The effect of the tumor-to-skin distance on axillary lymph node metastasis in breast cancer

SUMMARY

OBJECTIVE:

Tumor-to-skin distance is known to have an effect on axillary lymph node metastasis but has no clinical use with nomograms. This study aimed to investigate the effect of tumor-to-skin distance on axillary lymph node metastasis alone and in combination with nomogram for clinical use.

METHODS:

A total of 145 patients who underwent surgery for breast cancer (T1–T2 stage) and whose axillary lymph nodes were evaluated (axillary dissection or sentinel lymph node biopsy) between January 2010 and December 2020 were included in the study. Tumor-to-skin distance and other pathological data of the patients were evaluated.

RESULTS:

Of the 145 patients, 83 (57.2%) had metastatic lymph nodes in the axilla. Tumor-to-skin distance was different in terms of lymph node metastasis (p=0.045). In the receiver operating characteristic curve for tumor-to-skin distance, area under curve was 0.597 (95%CI 0.513–0.678, p=0.046), area under curve of the nomogram was 0.740 (95%CI 0.660–0.809), p<0.001) and nomogram+tumor-to-skin distance was 0.753 (95%CI 0.674–0.820), p<0.001). No statistical difference was found for axillary lymph node metastasis between the nomogram+tumor-to-skin distance and the nomogram alone (p=0.433).

CONCLUSION:

Although tumor-to-skin distance demonstrated a significant difference in axillary lymph node metastasis, it had a poor association with an area under curve value of 0.597 and did not produce a significant improvement in predicting lymph node metastasis when combined with the nomogram. The tumor-to-skin distance may be unlikely to enter clinical practice.

Keywords
Axilla; Breast; Lymph nodes; Lymphatic metastasis; Neoplasms

INTRODUCTION

Breast cancer is the most common type of cancer among women worldwide. According to the World Health Organization, there were 2.3 million new diagnoses and 685,000 deaths in 2020¹. Regional metastases to lymph nodes and surrounding tissues and distant metastases to organs such as bone, liver, and brain are considered to be among the factors that adversely affect the prognosis in breast cancer²,³. Evaluation of axillary lymph node metastases in breast cancer is important in planning the patient’s treatment⁴.

The most commonly used methods to evaluate axillary lymph nodes are preoperative Tru-Cut biopsy or intraoperative sentinel lymph node biopsy (SLNB)⁵. Both of these techniques are invasive and require a pathologist. Therefore, noninvasive efforts have been tried to be developed to detect metastatic lymph nodes. As part of these efforts, studies have been conducted to examine the relationship between tumor-to-skin distance (TSD) and metastatic axillary lymph nodes in breast cancer. The rate of lymph node metastasis was found to be high in tumors close to skin⁴,⁶,⁷,⁸,⁹,¹⁰. Although the results of these studies presented similar characteristics, there has been no use of TSD in clinics or nomograms¹¹,¹². This has led us also to investigate the diagnostic value of TSD.

The present study aimed to examine the area under curve value of TSD alone and in combination with nomogram and evaluate their potential use in clinical practice for axillary lymph node metastasis in breast cancer.

METHODS

Ethical approval

This study was conducted with the approval of Ethics Committee for Non-Interventional Research (No. 2021/02-18, Date: 04.02.2021) of the Fırat University.

Patient selection

Between January 2010 and December 2020, 177 female patients who underwent surgery for T1–T2 stage breast cancer were selected using the hospital data system. The exclusion criteria were as follows:

• Having neoadjuvant chemoradiotherapy,

• T3–T4 stage tumor,

• Paget’s disease,

• Recurrence of disease,

• Skin invasion,

• Multicentric cancer,

• Previous breast surgery in the studied breast,

• Skin-sparing mastectomy or excisions,

• Failure to investigate axillary lymph node status,

• Diseases that may cause alterations in the breast tissue (chronic heart/kidney disease).

This study included 145 female patients aged between 18 and 90 years who did not meet any of the exclusion criteria. The patients were classified into two groups based on having axillary lymph node metastasis. Group 1 comprised axillary nonmetastatic (N0) patients, and group 2 comprised metastatic (N+) patients. Memorial Sloan Kettering Cancer Center nomogram was used for prediction of axillary lymph node metastasis¹². Demographic and histopathological data were recorded for evaluation.

Pathological and immunohistochemical examination

All resection samples were examined by the pathology department of our institution based on standard clinical protocols. Immunohistochemistry was performed using 4-μm-thick histological tissue slides obtained from the paraffin blocks. The Olympus Microscope Digital Camera model DP71 (Olympus Co., Shinjuku, Tokyo, Japan) software imaging system was used for histological analysis of estrogen, and progesterone status was defined as positive when 10% or more of nuclei showed positive staining. For HER2 status, tumors with IHC staining of 3+ (uniform, intense membrane staining of 30% of invasive tumor cells) were considered HER2-positive. Cases with an IHC staining of 2+ were considered positive if they turned out to be positive in subsequent HER2/neu gene amplification (fluorescence in situ hybridization). HER2 1+ status and the absence of staining were considered negative. Breast cancer is divided into five molecular subtypes¹³. Histological grades were determined using the Modified Bloom-Richardson Nottingham Score index. TSD was measured macroscopically or under a light microscope.

Statistical analyses

Data analysis was performed using the SPSS 21.0 software (SPSS Inc., Chicago, IL, USA). Kolmogorov-Smirnov/Shapiro-Wilk test was applied to assess the distribution normality of the data. Nonparametric data were presented as median (minimum-maximum) and parametric data as mean±standard deviation. The independent sample t-test was applied for normally distributed data, the Mann–Whitney U test was used for non-normally distributed data, and the chi-square or Fisher’s exact test was used for nominal/ordinal data. Logistic regression analysis was performed to determine significant predictors of the axillary lymph node metastasis with univariate and multivariate analyses. Odds ratios (OR) with 95% confidence intervals were used. The optimal cutoff point of the TSD was evaluated by receiver operating characteristic (ROC) curve analysis. The comparison of nomogram with TSD+nomogram area under curves (AUCs) and the examination of the statistical significance of the AUCs were performed via the DeLong method¹⁴. An AUC-ROC >0.7 indicates a good discrimination model¹⁵. Binary logistic regression formula for TSD+nomogram was “Y=[(100−Nomogram)*(-0.04539)]+[TSD*(-0.03779)]+3.12636.” The statistical level of significance for all tests was considered to be <0.05.

RESULTS

A total of 145 female patients [62 (42.8%) in group 1 and 83 (57.2%) in group 2] were included in the study. The demographic characteristics of the patients are presented in Table 1. There was a significant difference between the groups for TSD [20 mm in group 1 (1–55) and 15 mm in group 2 (2–45), p=0.045). A significant difference was observed between the groups for LVI and tumor grade (p<0.001 and p=0.008, respectively). No significant difference was observed between groups in terms of other data (p>0.05).

Table 1.
Distribution of data within groups.

Variables		Group 1 (n=62) mean±std/median (min–max)	Group 2 (n=83) mean±std/median (min–max)	Total (n=145)	OR	p-value
Age		60.29±1.586	60.78±1.257	60.57±0.986		0.806a
Tumor-to-skin distance (mm)		20 (1–55)	15 (2–45)	17 (1–55)		0.045*
Tumor diameter (mm)		26 (7–50)	27 (9–47)	26 (7–50)		0.307*
Tumor site	Right	32	45	77	1.05	0.756**
	Left	30	38	68	0.95
Quadrant	UOQ	27	39	66		0.970***
	LOQ	8	11	19
	LIQ	8	12	20
	UIQ	17	19	36
	Central	2	2	4
ER	Negative	10	14	24	0.95	0.906**
	Positive	52	69	121	1
PR	Negative	9	18	27	0.67	0.272**
	Positive	53	65	118	1.09
HER2	Negative	39	48	87	1.09	0.537**
	Positive	23	35	58	0.88
Luminal types	Type A	5	11	16		0.698***
	Type B HER2-negative	29	30	59
	Type B HER2-positive	19	30	49
	HER2-positive	4	5	9
	Triple negative	5	7	12
Menopausal status	Premenopausal	17	18	35	1.26	0.425**
	Postmenopausal	45	65	110	0.927
Grade	1	27	19	46	1.90	0.008**
	2/3	35	64	99	2.598
Lymphovascular invasion	Negative	43	11	54	0.36	<0.001**
	Positive	19	72	91	4.43
Histological type of cancer	IDC	47	65	112		0.646***
	ILC	6	10	16
	Mix type and others	9	8	17

• ^a Independent sample t-test; *Mann-Whitney U test; **Chi-square test; ***Fisher’s exact test. UOQ: upper outer quadrant; LOQ: lower outer quadrant; LIQ: lower inner quadrant; UIQ: upper inner quadrant; IDC: invasive ductal cancer; ILC: invasive lobular cancer. Bold values indicate statistical significance at the p<0.05 level.

Univariate analysis showed that the TSD, LVI, nomogram, and grade were significantly associated with a higher risk of metastasis (p<0.05). In multivariate analysis, TSD (p=0.015) and LVI (p<0.001) were the independent risk factors of axillary lymph node metastasis. Nomogram was not entered into multivariate analysis owing to multicollinearity.

In the ROC curve analysis, the cutoff value of TSD for lymph node metastasis was calculated to be 21 mm (sensitivity 77.1%, specificity 38.7%, p=0.046). The cutoff value of the nomogram for lymph node metastasis was calculated to be 52.5 (sensitivity 69.9%, specificity 72.6%, p<0.001), and nomogram+TSD was 0.404 (sensitivity 89.1%, specificity 48.4%, p<0.001) (Figure 1). The comparison of the nomogram and TSD+nomogram curves indicated that the difference was not significant (p=0.433) (Table 2).

Figure 1.
Areas under the receiver operating characteristic curve for comparisons of tumor-to-skin distance, nomogram, and tumor-to-skin distance+nomogram. The area under curve of the tumor-to-skin distance, nomogram, and nomogram+tumor-to-skin distance were 0.597 (95%CI 0.513–0.678), 0.740 (95%CI 0.660–0.809), and 0.753 (95%CI 0.674–0.820), respectively.

Table 2.
Comparison of groups’ discriminative abilities for axillary lymph node metastasis.

Comparison of groups	Difference between AUCs	Significance level (p)
Nomogram vs. TSD	0.143	0.024
Nomogram vs. nomogram+TSD	0.0128	0.433
TSD vs. nomogram+TSD	0.155	0.003

• TSD: tumor-to-skin distance; AUC: area under curve. Bold values indicate statistical significance at the p<0.05 level.

DISCUSSION

This study aimed to examine the effect of TSD on axillary lymph node metastasis in breast cancer and its feasibility in clinical practice. According to the results of the study, although the TSD differed significantly between the groups, it was not superior to the nomogram, and it was not effective when used together with the nomogram. Low AUC-ROC curve value of TSD and lack of a significant difference for its combined use with nomogram compared to nomogram alone indicated that it was an inefficient parameter for clinical use in breast cancer.

Axillary lymph node metastasis plays an important role in the prognosis of breast cancer, such as the surgical margin¹⁶,¹⁷. Therefore, the most reliable method for axillary lymph node involvement today is the SLNB examination¹⁸. There have been efforts to predict the status of breast regional lymph nodes using other methods without resorting to SLNB¹¹,¹²,¹⁹,²⁰,²¹,²². TSD is one of them. This method was reported to be associated with the tumor’s proximity to dermal lymphatics²³.

Studies reported a significant relationship between TSD and axillary lymph node metastasis⁴,⁶–¹⁰. Only a study by Lee et al., reported no significant difference for TSD²⁰. The studies indicated different cutoff values in the range of 3–14 mm⁴,⁶–¹⁰. The lack of a clear cutoff value complicates the clinical use of TSD. In the present study, cutoff value was found at 21 mm. In tumors close to the skin, the rate of lymph node metastasis was higher than that in tumors far from the skin.

In the present study, the AUC-ROC value of TSD was <0.7. Therefore, the discrimination ability was categorized as weak¹⁵. In addition, the low sensitivity and specificity of the cutoff value is a reflection of this weak effect. In our study, although there was a small increase in AUCs in nomogram+TSD compared to nomogram, no significant difference was observed between the groups in the DeLong method. Nomogram alone indicated a significant statistical superiority to the TSD. Therefore, TSD may only be a statistical parameter for predicting axillary lymph node metastasis and it may not be clinically useful. Torstenson et al., reported that TSD measurement increased the AUC-ROC from 0.71 to 0.75 on the nomogram¹⁰. However, this study made no mentioning of the pairwise comparison of ROC curves analysis that was performed for the AUC-ROC values. Therefore, it may not be possible to make a statistical comment about the AUC-ROC values. The present study is the first to compare the AUC-ROC values of nomogram with nomogram+TSD combination in the literature.

Different measurement methods can affect the results of studies. While USG was used in many studies, Ojha et al., used pathology specimens⁴,⁶–¹⁰. The inability of radiologists to stabilize probe pressure while measuring with USG may have led to varying results. In our study, pathological specimens were used for the measurement.

Hormonal status and T stage are important factors that affect the treatment of breast cancer. In the present study, hormonal status of the tumor did not indicate any difference in groups. There are opposing views on hormonal status and tumor diameter⁴,⁶–¹⁰. Song et al., reported no difference between groups for the luminal types, while a study reported difference between luminal A and triple negative¹³,²⁴. In our study, no difference was observed among the luminal types for groups. Eom et al.,¹⁰ did not report any difference among patient groups with different tumor sizes for lymph node metastases, while other studies found varying rates of difference⁴,⁶–⁸. In the present study, no relationship was found between lymph node metastasis and tumor diameter.

Lymphovascular invasion affects tumor aggressiveness and lymph node metastasis. Eom et al., found no association between lymph node metastasis and LVI⁸. However, several other studies and the present study found a significant correlation between LVI and axillary lymph node metastasis⁴,⁷,¹⁰. In the present study, LVI was the independent risk factor for axillary lymph node metastasis. Moreover, there was a significant difference in grade between groups. The grade was high in the axillary metastatic group. Similar to our findings, Cunningham et al.⁷ reported higher lymph node metastasis in higher grades while other studies reported no difference¹⁰,²⁴,²⁵.

Among the study limitations are the retrospective and single-center nature of the study. Besides, the TSD measurements were conducted by multiple pathologists, and on a limited number of patients, which may have affected the study results.

CONCLUSION

The effect of TSD in predicting axillary lymph node metastasis is poor due to the low AUC value. Moreover, TSD was not superior to the nomogram. Due to the low AUC value of TSD and lack of a significant difference between nomogram+TSD and nomogram AUC-ROC curves, the TSD may not constitute a useful parameter for clinical use in breast cancer.

REFERENCES

• ^1.
  World Health Organization. Breast cancer. Available from: https://www.who.int/news-room/fact-sheets/detail/breast-cancer
  » https://www.who.int/news-room/fact-sheets/detail/breast-cancer
• ^2.
  Robertson JF, Dixon AR, Nicholson RI, Ellis IO, Elston CW, Blamey RW. Confirmation of a prognostic index for patients with metastatic breast cancer treated by endocrine therapy. Breast Cancer Res Treat. 1992;22(3):221-7. https://doi.org/10.1007/bf01840835
  » https://doi.org/10.1007/bf01840835
• ^3.
  Gomes Cunha JP, Goncalves R, Silva F, Aguiar FN, Mota BS, Chequim BB, et al. Validation of the Residual Cancer Burden Index as a prognostic tool in women with locally advanced breast cancer treated with neoadjuvant chemotherapy. J Clin Pathol. 2021:jclinpath-2021-207771. https://doi.org/10.1136/jclinpath-2021-207771
  » https://doi.org/10.1136/jclinpath-2021-207771
• ^4.
  Ansari B, Morton MJ, Adamczyk DL, Jones KN, Brodt JK, Degnim AC, et al. Distance of breast cancer from the skin and nipple impacts axillary nodal metastases. Ann Surg Oncol. 2011;18(11):3174-80. https://doi.org/10.1245/s10434-011-1957-z
  » https://doi.org/10.1245/s10434-011-1957-z
• ^5.
  Wang M, Zhou W, Zhao Y, Xia T, Zha X, Ding Q, et al. A novel finding of sentinel lymphatic channels in early stage breast cancer patients: which may influence detection rate and false-negative rate of sentinel lymph node biopsy. PLoS One. 2012;7(12):e51226. https://doi.org/10.1371/journal.pone.0051226
  » https://doi.org/10.1371/journal.pone.0051226
• ^6.
  Bae MS, Shin SU, Song SE, Ryu HS, Han W, Moon WK. Association between US features of primary tumor and axillary lymph node metastasis in patients with clinical T1-T2N0 breast cancer. Acta Radiol. 2018;59(4):402-8. https://doi.org/10.1177/0284185117723039
  » https://doi.org/10.1177/0284185117723039
• ^7.
  Cunningham JE, Jurj AL, Oman L, Stonerock AE, Nitcheva DK, Cupples TE. Is risk of axillary lymph node metastasis associated with proximity of breast cancer to the skin? Breast Cancer Res Treat. 2006;100(3):319-28. https://doi.org/10.1007/s10549-006-9256-2
  » https://doi.org/10.1007/s10549-006-9256-2
• ^8.
  Eom YH, Kim EJ, Chae BJ, Song BJ, Jung SS. The distance between breast cancer and the skin is associated with axillary nodal metastasis. J Surg Oncol. 2015;111(7):824-8. https://doi.org/10.1002/jso.23898
  » https://doi.org/10.1002/jso.23898
• ^9.
  Ojha S, Jain R, Nilkanthe R, Meenai F, Abhishek S, Amit H. Distance of tumor to skin as a predictive marker for axillary lymph node metastasis in cases of breast carcinoma - A retrospective study. Indian J Med Paediatr Oncol. 2018;39(3):321-5. https://doi.org/10.4103/ijmpo.ijmpo_26_17
  » https://doi.org/10.4103/ijmpo.ijmpo_26_17
• ^10.
  Torstenson T, Shah-Khan MG, Hoskin TL, Morton MJ, Adamczyk DL, Jones KN, et al. Novel factors to improve prediction of nodal positivity in patients with clinical T1/T2 breast cancers. Ann Surg Oncol. 2013;20(10):3286-93. https://doi.org/10.1245/s10434-013-3110-7
  » https://doi.org/10.1245/s10434-013-3110-7
• ^11.
  The University of Texas MD Anderson Cancer Center. Breast cancer nomogram to predict positive sentinel lymph nodes, without neoadjuvant chemotherapy. Available from: http://www3.mdanderson.org/app/medcalc/bc_nomogram3/index.cfm?pagename=sln
  » http://www3.mdanderson.org/app/medcalc/bc_nomogram3/index.cfm?pagename=sln
• ^12.
  Memorial Sloan Kettering Cancer Center. Breast cancer nomogram: sentinel lymph node metastasis. 2021 [cited on Dec 9, 2021]. Available from: http://nomograms.mskcc.org/Breast/BreastSLNodeMetastasisPage.aspx
  » http://nomograms.mskcc.org/Breast/BreastSLNodeMetastasisPage.aspx
• ^13.
  Song BI, Kim HW, Won KS. Predictive value of ¹⁸F-FDG PET/CT for axillary lymph node metastasis in invasive ductal breast cancer. Ann Surg Oncol. 2017;24(8):2174-81. https://doi.org/10.1245/s10434-017-5860-0
  » https://doi.org/10.1245/s10434-017-5860-0
• ^14.
  DeLong ER, DeLong DM, Clarke-Pearson DL. Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics. 1988;44(3):837-45. PMID: 3203132
• ^15.
  Silva FX, Parpinelli MA, Oliveira-Neto AF, Valle CR, Souza RT, Costa ML, et al. Prognostic value of an estimate-of-risk model in critically ill obstetric patients in Brazil. Obstet Gynecol. 2022;139(1):83-90. https://doi.org/10.1097/AOG.0000000000004619
  » https://doi.org/10.1097/AOG.0000000000004619
• ^16.
  Hassan RA, Maesaka JY, Ricci MD, Soares JM, Dória MT, Baracat EC, et al. Predictive factors for positive surgical margins in the treatment of breast ductal carcinoma in situ. J Cancer Res Ther. 2016;12(2):995-8. https://doi.org/10.4103/0973-1482.172135
  » https://doi.org/10.4103/0973-1482.172135
• ^17.
  Araujo RA, Cordero Luz FA, Costa Marinho E, Mendes TR, Nascimento CP, Ribeiro Delfino PF, et al. Operable breast cancer: how not to worsen the prognosis, especially in triple negative and stage II tumors. Surg Oncol. 2021;38:101596. https://doi.org/10.1016/j.suronc.2021.101596
  » https://doi.org/10.1016/j.suronc.2021.101596
• ^18.
  Lyman GH, Giuliano AE, Somerfield MR, Benson AB, Green S, Hamroun-Yazid L, et al. American Society of Clinical Oncology guideline recommendations for sentinel lymph node biopsy in early-stage breast cancer. J Clin Oncol. 2005;23(30):7703-20. https://doi.org/10.1200/JCO.2005.08.001
  » https://doi.org/10.1200/JCO.2005.08.001
• ^19.
  Bae SJ, Youk JH, Yoon CI, Park S, Cha CH, Lee HW, et al. A nomogram constructed using intraoperative ex vivo shear-wave elastography precisely predicts metastasis of sentinel lymph nodes in breast cancer. Eur Radiol. 2020;30(2):789-97. https://doi.org/10.1007/s00330-019-06473-5
  » https://doi.org/10.1007/s00330-019-06473-5
• ^20.
  Lee SE, Sim Y, Kim S, Kim EK. Predictive performance of ultrasonography-based radiomics for axillary lymph node metastasis in the preoperative evaluation of breast cancer. Ultrasonography. 2021;40(1):93-102. https://doi.org/10.14366/usg.20026
  » https://doi.org/10.14366/usg.20026
• ^21.
  Zhang X, Liu Y, Luo H, Zhang J. PET/CT and MRI for identifying axillary lymph node metastases in breast cancer patients: systematic review and meta-analysis. J Magn Reson Imaging. 2020;52(6):1840-51. https://doi.org/10.1002/jmri.27246
  » https://doi.org/10.1002/jmri.27246
• ^22.
  Piato JR, Filassi JR, Dela Vega AJ, Coura-Filho GB, Aguiar FN, Porciuncula LM, et al. SPECT-CT-guided thoracoscopic biopsy of sentinel lymph nodes in the internal mammary chain in patients with breast cancer: a pilot study. Innovations (Phila). 2016;11(2):94-8. https://doi.org/10.1097/IMI.0000000000000257
  » https://doi.org/10.1097/IMI.0000000000000257
• ^23.
  Chao C, Edwards MJ, Abell T, Wong SL, Simpson D, McMasters KM. Palpable breast carcinomas: a hypothesis for clinically relevant lymphatic drainage in sentinel lymph node biopsy. Breast J. 2003;9(1):26-32. https://doi.org/10.1046/j.1524-4741.2003.09107.x
  » https://doi.org/10.1046/j.1524-4741.2003.09107.x
• ^24.
  Oliveira Filho HR, Doria MT, Piato JR, Soares Junior JM, Filassi JR, Baracat EC, et al. Criteria for prediction of metastatic axillary lymph nodes in early-stage breast cancer. Rev Bras Ginecol Obstet. 2015;37(7):308-13. https://doi.org/10.1590/S0100-720320150005343
  » https://doi.org/10.1590/S0100-720320150005343
• ^25.
  Choong WL, Evans A, Purdie CA, Wang H, Donnan PT, Lawson B, et al. Mode of presentation and skin thickening on ultrasound may predict nodal burden in breast cancer patients with a positive axillary core biopsy. Br J Radiol. 2020;93(1108):20190711. https://doi.org/10.1259/bjr.20190711
  » https://doi.org/10.1259/bjr.20190711

• Funding: none.

Publication Dates

• Publication in this collection
  21 Apr 2023
• Date of issue
  2023

History

• Received
  12 Jan 2023
• Accepted
  20 Jan 2023