Preoperative magnetic resonance imaging in locoregional breast Preoperative magnetic resonance imaging in locoregional breast cancer cancer

organizaron los resultados en tablas. Hubo 46 estudios retrospectivos y comparativos uni o multicéntricos, tres estudios prospectivos, aleatorizados y controlados y cuatro metaanálisis que incluyeron pacientes con carcinoma ductal o lobular inltrantes y carcinoma ductal in situ . Los resultados comparativos fueron antagónicos y discutibles, sin embargo, en los estudios más relevantes se demostró que: la RPM retrasa la cirugía; incrementa las mastectomías y las biopsias adicionales; aumenta la detección de enfermedad ipsilateral/contralateral no necesariamente maligna; no se estableció una diferencia signicativa en la tasa de recurrencia loco-regional o a distancia. Conclusiones: La RMP en cáncer de mama no avanzado tiene resultados controversiales en relación al tipo de cirugía, reoperaciones y supervivencia libre de progresión, siendo necesario contar con estudios adicionales de tipo prospectivo, multicéntrico, aleatorizado y comparativo que dena claramente su rol y benecio. Radical; Recurrencia Local de Neoplasia; Supervivencia sin Enfermedad. (Fuente: DeCS BIREME). ABSTRACT Introduction : Preoperative magnetic resonance imaging (MRI) is controversial in patients with breast cancer, and there is no consensus on its benet compared to standard images. The objective of this review was to evaluate the comparative studies of patients with non-advanced breast cancer, with or without the use of PROM. Methods: A search was done for medical articles published from January 1, 2000, to March 31, 2021, in MEDLINE/PUBMED, LILACS, and SCIELO, and publications that met the inclusion criteria were included. Results: There were 3 828 publications, of which 53 met the inclusion criteria; the selected articles were reviewed, and the results were organized in tables. There were 46 single- or multicenter retrospective and comparative studies, three prospective, randomized, controlled studies, and four meta-analyses that included patients with inltrating ductal or lobular carcinoma and ductal carcinoma in situ. The comparative results were antagonistic and debatable; however, in the most relevant studies, it was shown that: PROM delays surgery; increases mastectomies and additional biopsies; increases detection of ipsilateral/contralateral disease not necessarily malignant; no signicant diﬀerence was established in the rate of loco-regional or distant recurrence. Conclusions: MRI in non-advanced breast cancer has controversial results in the type of surgery, reoperations, and progression-free survival. It is necessary to have additional prospective, multicenter, randomized, and comparative studies that clearly dene its role and benet.


Pág. 487
Breast magnetic resonance imaging (MRI) with dynamic images provides information on the transversal morphology of the lesion, functional characteristics, vascularization/per fusion, and permeability. This is the reason for its current nomenclature of Dynamic Contrast Enhancement (1)(2)(3) Breast MR Imaging (DCE-MRI) .
Breast MRI has evolved to high-resolution images that evaluate multiple parameters, unlike the initial conventional approach, which used only contrastenhanced sequences to evaluate tumors. The interpretation must be made with radiologists experienced in breast images because, like all evolving technology, the learning curve requires sufficient time (4) for greater certainty of the information .
The sensitivity of MRI in breast carcinoma is 88 to 100; its (5,6) speci city reaches 72% . Breast MRI is indicated nally, the preoperative staging places it in category 2B (7,8) (based on low levels of evidence) .
The use of preoperative MR images in patients with breast cancer remains controversial. There is no consensus on whether it confers a bene t since it has not shown any advantages over standard images. Therefore, we need to know the real bene t that patients achieve in relation to the surgical decision based on this tool. Meta-analysis; prospective or retrospective, singlecenter or multi-center clinical studies; Observational, retrospective clinical studies, with a control group, in patients with breast cancer (invasive, ductal carcinoma in situ and/or in ltrating lobular or ductal carcinoma), comparing preoperative MRI versus no MRI.

Study inclusion criteria
All clinical studies presented two groups of patients: patients with preoperative MRI (MRI) and another group of patients with breast cancer who only used mammography and breast ultrasound, but not MRI (noMRI).

METHODS
Searched results. Rates of: lumpectomy or mastectomy, surgical reoperation, loco-regional or distant recurrence, disease-free or progression-free, and overall survival.

Study exclusion criteria
•Clinical studies are evaluating breast cancer with histologies other than mammary adenocarcinoma.
•Clinical studies included patients with neoadjuvant treatment.
•Clinical studies included patients with metastases or patients with other synchronous cancers.
The search was carried out from January 1, 2000, to March 31, 2021, with three different data engines: M E D L I N E / P U B M E D , L I L A C S , a n d S C I E L O . MEDLINE/PUBMED was searched for all medical articles containing the word "Preoperative Magnetic Resonance AND Breast Cancer"; 3606 were found. In the LILACS platform and SCIELO, the words "Magnetic Resonance AND Breast Cancer" and "Magnetic Resonance AND Breast Cancer" were searched, 152 and 70 articles were found, respectively. Of the total collected, compliance with the inclusion and exclusion criteria was evaluated, and nally, 53 articles were selected for review. The information was transferred to tables designed to order the information based on the desired result.

INTRODUCTION
16% of women with breast cancer, the predictive prognostic value was 66%, and the ratio of true positives/false positives was 1,91; conversion from a wide local excision to a mastectomy was 8,1%, showing that MRI in this context caused a greater extension of (37) the surgery in a signi cant group of women .
Subsequently, the MONET Study, with 211 patients in the noRPM group and 207 patients in the PRP group, showed that conservative surgery (BCS) was similar in both groups (68% versus 66%); Reoperations for positive margins after BCS were signi cantly higher in (13,30) the MRI group versus the control group .
A meta-analysis published by Houssami and colleagues in 2013 found a signi cant initial mastectomy rate of 16,4% versus 8,1%; there was no difference in reoperation rate after BCS and overall mastectomy in (40) the noRMP and RMP groups correspondingly . M e t a -a n a l y s i s a n d p r o s p e c t i v e a n d randomized studies of MRI in breast cancer.
The rst prospective, randomized, controlled, multicenter clinical study was published in 2010; 1623 patients with breast cancer were enrolled in 45 hospital centers in the United Kingdom. The COMICE study compared RMP (n=816), versus no RMP (n=807). It was shown that the use of MRI was not signi cantly (38) associated with a reduction in the reoperation rate .
A new meta-analysis by Houssami in 2014 with 3169 patients showed that local recurrence-free survival at eight years was similar in patients with MRI (97%) versus non-MRI (95%); 8-year distant recurrence-free survival (42) also did not differ between groups (89% versus 93%) .

Another meta-analysis published in 2017 by Houssami
In 2014, the POMB Study, which included 440 breast cancer patients under 56 years of age in Sweden, randomly assigned one group to RMP (n=220) and another to noRMP (n=220). The RMP group had a higher rate of BCS than the control group; however, there was a change in the mastectomy decision in 23/153 patients (15%). The reoperation rate was signi cantly lower in the MRI group: 11/220 (5%) versus 33/220 (15%) in the (41) control group .

1.-Breast cancer (All histological types)
Angarita in Canada and Grady in the USA found signi cant differences in the number of new tumors (11,12) found in RMP patients vs. non-RMP .
Yi in South Korea found that patients with MRI had better ipsilateral locoregional recurrence-free survival (31) than those without MRI . Hill in the United States found in a univariate analysis that locoregional recurrence was lower in patients undergoing MRI versus no MRI, with a mean follow-up of 8 years; however, multivariate analysis showed that MRI was not (32) associated with loco-regional recurrence . In contrast, long-term studies such as the one by Ryu, Zeng, and Gervais with a follow-up of >5 and 10 years, did not show a signi cant difference in loco-regional recurrence-free survival in the groups with and without (33)(34)(35) preoperative MRI . Finally, Onega analyzed a multicenter database (The Breast Cancer Surveillance Consortium) and showed that breast cancer-speci c and adjusted mortality was not signi cant between (36) both comparisons groups .
In 2008, a meta-analysis was published whose results showed that MRI detected the additional disease in Finally, Ha in 2019, in a single-center study, after a 9-year follow-up, found that the loco-regional recurrence rate for breast cancer with lobular or mixed components was not signi cant, nor was overall survival between (62) RMP versus non-RMP .
The multifocality/multicentricity of breast cancer, evaluated in detailed pathological examinations of the In most of these studies, MRI patients were younger and had higher breast density. Premenopausal women are more likely to have aggressive breast tumor p h e n o t y p e s a s we l l a s d e n s e r b re a s t s t h a n (65) postmenopausal women . These biases can alter the results of the studies and lead to controversial conclusions. 3.-Early breast cancer (lobular or duct-lobular carcinoma) (55,56) groups of patients with and without the use of MRI .

DISCUSSION
M o s t m a l i g n a n t b r e a s t n e o p l a s m s a r e adenocarcinomas, which constitute more than 95% of breast cancers and are classi ed as in situ or invasive. In carcinoma in situ, cells are restricted within the lobularductal system of the breast, whereas in invasive carcinoma, cells spread beyond that structure.
Therefore, invasive carcinomas (both ductal, lobular or mixed) and ductal carcinoma in situ have been considered for this systematic review, but not lobular carcinoma in situ, since it is regarded as a non-obligate (63,64) precursor of breast carcinoma .
In  recurrence is signi cantly reduced by up to 70% over a (78) 10-year period .
In the retrospective and comparative studies of this review, antagonistic results were found that do not allow evaluating the true differential weight between the use or not of MRI, in early breast cancer.
Regarding the prospective, randomized and controlled studies, the rst to be carried out was the COMICE38 , which did not nd that breast density signi cantly in uenced the reoperation rate; One year later, the MONET in patients with non-palpable breast lesions found that the reoperation rate was signi cantly higher (40) in patients with MRI. Finally In relation to meta-analyses, the rst one carried out by (37) Housami in 2008 showed that MRI detected the additional disease in 16% of women with breast cancer, however, the ratio of true positives/false positives was 2:1, that is, out of every three women diagnosed and biopsied through MRI, one was false positive; conversion to mastectomy was 8.1%; it was one of the rst studies to question the usefulness of the MRI. The (41,43) same author published 2 more meta-analyses showing that the use of MRI was associated with a higher rate of mastectomy, but not reoperation. There are criticisms of these meta-analyses on the basis that only three randomized trials were included and there (80) were serious methodological de ciencies .

Pág. 491
The Ha in 2019, with a 9-year follow-up, exposed this position when it found that loco-regional recurrence due to breast cancer with a lobular or mixed component was not signi cant in the groups with and without preoperative MRI.