MATERIALS AND METHODS

Article search method

To evaluate the diagnostic value of superparamagnetic iron oxide MRI in determining axillary lymph node metastasis in breast cancer patients, the researchers searched studies that had been registered on PubMed, Cochrane Library, ScienceDirect, SpringerLink, and Ovid databases up until December 2013, and searched for articles including the keywords and titles “magnetic resonance imaging AND axilla” and “superparamagnetic iron oxide AND axilla.”

Standard for article selection

Three researchers (J.G.P., D.W.R. and Y.S.K) reviewed all searched articles and decided whether they were suitable for analysis. The articles included in this study must be 1) original articles, 2) articles that determined metastasis based on biopsy, 3) articles published in English, or 4) articles that report or can calculate true positive, false positive, false negative and true negative values. Among the searched articles, ones that are not original articles (reviews, poster presentations, case reports, etc.) and ones published in languages other than English were excluded.

Qualitative evaluation of articles

All researchers planned and reviewed this study together. In the case where there were different opinions over the selection of articles, discussions were held until agreement was reached. For qualitative evaluation of the analyzed articles, the “QUADAS tool (quality assessment of diagnostic accuracy studies tool)” was used.

Analysis goal

The goal of this study is to evaluate the diagnostic value of superparamagnetic iron oxide MRI. To this end, true positive, false positive, false negative and true negative values included in the selected articles were used to analyze the sensitivity and specificity of superparamagnetic iron oxide MRI.

Statistical analysis

The heterogeneity of articles included in the study was evaluated through Cochran's Q test and I²-statistics. It was considered that there was heterogeneity if P-value is lower than 0.1 in Cochran's Q test.²² Moreover, the level of heterogeneity was determined by dividing the range of I²-statistics into three ranges (< 25%: low, 25–75%: moderate, > 75%: high) of heterogeneity.²³ A random model effect was used when heterogeneity was confirmed. The results of analysis were quantified using the SROC (summary receiver operating characteristics) curve, and AUC (area under curve) was identified to evaluate the diagnostic value of superparamagnetic iron oxide MRI. Furthermore, a subgroup analysis was conducted to verify the cause of heterogeneity among studies.

All studies were statistically analyzed using Meta-Disc 1.4 (Clinical BioStatistics Unit, Hospital Universitario Ramón y Cajal, Madrid, Spain) and STATA 12.0 (STATA Co., College Station, Texas, United States), and it was considered statistically significant when the P-value was lower than 0.05.

RESULTS

Article search results

Figure 1 shows the flow chart for the article selection in this meta-analysis. A total of 346 articles were searched, 7 of which were ultimately selected and analyzed in this study. Redundant articles were determined by checking the author names, titles and contents.

Characteristics of analyzed articles and qualitative evaluation

Table 1 shows the quality and characteristics of articles included in this study. All included articles were published between 2002 and 2011. Moreover, all articles had a QUADAS score of 8 or higher, indicating that they were generally acceptable studies.

Diagnostic accuracy and SROC curve

In this meta-analysis, superparamagnetic iron oxide MRI had a sensitivity of 0.83 (95% confidence interval: 0.75 – 0.89) and specificity of 0.97 (95% confidence interval: 0.94–0.98) (Fig. 2).

Figure 3 shows the SROC curve of superparamagnetic iron oxide MRI, and AUC turned out to be 0.9563.

Heterogeneity of articles and subgroup analysis

Table 2 presents the heterogeneity of studies on superparamagnetic iron oxide MRI. Both sensitivity (P = 0.0062, I² index = 66.7%) and specificity (P = 0.0987, I² index = 43.8%) showed moderate heterogeneity.

Furthermore, Table 3 shows the results of subgroup analysis. This study proved that the surgical method (sentinel lymphadenectomy) had the highest diagnostic accuracy for axillary lymphnode metastasis (sensitivity = 1.00 (95% confidence interval: 0.16–1.00), specificity = 1.00 (95% confidence interval: 0.63–1.00). The authors attempted to conduct a subgroup analysis according to stage and patient age as well, but failed due to the lack of data.

Meta-regression analysis

To determine the cause of heterogeneity, the authors conducted a meta-regression analysis on the year of publication (before 2009 vs. after 2010), area of publication (East vs. West), magnetic field strength (1.5 tesla vs. 1.0 tesla), number of research subjects (less than 10 persons vs. 10 persons or more), and QUADAS score (lower than 10 points vs. 10 points or higher) (Table 4). In this meta-analysis, QUADAS score (P = 0.010, relative diagnostic odds ratio (DOR) = 0.351) turned out to be the factor related to heterogeneity.

DISCUSSION

This study analyzed the diagnostic value of superparamagnetic iron oxide MRI in determining axillary lymph node metastasis of breast cancer patients by analyzing sensitivity and specificity. The authors searched previous studies that conducted a histological diagnosis after superparamagnetic iron oxide MRI and analyzed the accuracy of the aforementioned diagnostic imaging, and ultimately analyzed 7 articles conducted on a total of 118 patients.

This meta-analysis searched and analyzed clinical studies published up until December 2013 and thus it does not include data published after that. The QUADAS tool was used for qualitative evaluation of the articles included in this meta-analysis.

Studies included in this meta-analysis had heterogeneity in both sensitivity and specificity. Thus, a meta-regression analysis was conducted to analyze the cause of such heterogeneity, and it was discovered that the QUADAS was the biggest cause of heterogeneity.

Moreover, a subgroup analysis was conducted on studies included in this meta-analysis to analyze the factors related to the accuracy of diagnosing axillary lymph node metastasis, and it was found that sentinel lymph node biopsy was the factor that had the highest accuracy in diagnosing axillary lymph node metastasis. However, only in 1 of the 7 articles was sentinel lymph node biopsy performed, which indicates that more research is needed.

In this study, sensitivity and specificity of superparamagnetic iron oxide MRI in determining axillary lymph node metastasis were 0.83 (95% confidence interval: 0.75 – 0.89) and 0.97 (95% confidence interval: 0.94 – 0.98) respectively, revealing a high diagnostic value, which can be identified through the AUC value using the SROC curve (AUC = 0.9563).

This study has a few limitations. First, even though the database was generally used in meta-analysis such as EMBASE, this study failed to check the database for which the organization has no search rights. However, the authors made the most of the database for which the organization has search rights, such as PubMed, Cochrane Library, ScienceDirect, SpringerLink, and Ovid databases. Moreover, they checked each of the references for the articles included in this meta-analysis to minimize loss of data. Second, most studies included in this meta-analysis included both “patient to patient” data and “lesion to lesion” data, but some studies only presented either “patient to patient” or “lesion to lesion” data. This study gave priority to “patient to patient” data in analysis. Therefore, studies with no “patient to patient” data were analyzed by applying “lesion to lesion” data, which may have affected the study results. However, in studies that presented both “patient to patient” data and “lesion to lesion” data, each of the data sets did not show much a numerical gap in sensitivity and specificity, and thus may not have had a crucial effect on the study results. However, more research is needed for more accurate analysis.

In conclusion, superparamagnetic iron oxide MRI may be an effective diagnostic imaging technique for stage setting and prognostic prediction by determining axillary lymph node metastasis before surgery in breast cancer patients. Furthermore, it will help determine whether to perform invasive diagnosis such as sentinel lymphadenectomy or axillary lymph node dissection, which will be useful in preventing complications.

References

• 1. Banerjee M, George J, Song EY, Roy A, Hryniuk W. Tree-based model for breast cancer prognostication. J Clin Oncol 2004;22:2567–2575.ArticlePubMed
• 2. Cianfrocca M, Goldstein LJ. Prognostic and predictive factors in early-stage breast cancer. Oncologist 2004;9:606–616.ArticlePubMed
• 3. Fisher B, Bauer M, Wickerham DL, Redmond CK, Fisher ER, Cruz AB, et al. Relation of number of positive axillary nodes to the prognosis of patients with primary breast cancer. An NSABP update. Cancer 1983;52:1551–1557.ArticlePubMed
• 4. Wilking N, Rutqvist LE, Carstensen J, Mattsson A, Skoog L. Prognostic significance of axillary nodal status in primary breast cancer in relation to the number of resected nodes. Stockholm Breast Cancer Study Group. Acta Oncol 1992;31:29–35.ArticlePubMed
• 5. Swenson KK, Nissen MJ, Ceronsky C, Swenson L, Lee MW, Tuttle TM. Comparison of side effects between sentinel lymph node and axillary lymph node dissection for breast cancer. Ann Surg Oncol 2002;9:745–753.ArticlePubMed
• 6. Mansel RE, Fallowfield L, Kissin M, Goyal A, Newcombe RG, Dixon JM, et al. Randomized multicenter trial of sentinel node biopsy versus standard axillary treatment in operable breast cancer: the ALMANAC Trial. J Natl Cancer Inst 2006;98:599–609.ArticlePubMed
• 7. Veronesi U, Paganelli G, Viale G, Luini A, Zurrida S, Galimberti V, et al. A randomized comparison of sentinel-node biopsy with routine axillary dissection in breast cancer. N Engl J Med 2003;349:546–553.ArticlePubMed
• 8. Liu CQ, Guo Y, Shi JY, Sheng Y. Late morbidity associated with a tumour-negative sentinel lymph node biopsy in primary breast cancer patients: a systematic review. Eur J Cancer 2009;45:1560–1568.ArticlePubMed
• 9. Robertson IJ, Hand F, Kell MR. FDG-PET/CT in the staging of local/regional metastases in breast cancer. Breast 2011;20:491–494.ArticlePubMed
• 10. Taylor K, O'Keeffe S, Britton PD, Wallis MG, Treece GM, Housden J, et al. Ultrasound elastography as an adjuvant to conventional ultrasound in the preoperative assessment of axillary lymph nodes in suspected breast cancer: a pilot study. Clin Radiol 2011;66:1064–1071.ArticlePubMed
• 11. Sianesi M, Ceci G, Ghirarduzzi A, Del Rio P, Guazzi A, Pisanelli B, et al. Use of axillary ultrasonography in breast cancer: a useful tool to reduce sentinel node procedures. Ann Ital Chir 2009;80:315–318.PubMed
• 12. Cooper KL, Meng Y, Harnan S, Ward SE, Fitzgerald P, Papaioannou D, et al. Positron emission tomography (PET) and magnetic resonance imaging (MRI) for the assessment of axillary lymph node metastases in early breast cancer: systematic review and economic evaluation. Health Technol Assess 2011;15:iii–iiv. 1–134.Article
• 13. Ahmed M, Douek M. What is the future of magnetic nanoparticles in the axillary management of breast cancer? Breast Cancer Res Treat 2014;143:213–218.ArticlePubMed
• 14. Deserno WM, Harisinghani MG, Taupitz M, Jager GJ, Witjes JA, Mulders PF, et al. Urinary bladder cancer: preoperative nodal staging with ferumoxtran-10-enhanced MR imaging. Radiology 2004;233:449–456.ArticlePubMed
• 15. Keller TM, Michel SC, Fröhlich J, Fink D, Caduff R, Marincek B, et al. USPIO-enhanced MRI for preoperative staging of gynecological pelvic tumors: preliminary results. Eur Radiol 2004;14:937–944.ArticlePubMed
• 16. Koh DM, Brown G, Temple L, Raja A, Toomey P, Bett N, et al. Rectal cancer: mesorectal lymph nodes at MR imaging with USPIO versus histopathologic findings--initial observations. Radiology 2004;231:91–99.ArticlePubMed
• 17. Rockall AG, Sohaib SA, Harisinghani MG, Babar SA, Singh N, Jeyarajah AR, et al. Diagnostic performance of nanoparticle-enhanced magnetic resonance imaging in the diagnosis of lymph node metastases in patients with endometrial and cervical cancer. J Clin Oncol 2005;23:2813–2821.ArticlePubMed
• 18. Harisinghani MG, Barentsz J, Hahn PF, Deserno WM, Tabatabaei S, van de Kaa CH, et al. Noninvasive detection of clinically occult lymph-node metastases in prostate cancer. N Engl J Med 2003;348:2491–2499.ArticlePubMed
• 19. Yezhelyev M, Yacoub R, O'Regan R. Inorganic nanoparticles for predictive oncology of breast cancer. Nanomedicine (Lond) 2009;4:83–103.ArticlePubMed
• 20. Harada T, Tanigawa N, Matsuki M, Nohara T, Narabayashi I. Evaluation of lymph node metastases of breast cancer using ultrasmall superparamagnetic iron oxide-enhanced magnetic resonance imaging. Eur J Radiol 2007;63:401–407.ArticlePubMed
• 21. Motomura K, Ishitobi M, Komoike Y, Koyama H, Noguchi A, Sumino H, et al. SPIO-enhanced magnetic resonance imaging for the detection of metastases in sentinel nodes localized by computed tomography lymphography in patients with breast cancer. Ann Surg Oncol 2011;18:3422–3429.ArticlePubMed
• 22. Fleiss JL. Analysis of data from multiclinic trials. Control Clin Trials 1986;7:267–275.ArticlePubMed
• 23. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ 2003;327:557–560.ArticlePubMedPMC
• 24. Memarsadeghi M, Riedl CC, Kaneider A, Galid A, Rudas M, Matzek W, et al. Axillary lymph node metastases in patients with breast carcinomas: assessment with nonenhanced versus uspio-enhanced MR imaging. Radiology 2006;241:367–377.ArticlePubMed
• 25. Nakai G, Matsuki M, Harada T, Tanigawa N, Yamada T, Barentsz J, et al. Evaluation of axillary lymph nodes by diffusion-weighted MRI using ultrasmall superparamagnetic iron oxide in patients with breast cancer: initial clinical experience. J Magn Reson Imaging 2011;34:557–562.ArticlePubMed
• 26. Kimura K, Tanigawa N, Matsuki M, Nohara T, Iwamoto M, Sumiyoshi K, et al. High-resolution MR lymphography using ultrasmall superparamagnetic iron oxide (USPIO) in the evaluation of axillary lymph nodes in patients with early stage breast cancer: preliminary results. Breast cancer 2010;17:241–246.ArticlePubMed
• 27. Michel SC, Keller TM, Fröhlich JM, Fink D, Caduff R, Seifert B, et al. Preoperative breast cancer staging: MR imaging of the axilla with ultrasmall superparamagnetic iron oxide enhancement. Radiology 2002;225:527–536.ArticlePubMed
• 28. Stadnik TW, Everaert H, Makkat S, Sacré R, Lamote J, Bourgain C. Breast imaging. Preoperative breast cancer staging: comparison of USPIO-enhanced MR imaging and 18F-fluorodeoxyglucose (FDC) positron emission tomography (PET) imaging for axillary lymph node staging--initial findings. Eur Radiol 2006;16:2153–2160.ArticlePubMed
• 29. Stets C, Brandt S, Wallis F, Buchmann J, Gilbert FJ, Heywang-Köbrunner SH. Axillary lymph node metastases: a statistical analysis of various parameters in MRI with USPIO. J Magn Reson Imaging 2002;16:60–68.ArticlePubMed

Fig. 1

Flow chart of this meta-analysis.

Fig. 2

Forest plots of all studies. Sensitivity and specificity of SPIO enhanced MRI.

SPIO = superparamagnetic iron oxide; MRI = magnetic resonance imaging; CI = confidence interval

Fig. 3

Summary receiver operating characteristics curve of SPIO enhanced MRI.

SPIO = superparamagnetic iron oxide; MRI = magnetic resonance imaging; SROC = summary receiver operating characteristic; AUC = area under curve; SE = standard error

Table 1

Characteristics and quality of the included studies

T = tesla; QUADAS = quality assessment of diagnostic accuracy studies; ALND = axillary lymph node dissection; SNB = sentinel node biopsy.

Table 2

Test for heterogeneity of SPIO enhanced MRI

SPIO = superparamagnetic iron oxide; MRI = magnetic resonance image

Table 3

Summary estimates of pooled sensitivity and specificity for SPIO enhanced MRI studies.

SPIO = superparamagnetic iron oxide; MRI = magnetic resonance imaging; CI = confidence interval; T = tesla; ALND = axillary lymph node dissection; SNB = sentinel node biopsy; QUADAS = quality assessment of diagnostic accuracy studies

Table 4

Meta-regression for the potential source of heterogeneity

DOR = diagnostic odds ratio; T = tesla; QUADAS = quality assessment of diagnostic accuracy studies.

Figure & Data

References

Citations

Citations to this article as recorded by