7%, P < 0.001) and
variable as rounded-off integer values based on the estimation nuclear grade (NG; 22.4% vs. 7.1%, P < 0.001) were higher, and
values from the multivariate analysis. The simplified scoring the size of invasive breast cancer was larger (2.0 ± 0.8 cm vs .
system was validated both internally and externally with 2 1.6 ± 0.6 cm, P < 0.001) in the high-risk group. However, there
cohorts. We analyzed the receiver operating characteristic curve were no differences between the 2 groups regarding axillary
and determined the area under the curve (AUC). All statistical nodal burden, lymphovascular invasion, or multicentricity.
analyses were conducted using SAS software ver. 9.3 (SAS The ER score was predominantly high in both groups, with no
Institute) and R software ver. 3.1.1 (The R Foundation), with significant differences observed. However, the high-risk group
P-values less than 0.05 deemed significant. had a greater proportion of individuals with low PR expression
(48.2% vs. 27.4%, P < 0.001), HER2 2 (+) status (41.2% vs. 24.3%,
RESULTS P = 0.012), and high Ki-67 (38.8% vs. 15.9%, P < 0.001).

Characteristics of patients Development of a simplified risk scoring system

The risk classification was based on an ODX RS of 25 and Between January 2015 and December 2022, 311 patients
the reported results for MMP. Characteristics according to the
GET results are shown in Table 1. No significant differences
Table 1. Continued
Low-risk High-risk
Variable P-value
group group
Table 1. Clinicopathological factors of the patient cohort (n
= 311) TNM stage <0.001
IB 55 (24.3) 12 (14.1)
Low-risk High-risk
Variable P-value IIA 121 (53.5) 32 (37.6)
group group
IIB 50 (22.1) 41 (48.2)
No. of patients 226 85 Lymphovascular invasion 0.509
Gene expression test <0.001 Absent 187 (82.7) 71 (83.5)
Oncotype DX 95 (42.0) 11 (12.9) Present 39 (17.3) 14 (16.5)
MammaPrint 131 (58.0) 74 (87.1) Histologic grade <0.001
Age (yr) 0.533 1 90 (39.8) 21 (24.7)
≤50 107 (47.3) 40 (47.1) 2 130 (57.5) 49 (57.6)
>50 119 (52.7) 45 (52.9) 3 6 (2.7) 15 (17.6)
Body mass index (kg/m2) 0.093 Nuclear grade <0.001
<23 98 (43.4) 30 (35.3) 1 17 (7.5) 1 (1.2)
23–25 67 (29.6) 21 (24.7) 2 193 (85.4) 65 (76.5)
>25 61 (27.0) 34 (40.0) 3 16 (7.1) 19 (22.4)
Menopause 0.262 ER score 0.618
Premenopausal 120 (53.1) 41 (48.2) 1–20 or Alled score 3–6 2 (0.9) 1 (1.2)
Postmenopausal 106 (46.9) 44 (51.8) >20 or Alled score 7–9 224 (99.1) 84 (98.8)
Histology 0.002 PR score <0.001
IDC 183 (81.0) 80 (94.1) 1–20 or Alled score 3–6 62 (27.4) 41 (48.2)
ILC 29 (12.8) 1 (1.2) >20 or Alled score 7–9 164 (72.6) 44 (51.8)
Others 14 (6.2) 4 (4.7) HER2 0.012
Multicentricity 0.267 Negative 54 (23.9) 18 (21.2)
Unifocality 146 (64.6) 51 (60.0) 1+ 117 (51.8) 32 (37.6)
Multifocality/multicentricity 80 (35.4) 34 (40.0) Equivocal 55 (24.3) 35 (41.2)
Pathologic T stage <0.001 Ki-67 index <0.001
T1 176 (77.9) 44 (51.8) ≤20 190 (84.1) 52 (61.2)
T2 50 (22.1) 41 (48.2) >20 36 (15.9) 33 (38.8)
Pathologic N stage 0.429
Micromets 66 (29.2) 19 (22.4) Values are presented as number (%).
IDC, invasive ductal carcinoma; ILC, Invasive lobular carcinoma;
Macro 1, 2 142 (62.8) 60 (70.6)
ER, estrogen receptor; PR, progesterone receptor; HER2, human
Marco 3 18 (8.0) 6 (7.1) epidermal growth factor receptor 2.

362
 Kwang Hyun Yoon, et al: Traditional factors can predict gene expression result in ALN metastases

satisfied the inclusion criteria. A simplified risk scoring and between the T and TNM stages, utilizing the variance
system was developed based on 75% of the patient cohort, inflation factor value. A multivariate regression analysis was
which accounted for 252 patients (76.1%) up until June conducted after excluding the NG and T stages. The results of
2021. The remaining patients were used for validation. No the multivariate regression analysis indicated that T stage, HG,
differences exist in the clinicopathological factors between PR score, HER2 score, and Ki-67 were significant (Table 2). The
the development and validation cohorts. For simplicity in Hosmer-Lemeshow test yielded a chi-square (χ²) value of 2.686
clinical practice, multivariate factors were converted into binary with a P-value of 0.388. To assess the overall impact and for
variables based on medical evidence or distribution. Univariate ease of calculation, scores were assigned to these 5 variables
regression analysis was conducted on the development cohort, based on their β -coefficients. A simple scoring system was
and significant associations were observed between the created by assigning integer values to each variable, rounded up
outcome variable and several factors, including BMI, histology, considering their β-coefficients, and using 1 point as a reference
T stage, TNM stage, HG, NG, PR expression, HER2 score, and point for a PR score of 0.613. Therefore, a score of 3 was assigned
Ki-67. Multicollinearity was examined between NG and HG for HG 3, a score of 2 for pathologic T2 or above, and a score of

Table 2. Univariate and multivariate analyses of the predictive variables of gene expression test high risk on the
development cohort
Univariate analysis Multivariate analysis
Variable
OR (95% CI) P-value β-coefficient (SE) OR (95% CI) P-value

Body mass index (kg/m2)

≤25 Reference
>25 2.094 (1.165–3.765) 0.014
Histology
IDC Reference
ILC 0.079 (0.011–0.589) 0.013
Others 0.654 (0.209–2.047) 0.465
Pathologic T stage
T1 Reference Reference
T2 3.280 (1.933–5.566) <0.001 1.15 (0.30) 3.165 (1.754–5.721) <0.001
Pathologic N stage
Micrometastasis Reference
Macro 1, 2 1.468 (0.811–2.655) 0.205
Marco 3 1.158 (0.403–3.327) 0.785
TNM stage
IB Reference
IIA 1.222 (0.581–2.530) 0.608
IIB 3.758 (1.778–7.946) 0.001
Histologic grade
1, 2 Reference Reference
3 7.857 (2.936–21.023) <0.001 1.42 (0.57) 3.383 (1.102–10.379) 0.003
Nuclear grade
1, 2 Reference
3 3.778 (1.839–7.764) <0.001
PR score
>20 or Alled score 7–9 Reference Reference
1–20 or Alled score 3–6 2.294 (1.288–4.088) 0.005 0.613 (0.29) 1.863 (1.201–3.508) 0.036
HER2
Negative or 1+ Reference Reference
Equivocal 2.447 (1.383–4.331) 0.002 0.827 (0.30) 2.287 (1.266–4.132) 0.006
Ki-67
≤20 Reference Reference
>20 2.337 (1.258–4.344) <0.001 0.945 (0.33) 2.573 (1.339–4.946) 0.005

OR, odds ratio; CI, confidence interval; SE, standard error; IDC, invasive ductal carcinoma; ILC, invasive lobular carcinoma; PR,
progesterone receptor; HER2, human epidermal growth factor receptor 2.

Annals of Surgical Treatment and Research 363

 Annals of Surgical Treatment and Research 2023;105(6):360-368

1 for a lower PR (1–20 or Alled score 3–6), HER2 2 (+), and high the sensitivity was 0.580 (95% confidence interval [CI], 0.482–
Ki-67 (>20) (Table 3). The sum of each score becomes the total 0.665) (Table 6), specificity was 0.880 (95% CI, 0.843–0.912),
cumulative score, representing the probability of attaining GET positive predictive value (PPV) was 0.645 (95% CI, 0.537–0.740),
high risk. negative predictive value (NPV) was 0.847 (95% CI, 0.812–0.878),
and overall accuracy was 0.847 (95% CI, 0.730–0.919). In the
Model performance and external validation validation cohort, the sensitivity was 0.625 (95% CI, 0.409–
To identify the optimal cutoff point for the AUC, we analyzed 0.757), specificity was 0.930 (95% CI, 0.850–0.979), PPV was 0.769
Youden’s J statistic, which reached its peak at 3.6. To ascertain (95% CI, 0.504–0.931), NPV was 0.870 (95% CI, 0.795–0.915), and
the estimate of GET high risk, we utilized the logistic regression overall accuracy was 0.847 (95% CI, 0.730–0.919). The receiver
formula: 1 / [1 + EXP {−(−1.83 + 0.613 × point)}], to calculate operating characteristic curve’s AUC value for the training set
the estimate of GET high risk for each point (Table 4). When 3.6 was 0.837 (95% CI, 0.710–0.964), with a P-value of less than 0.001,
is used as the cutoff to separate into low and high groups, there indicating that the AUC value was statistically significant (Fig.
is a 65.1% probability that point 4 will be classified as high 1). Similarly, for the validation set, the AUC value was 0.743
through gene expression assays (Table 5). In the training cohort, (95% CI, 0.647–0.811), and it had a P-value of less than 0.001,
demonstrating statistical significance (Fig. 1).

Table 3. The risk score for each variable

DISCUSSION
Variable Point
ALN metastasis plays a crucial role in patients with breast
Pathologic T2 2
Histologic grade 3 3 cancer, and it is a supplementary indicator for administering
PR score (<20 or Alled score 3–6) 1 adjuvant systemic CTx following surgery and a prognostic
HER2 2 (+) 1 factor for poor survival rates [2]. In patients with HR (+), HER2
Ki-67 >20 1 (−) breast cancer, even when ALN metastasis is observed, a
The sum of each individual risk score determined the total risk
score.
PR, progesterone receptor; HER2, human epidermal growth Table 5. The cutoff value for GET risk of sensitivity and
factor receptor 2. specificity
Variable GET risk low GET risk high

Table 4. The estimated risk percentage at each point in the Development cohort (n = 252)
GET scoring system Low-risk score 161 29
High-risk score 22 40
Risk group Points total Estimate of GET high risk Validation cohort (n = 59)
Low risk 0 0.1382 Low-risk score 40 6
Low risk 1 0.2285 High-risk score 3 10
Low risk 2 0.3534 ODX cohort (n = 106)
Low risk 3 0.5022 Low-risk score 86 3
High risk 4 0.6507 High-risk score 9 8
High risk 5 0.7747 MMP cohort (n = 205)
High risk 6 0.8639 Low-risk score 115 32
High risk 7 0.9214 High-risk score 16 42
High risk 8 0.9558 GET, gene expression test.
GET, gene expression test. Cutoff value is 3.6.

Table 6. Performance of the risk scoring system in the training and validation cohorts
Cohort Sensitivity Specificity PPV NPV Accuracy

Training cohort 0.580 (0.482–0.665) 0.880 (0.843–0.912) 0.645 (0.537–0.740) 0.847 (0.812–0.878) 0.847 (0.730–0.919)
Validation cohort 0.625 (0.409–0.757) 0.930 (0.850–0.979) 0.769 (0.504–0.931) 0.870 (0.795–0.915) 0.798 (0.744–0.844)
ODX cohort 0.727 (0.421–0.922) 0.905 (0.870–0.928) 0.471 (0.272–0.597) 0.966 (0.928–0.990) 0.887 (0.823–0.927)
MMP cohort 0.568 (0.480–0.640) 0.878 (0.828–0.919) 0.724 (0.613–0.817) 0.782 (0.738–0.819) 0.766 (0.703–0.818)

GET, gene expression test; ODX, Oncotype DX (Genomic Health); MMP, MammaPrint (Agendia); PPV, positive predictive value; NPV,
negative predictive value.

364
 Kwang Hyun Yoon, et al: Traditional factors can predict gene expression result in ALN metastases

A B
1.0 1.0

0.8 0.8
Sensitivity

Sensitivity
0.6 0.6

0.4 0.4

0.2 0.2
AUC, 0.837 (95% CI, 0.710 0.964) AUC, 0.743 (95% CI, 0.647 0.811)
P < 0.001 P < 0.001

0 0.2 0.4 0.6 0.8 1.0 0 0.2 0.4 0.6 0.8 1.0
1-Specificity 1-Specificity

C D
1.0 1.0

0.8 0.8
Sensitivity

Sensitivity

0.6 0.6

0.4 0.4 Fig. 1. Receiver operating charac­

teristic curve of a simplified risk
0.2 0.2 scoring system. (A) Development
AUC, 0.898 (95% CI, 0.829 0.969) AUC, 0.738 (95% CI, 0.667 0.803)
cohort. (B) Validation cohort. (C)
P < 0.001 P < 0.001 Oncotype DX (Genomic Health)
cohort. (D) MammaPrint (Agendia)
0 0.2 0.4 0.6 0.8 1.0 0 0.2 0.4 0.6 0.8 1.0 cohort. AUC, area under the
1-Specificity 1-Specificity curve; CI, confidence interval.

good prognosis can be expected with surgery alone, rendering groups with HR (+), HER2 (−) breast cancer, which is predicted
adjuvant systemic CTx unnecessary. The key focus here is on to have a good prognosis, those with ALN metastasis tend to
the selection of patients who are suitable for these treatment be classified as clinically high risk and are often considered for
strategies. Accurate patient classification enables personalized CTx. Due to economic concerns, disparities in medical services
therapies while maintaining favorable oncologic outcomes. and unnecessary treatments should be avoided [14]. For these
Substantial evidence from randomized controlled trials has reasons, numerous studies have been conducted to predict
been discussed thus far. The focus of these studies was on genetic test results based on patient factors and traditional
ODX and MMP, and they have enabled the prediction of patient clinicopathological factors [15-18].
prognosis based on the results of the gene signature [7,8]. Most previous studies primarily focused on ODX or MMP.
Genomic signatures have been incorporated into the Moreover, they were conducted on all patients who underwent
decision-making tools within the guidelines of major medical these tests. While a few studies measured the performance
and oncology societies, such as the American Society of Clinical regarding the major outcome of distant recurrence in the
Oncology, the National Comprehensive Cancer Network, the patient population, our research focused on the results of the
St. Gallen Consensus Conference, and the European Society 2 genetic tests. Therefore, an analysis of the RxPONDER and
for Medical Oncology. However, it is challenging to administer MINDACT trial results is required to apply our research in a
these tests across the board. Financial burdens and issues clinical setting. Through the RxPONDER trial, ODX has been
include delays in adjuvant systemic treatment while waiting for recognized as a tool to prognosticate oncologic outcomes and
the test results. In cases where patients have limitations with predict CTx responses in patients with ALN metastases. In the
these tests, the decision to proceed with CTx is determined case of the RxPONDER trial, all patients had ALN metastases.
by factors such as anatomical stage, breast cancer subtype, However, patients aged under 50 years constituted 21.5% of
Eastern Cooperative Oncology Group status, patient age, and the total. The research results indicated that CTx could be
the preferences of the patients or physicians. Even in patient omitted in cases with an RS of less than 25. However, caution

Annals of Surgical Treatment and Research 365

 Annals of Surgical Treatment and Research 2023;105(6):360-368

was needed in the subgroup analysis for patients aged under adjuvant CTx without testing. As such, it is thought that these
50 years. MMP is recognized as a tool for estimating survival patients were likely not included in our study. Physicians at
rates after surgery. Of the 1,550 patients in the MINDACT trial, the 2 institutions conducting the research believe that if ER
47.6% had ALN metastases. According to the recent update on is low, most patients may need CTx. Therefore, there are not
the long-term results of MINDACT, in patients with clinically many patients with low ER in our study. Even excluding ER,
high risk/genetically low risk, there is an absolute benefit of 5% the degree of PR expression was a significant figure in this
± 2.8% in the distant metastasis survival rate with additional study that included both GET. In previous studies examining
CTx. Due to these results, there is a need to be more cautious ODX, PR has been reported as a strong factor that can explain
regarding omitting CTx in patients. up to 23% of the variability in total ODX scores. Additionally,
We conducted a study targeting patients with breast cancer in studies related to PR and breast cancer prognosis, the degree
with ALN metastasis who underwent ODX or MMP. The objective of PR expression has been reported to be associated with the
was to investigate the correlation between the results of GET prognosis of breast cancer [19].
and traditional clinical and pathological factors. In this study, the Our study has several limitations. First, as it was a
multigene analysis revealed a strong association between high retrospective study, selection bias may exist. As a study design
risk and the following factors: pathologic T2, HG 3, PR score (<20 limitation, our cohort has a low proportion of low ER subjects.
or Allred score 3–6), HER2 2 (+), and Ki-67 of >20. Furthermore, Moreover, the rate at which each test was used was influenced
a simple and efficient scoring system was developed using the by the timing of the release of the RxPONDER and MINDACT
estimation of β-coefficients values for each variable. This scoring trials. In other words, the rate of test implementation varies
system allows for easy and rapid assessment of individuals based slightly by year. Furthermore, conducting research solely based
on the identified factors, enabling a practical approach to risk on test results without including actual patient survival data
stratification in clinical settings. In our study, all subjects had requires caution in interpreting and applying the results in
lymph node metastases; approximately 47% were aged under 50 a clinical setting. Caution is necessary when applying and
years, and approximately 65% used MMP for the GET. This study interpreting our study’s results for women aged under 50 years
only confirms the test results without survival data, such as based on the findings of the RxPONDER trial. Similarly, careful
distant metastases. consideration is essential when deciding to omit anticancer
According to previous studies predicting MMP results, there treatment based on the results of the MINDACT trial. Our study
was a negative correlation between MMP results and age. Our is meaningful because it includes the 2 most widely used tests,
study included only patients with ALN metastases among ODX and MMP, and focuses on patients with ALN metastases.
those who underwent MMP. Moreover, it included patients Additionally, we utilized easily assessable traditional clinical
who had undergone ODX, and there was no correlation with pathological factors in breast cancer patients and assigned
age. However, as previously described, caution is needed corresponding scores without complex tools. These scores
in patients aged under 50 years, as with the results of the can be calculated easily and quickly without sophisticated
RxPONDER trial. In actual clinical practice, the extent of ALN instruments. Furthermore, according to performance validation,
metastasis acts as a powerful factor in predicting prognosis. our study showed a high NPV. Thus, our study demonstrates
Macrometastases have worse outcomes than micrometastases, that it can be an excellent tool for effectively identifying
and the need for adjuvant CTx increases as the number of individuals with a high likelihood of being predicted as
lymph node metastases increases. However, in our study, the negative by the GET, even if they have ALN metastases.
number of metastases, and micrometastasis or macrometastasis We developed a simple scoring system to predict GET
did not correlate with the test results. According to this study, results in patients with HR (+), HER2 (−), and ALN metastases.
in breast cancer patients with HR (+), HER2 (−), and ALN (+), According to our study, a score of 3 was assigned for HG 3.
the presence or absence of micrometastasis and the number of Scores of 2 were given for pathologic T2 or above, while scores
lymph node metastases do not affect the results of the GET. of 1 were assigned for PR (1–20 or Alled score 3–6), HER2 2 (+),
In our study, the proportion of patients with a low ER status and high Ki-67 (>20). The sum of each score becomes the total
was tiny, at just 1%. Therefore, verifying any correlation between cumulative score, representing the probability of attaining GET
the ER score and the results of GET was difficult. As Korean high risk. If the total cumulative score is 4 points, there is a 65%
health insurance does not cover the GET and is expensive, probability that the test result will be classified as high risk. In
careful consideration must be given when selecting patients for cases with a lower score, omitting the costly decision-making
testing. Patients with low ER are likely to have a poor response test may be possible.
to hormone therapy. Moreover, they have been diagnosed
with ALN metastases. Due to the physician’s preference, they
may have undergone neoadjuvant CTx or directly received

366
 Kwang Hyun Yoon, et al: Traditional factors can predict gene expression result in ALN metastases

ACKNOWLEDGEMENTS Yujin Kim: [Link]

Jee Hyun Ahn: [Link]
Fund/Grant Support Jee Ye Kim: [Link]
This research was supported by a grant of the medical data- Hyung Seok Park: [Link]
driven hospital support project through the Korea Health Seung Il Kim: [Link]
Information Service (KHIS), funded by the Ministry of Health & Seho Park: [Link]
Welfare, Republic of Korea.
Author Contribution
Conflict of Interest Conceptualization, Methodology: SP
No potential conflict of interest relevant to this article was Resources: JHA, JYK, HSP, SIK, SP
reported. Investigation: KHY, SJL, YK
Formal Analysis: SJL
ORCID iD Project Administration: SIK
Kwang Hyun Yoon: [Link] Writing – Original Draft: KHY
Suk Jun Lee: [Link] Writing – Review & Editing: All authors

REFERENCES

1. Chen C, Dhanda R, Tseng WY, Forsyth M, Waheed S, Stacey-Clear A. Lymph node patients with breast cancer treated with
Patt DA. Evaluating use characteristics for ratio can further stratify prognosis in anastrozole or tamoxifen: a TransATAC
the oncotype dx 21-gene recurrence score subpopulations of breast cancer patients study. J Clin Oncol 2010;28:1829-34.
and concordance with chemotherapy use with axillary nodal metastases. Future 12. Habel LA, Shak S, Jacobs MK, Capra A,
in early-stage breast cancer. J Oncol Pract Oncol 2013;9:1425-31. Alexander C, Pho M, et al. A population-
2013;9:182-7. 7. Cardoso F, van’t Veer LJ, Bogaerts J, Slaets based study of tumor gene expression and
2. Fisher B, Bauer M, Wickerham DL, L, Viale G, Delaloge S, et al. 70-Gene risk of breast cancer death among lymph
Redmond CK, Fisher ER, Cruz AB, et al. signature as an aid to treatment decisions node-negative patients. Breast Cancer Res
Relation of number of positive axillary in early-stage breast cancer. N Engl J Med 2006;8:R25.
nodes to the prognosis of patients with 2016;375:717-29. 13. Andre F, Ismaila N, Allison KH, Barlow
primary breast cancer: an NSABP update. 8. Kalinsky K, Barlow WE, Gralow JR, Meric- WE, Collyar DE, Damodaran S, et al.
Cancer 1983;52:1551-7. Bernstam F, Albain KS, Hayes DF, et al. Biomarkers for adjuvant endocrine and
3. Elston C W, El l is IO. Pat holog ic a l 21-Gene assay to inform chemotherapy chemotherapy in early-stage breast
prognostic factors in breast cancer. I. benefit in node-positive breast cancer. N cancer: ASCO guideline update. J Clin
The value of histological grade in breast Engl J Med 2021;385:2336-47. Oncol 2022;40:1816-37.
cancer: experience from a large study 9. Caparica R, Brandão M, Piccart M. 14. Lim H, Kim SI, Hyun S, Lee GB, Seol
with long-term follow-up. Histopathology Systemic treatment of patients with early A, Lee M. Uptake rate of risk-reducing
1991;19:403-10. breast cancer: recent updates and state of salpingo-oophorectomy and surgical
4. Colditz GA, Rosner BA, Chen WY, Holmes the art. Breast 2019;48 Suppl 1:S7-20. outcomes of female germline BRCA1/2
MD, Hankinson SE. Risk factors for 10. National Comprehensive Cancer Network mutation carriers: a retrospective cohort
breast cancer according to estrogen and (NCCN). NCCN Guidelines version 4.2023: study. Yonsei Med J 2021;62:1090-7.
progesterone receptor status. J Natl Breast Cancer [Internet]. NCCN; 2023 15. Eaton AA, Pesce CE, Murphy JO, Stempel
Cancer Inst 2004;96:218-28. [cited 2023 Jul 1]. Available from: https:// MM, Patil SM, Brogi E, et al. Estimating
5. Choi H, Ahn SG, Bae SJ, Kim JH, Eun NL, [Link]/professionals/physician_ the OncotypeDX score: validation of
Lee Y, et al. Comparison of programmed gls/pdf/[Link]. an inexpensive estimation tool. Breast
cell death ligand 1 status between core 11. Dowsett M, Cuzick J, Wale C, Forbes J, Cancer Res Treat 2017;161:435-41.
needle biopsy and surgical specimens of Mallon EA, Salter J, et al. Prediction of 16. Lee SB, Kim J, Sohn G, Kim J, Chung IY,
triple-negative breast cancer. Yonsei Med J risk of distant recurrence using the 21- Kim HJ, et al. A nomogram for predicting
2023;64:518-25. gene recurrence score in node-negative the oncotype DX recurrence score in
6. L aw n A M, Fr a mp t on A E , K rel l J, and node-positive postmenopausal women with T1-3N0-1miM0 hormone

Annals of Surgical Treatment and Research 367

 Annals of Surgical Treatment and Research 2023;105(6):360-368

receptor positive, human epidermal 18. Kim MC, Kwon SY, Choi JE, Kang SH, DS, Park YH, et al. Progesterone receptor
growth factor 2 (HER2) negative breast Bae YK. Prediction of oncotype DX expression level predicts prognosis of
cancer. Cancer Res Treat 2019;51:1073-85. recurrence score using clinicopathological estrogen receptor-positive/HER2-negative
17. Lee YJ, Hwang YS, Kim J, Ahn SH, Son BH, variables in estrogen receptor-positive/ young breast cancer: a single-center
Kim HJ, et al. A nomogram for predicting human epidermal growth factor receptor prospective cohort study. Cancers (Basel)
probability of low risk of MammaPrint 2-negative breast cancer. J Breast Cancer 2023;15:3435.
results in women with clinically high-risk 2023;26:105-16.
breast cancer. Sci Rep 2021;11:23509. 19. Kwak Y, Jang SY, Choi JY, Lee H, Shin

368