Clinical and MRI variables associated with close or positive margins during breast-conserving surgery using MRI projection mapping in breast carcinoma with nonmass enhancement

Abstract

Purpose

To evaluate the utility of a magnetic resonance imaging (MRI) projection mapping system (PMS) for determining the resection lines during breast-conserving surgery (BCS) in patients with breast cancer presenting with nonmass enhancement (NME) and identify the clinical or MRI variables associated with close or positive margins.

Materials and methods

Forty-one patients with breast cancer exhibiting NME were enrolled. In the operating room, a maximum intensity projection image generated from supine MRI was projected onto the breast using a PMS, which employed a structured light method to measure the surface of the breast. Cancer contours delineated on the MRI-PMS, with an additional safety margin, served as the resection lines for cylindrical BCS. Margins were pathologically categorized as negative (> 2 mm), close (≤ 2 mm), or positive. The association between margin status and clinical or MRI variables was analyzed.

Results

Surgical margins were negative in 24 patients (58.5 %), close in 15 (36.6 %), and positive in 2 (4.9 %). There were significant differences in the maximum diameter of nonmass components (NMCs) shown by pathology, that of NME on MRI, and the discrepancy between the two diameters between patients with negative margin and those with close or positive margin (< 0.05 for all). Receiver operating characteristics revealed that threshold of 40 mm for NMEs provided high specificity of 91.7 %.

Conclusion

The MRI-PMS led to a low rate of positive margins during BCS in patients with breast cancer with NMEs. Large NMCs and NMEs are associated with positive or close margin.

Graphical Abstract

Highlights

• •A partial mastectomy using projection mapping was done for nonmass breast cancer.
• •An MIP image from supine MRI was projected on the breast skin.
• •The rate of positive margins was quite low.
• •Larger nonmass components on pathology were related to close or positive margins.
• •Larger nonmass enhancement on MRI was related to close or positive margins.

1. Introduction

Breast cancer presenting mass morphology is palpable and easily identified using ultrasonography. In addition, it is a good indication for breast-conserving surgery (BCS) during the early stages [1]. The recurrence rate for breast cancer when BCS is conducted with a negative surgical margin is comparable to that of a total mastectomy [1]. It is also well known that negative margins reduce the odds of local recurrence [2], [3]. Patients with ductal carcinoma in situ (DCIS) or invasive lobular carcinoma (ILC) have a significantly higher risk of positive margins [4], [5]. Magnetic resonance imaging (MRI) often shows nonmass enhancement (NME), which may be external intraductal components with invasive cancer or DCIS, as well as invasive ductal carcinoma (IDC) presenting a mass [5], [6], [7]. In addition, ILC exhibits NME on MRI [5], [6], [8], [9]. The NME can induce high rates of positive margins and recurrence following BCS [10], [11], [12]. Therefore, contrast-enhanced MRI is essential for determining the resection lines for BCS. This is especially the case in patients with DCIS or ILC showing NME, which may be consistent with nonmass components (NMCs) shown by pathology [8], [13], [14]. Nonetheless, the clinical utility of preoperative MRI is limited for BCS [9] because the patients’ position differs between the MRI (i.e., prone position) and the operating bed (i.e., supine position). The shapes of “soft” breasts as well as those of the NMCs change according to the patients’ positions during examination [15].

So far, there are no clear anatomical landmarks other than the nipple when identifying tumor location in BCS [16]. Several studies have examined methods for visualizing the extent of breast cancer on the breast skin using MRI, including real-time virtual sonography (RVS) and three-dimensional (3D) printing [17], [18]. We have developed prototype MRI projection mapping system (PMS) to maximize the utility of MRI by projecting breast cancer visualized by contrast-enhanced supine MRI onto the patient’s body directly when delineating the resection lines in the operating room [19].

In the present study, we assessed whether the MRI-PMS delineated the resection lines for BCS appropriately in patients with breast cancer with NME. We aimed to identify clinical and MRI variables associated with pathological close or positive margins during BCS performed using the MRI-PMS.

2. Materials and methods

2.1. Participants

This retrospective study was approved by our Institutional Review Board (approval number: 20240302), and all patients gave written informed consent to participate in this study. We enrolled 41 women with breast cancer with NME shown on prone MRI, who underwent BCS using the MRI-PMS from April 2021 to March 2025. We excluded patients with breast cancer who had received neoadjuvant chemotherapy before undergoing BCS and those who had undergone BCS before this study.

2.2. MRI acquisition

MRI examinations were performed using a 3.0 T imager (Ingenia Achieva; Philips Healthcare, Best, the Netherlands). A 16-channel phased array breast coil was used for signal reception in the MRI examination performed in the prone position, and a 28-channel coil (anterior phased array torso, 16ch; posterior, 12ch) phased array torso coil was used in supine examinations. Transverse fat-suppressed contrast-enhanced T1-weighted 3D gradient-echo imaging in the prone position was performed with the typical imaging parameters as follows: repetition time (TR), 3.9 ms; echo time (TE), 1.7 ms; flip angle, 11°; field of view (FOV), 340 × 340 mm²; matrix, 560 × 560; slice thickness 2 mm; receiver bandwidth (RBW), 722 Hz/pixel; one acquisition; and sensitivity encoding (SENSE) with a reduction factor of 2.0. Gadobutrol (Gadovist®; Bayer AG, Leverkusen, Germany) at a dose of 0.1 mmol/kg was administered at a flow rate of 1 ml/s. This imaging covered both breasts. One day before the surgery, transverse fat-suppressed contrast-enhanced T1-weighted 3D gradient-echo imaging in the supine position was acquired 60 s after a gadobutrol injection at a dose of 0.1 mmol/kg with the typical imaging parameters as follows: TR, 3.5 ms; TE, 1.8 ms; flip angle, 10°; FOV, 340 × 340 mm²; matrix, 380 × 380; slice thickness, 1.0 mm; RBW, 783 Hz/pixel; one acquisition; and SENSE with a reduction factor of 3.4. This imaging covered the unilateral breast with breast cancer and skin markers. The patient’s arms were raised using a homemade armrest that allowed the arm position to be identical to that in the operating room [18]. Nifedipine capsules (Adalat®; Bayer Healthcare, Osaka, Japan) showing high signal intensity on the fat-suppressed contrast-enhanced T1-weighted images were used as the rigid and nonrigid skin markers, and their locations were marked using a pen before removing the Adalat® capsules from the patient’s chest.

2.3. Projection mapping system

The PMS was described in detail previously [19] and consisted of a projector (350 lm WXGA DLP projector, ASUS P2E; ASUSTeK Computer, Taipei, Taiwan), a color digital camera (CMOS Color: JAI GO-5000C; JAI, Kanagawa, Japan), and a 16 mm FA lens fixed-aperture f 8.0 (RICOH FL-BC1618–9; RICOH, Tokyo, Japan; Fig. 1a). The PMS had a movable arm tilting head to project a maximum intensity projection (MIP) image generated from the supine contrast-enhanced 3D MRI onto the patient’s breast perpendicularly from above. The depth and surface information of the “curved” breast were analyzed using the structured light method (Fig. 1b). Thereafter, the PMS projected the MIP image, which showed both breast cancer and skin markers, onto the patient’s breast as the skin markers coincided with between the MIP image and the patient’s breast surface in the operating room. The cancer contour shown by the PMS was traced onto the breast surface. During this procedure, the patient was under general anesthesia, and their arms were raised on the housemade armrest used in the supine MRI.

Fig. 1.

The projection mapping system (a) has a movable arm and tilting head with a color digital camera and a projector. The depth and surface information of the curved and soft breast are analyzed using the structured light method (b).

2.4. Determination of surgical lines for BCS

In the operating room, the resection line was delineated around the traced cancer contour, which was determined using the MRI-PMS, adding the safety margin of 10 or 15 mm. Indocyanine green staining of the resection lines was then performed for the cylindrical BCS [20]. Thereafter, the armrest was removed to set the patient’s arms sideways (i.e., a common position in breast surgery). Fig. 2 shows a representative case.

Fig. 2.

A 67-year-old woman with ductal carcinoma in situ (DCIS) in the right upper breast. The DCIS was identified as nonmass enhancement (NME) in the transverse (a) and maximum intensity projection (MIP) images (b) of the prone contrast-enhanced MRI and in the source transverse (c) and MIP images (d) of the supine MRI performed one day before breast-conserving surgery. The supine MRI shows the skin marker (green arrow, c and d) and NME in the outer part rather than the prone MRI. A breast marker (red arrow) at the tumor bed, placed during biopsy, is also seen. The projection mapping system (PMS) projects the MIP image generated from the supine MRI, which shows both the enhancing breast cancer and skin markers, on the patient’s breast surface in the operating room (e). The resection line (red dots) is delineated as the cancer contour visualized by the MRI-PMS (red margin) plus the safety margin of 10 mm (f). A black circle is a hypoechoic area of DCIS delineated using ultrasonography. An X-ray of the resected specimen shows that the breast marker at the tumor bed is at the center of the specimen (g). The pathological examination reveals DCIS (black dots), which does not extend the surgical margin (i.e., negative margin) (h). Other investigated variables are as follows: body mass index, 25.6; BPE shown on prone and supine MRI, low; the maximum diameter of non mass enhancement (NME) shown on supine MRI, 36 mm; non mass components (NMC) shown by pathology, 40 mm; and the discrepancy of the distance between NMC and NME (in the direction from cancer to the nipple), 4 mm.

2.5. Evaluation of the availability of the MRI-PMS

We assessed whether the nipple, rigid, or nonrigid skin markers on the MIP image were projected onto the patient’s nipple and marks on the breast skin appropriately.

2.6. Pathological data

The resected specimens were sliced into approximately 5-mm-thick specimens, perpendicular to the line connecting the nipple and the tumor. All specimens were processed into pathological slides to evaluate the surgical margins. Tumor size was measured in two directions: along the axis from the tumor to the nipple and perpendicular to that axis [21]. Additionally, the shortest lateral pathological margin was measured and recorded (Fig. 3a). Based on the surgical margin evaluation criteria recommended in the Japanese Breast Cancer Society’s Clinical Practice Guidelines for Breast Cancer, Part 1: Treatment [22], [23], the margin status was classified into the following three categories: 1. Negative margin: No cancer cells are within 2 mm of the margin; 2. Close, but negative margin: Cancer cells are within 2 mm of the margin but not at the margin; and 3. Positive margin: Cancer cells are exposed at the resection margin. Two board-certified pathologists performed all pathological diagnoses.

Fig. 3.

The surgical specimen of a 43-year-old woman with invasive ductal carcinoma (IDC) in the right outer breast shows IDC (red) and ductal carcinoma in situ (DCIS, black) spreading in a mix, presenting the nonmass components as a whole (a). Slice lines perpendicular to the nipple-cancer direction are shown by purple lines. The maximum diameter in the slice direction (solid blue arrow) and that in the nipple-cancer direction perpendicular to it (blue dashed arrow) are measured (30 mm, 55 mm, respectively). In this case, the shortest margin was in the slice direction (white arrows), and less than 2 mm, which means a close margin. The maximum diameters of nonmass enhancement on a maximum intensity projection image of supine MRI in the two directions are measured (23 mm, 45 mm, respectively), in the same way as pathology (b). The resection line (red dots) is delineated as the cancer contour visualized by the MRI-projection mapping system (red margin) plus the safety margin of 15 mm (c). Discrepancy of the distance in the direction of the shortest lateral margin (slice direction in this case),7 mm.

2.7. Evaluation of MRI

A radiologist with 16 years of experience in breast MRI assessed the MRI findings. First, background parenchymal enhancement (BPE) of the affected breast shown on prone and supine MRI was assessed on a two-level classification (i.e., minimal/mild vs. moderate/marked). The maximum diameters of the NME on the MIP image generated from supine MRI were measured in the two directions, which were identical to the directions in the pathological evaluation noted above (Fig. 3b).

2.8. Statistical analysis

Patients were divided into a negative-margin group and a close or positive-margin group according to whether additional treatment beyond whole-breast irradiation was required after BCS. Recently, additional resection or reoperation has been replaced by boost irradiation to the tumor bed in patients with a close surgical margin during BCS [24]. The differences in patients’ characteristics were evaluated between the two groups. These characteristics were patients’ ages and body mass index (BMI), the maximum diameters of the NMEs shown on supine MRI and that of the NMCs on pathological specimens, and the discrepancy of the diameters between NMC shown by pathology and NME shown by supine MRI in the direction of the shortest lateral pathological margin. The discrepancy was calculated as the diameter of the pathology minus that of the supine MRI. When appropriate, the Student t-test or Mann–Whitney U-test was used to compare the two patient groups. The differences in BPE shown on prone or supine MRIs and the width of the safety margin (i.e., 10 mm vs. 15 mm) were evaluated between the two patient groups using Fisher’s exact test. Agreement between the maximum diameters of NMEs shown on supine MRIs and NMCs shown by pathology was evaluated using Bland–Altman analysis. Receiver operating characteristics (ROC) analysis was used to determine the threshold of MRI variables which predicted close or positive margins following BCS. A P-value < 0.05 was considered a statistically significant difference for all analyses in this study. SPSS software (v. 28.0; IBM SPSS, Armonk, NY, USA) was used for statistical analyses.

3. Results

3.1. Lesion characteristics

The MRI-PMS was used for 43 patients with breast cancer who exhibited NMEs on MRI. Two patients were excluded because the skin markers projected by the MRI-PMS were localized > 20 mm apart from the skin markers on the patients’ breasts in the operating room, probably due to a mechanical error of the PMS. Consequently, 41 patients with NMEs were assessed (age, 35–74 years; mean, 51.8 years; BMI, 16.8–34.9 kg/m²; median, 21.9 kg/m²). Their pathological diagnoses were as follows: pure DCIS (n = 16), IDC (n = 17), IDC with predominant DCIS (n = 3), ILC (n = 3), and microinvasive carcinoma (n = 2). Among invasive cases, T-classifications were T1 (n = 21), T2 (n = 3), and T3 (n = 1), with luminal (n = 23) and HER2-enriched (n = 2) immunohistochemical subtypes. The safety margin widths were 10 mm (n = 7) and 15 mm (n = 34). The patients’ characteristics are summarized in Table 1.

Table 1.

Baseline characteristics of participants.

Cancer characteristics	N	%
Age (mean±SD)	51.7 ± 9.5 (35–74)
BMI (median± IQR)	21.9 ± 5.5 (16.8–34.9)
Histological pathology	41
DCIS	16	39.0
IDC	17	41.5
IDC with predominant DCIS	3	7.3
microinvasive carcinoma	2	4.9
ILC	3	7.3
T-classification	41
Tis	16	39.0
T1mi	2	4.9
T1a	3	7.3
T1b	7	17.1
T1c	9	22.0
T2	3	7.3
T3	1	2.4
Molecular subtypes of invasive cancers	25
Luminal A	19	76.0
Luminal B, HER2-	4	16.0
Luminal B, HER2 +	1	4.0
HER2	1	4.0
Triple negative	0	0

BMI, body mass index; DCIS, ductal carcinoma in situ; IDC, invasive ductal carcinoma in situ; ILC, invasive lobular carcinoma; HER2, human epidermal receptor 2

3.2. Pathological findings for surgical margin status

Surgical margins delineated by the MRI-PMS were cancer-negative in 24 patients (58.5 %), close in 15 (36.6 %), and positive in 2 (4.9 %). The pathological findings of the shortest margins showed invasive cancer in four cases, all of which were associated with negative margins, whereas the remaining 37 cases revealed non invasive cancer. In the two patients with positive margins, additional resection was performed in one patient, and reoperation was conducted in the other. In the patients with close margins, whole breast irradiation was performed, followed by boost radiation on the tumor bed.

3.3. Evaluation of the associations between the clinical or MRI variables and the surgical margin status (Table 2)

Table 2.

Summary of clinical and MRI variables and association with surgical margin status.

		Surgical margin
	Total (n = 41)	Negative (n = 24)	Close (n = 15), Positive (n = 2)
Patient's baseline characteristics				p value
Age (years)	51 [14] (35–74)	51 [10] (40–74)	47 [13.8] (35–72)	0.307a
BMI (kg/m²)	21.9[5.4] (16.8–34.9)	21.3 [5.0] (17.6–28.9)	22.7[6.0] (16.8–34.9)	0.751a
Pathological findings
Direction of shortest lateral margin
From the nipple to cancer	16 (39)
Perpendicular to above	25 (61)
NMC diameter in the shortest lateral margin (mm)	23 [28.5] (3–135)	18 [14.5] (3–75)	38 [28.55] (10–135)
Maximum diameter of NMC (mm)	30 [34.3] (3–135)	20.5 [19.0] (7−75)	50 [26.8] (13–135)	0.002*,a
Width of safety margin (mm)				0.105b
10 mm	7 (17.1)	2	5
15 mm	34(82.9)	22	12
Imaging variables on MRI
BPE on prone MRI				0.812b
Minimal/Mild	25 (61)	15	10
Moderate/Marked	16 (39)	9	7
BPE on supine MRI				0.507b
Minimal/Mild	29 (70.7)	18	11
Moderate/Marked	12 (29.3)	6	6
NME diameter in the shortest lateral margin (mm)	20 [17.8] (0–80)	14.5[13.0] (0–80)	23 [24.5] (0–80)
Maximum diameter of NME (mm)	27 [21.0] (0–80)	21.5[19.8] (0–80)	30 [28.3] (0–80)	0.028*,a
Discrepancy of the diameters between NMC and NME (mm)	6 [14.5] (−24–55)	0.5 [12.3] (−24–32)	10 [16.3] (−7–55)	0.010*,a

Median [IQR]

Negative; > 2 mm; Close: = or < 2 mm; Positive: no tumor on ink

NMC, nonmass component shown by pathology; NME, nonmass enhancement shown by supine MRI

^aMann-Whitney U test

^bFisher’s Exact Test

^*p < 0.05

The maximum diameters of the NMEs on supine MRI ranged from 0 to 80 mm (median, 27 mm; interquartile range [IQR], 21.0 mm), and those of the NMCs shown by pathology were from 3 to 135 mm (median, 30 mm; IQR, 34.3 mm). The discrepancy of the diameters between NMC and NME ranged from −24 to 55 mm (median, 6 mm; IQR, 14.5 mm). Prone MRI showed minimal/mild BPE in 25 cases and moderate/marked BPE in 16 cases, while prone MRI showed minimal/mild BPE in 29 cases and moderate/marked BPE in 12 cases.

There were significant differences in the maximum diameters of NMCs shown by pathology (negative margin: 20.5 mm, IQR, 19.0 mm; positive or close margin: median, 50 mm; IQR, 26.8 mm; P = 0.002) and the maximum NME diameter shown on supine MRI (negative margin: 21.5 mm, IQR, 19.8 mm; positive or close margin: median, 30 mm, IQR, 28.3 mm; P = 0.028) between the two groups (Fig. 4a, b). Significant differences were also found in the discrepancy of diameters between NMC and NME between the two groups (negative margin: 0.5 mm, IQR, 12.3 mm; positive or close margin: median, 10 mm, IQR, 16.3 mm; P = 0.010, Fig. 4c). In the ROC analysis of maximum NME diameter, the cutoff value between negative and close or positive margins was 40 mm, with an AUC of 0.703 (95 % CI, 0.538–0.869), sensitivity of 47.1 %, and specificity of 91.7 % (Fig. 5). Bland–Altman analysis showed high agreement between the maximum diameters of NMCs shown by pathology and NME shown on supine MRI, with a mean difference of 6.2 mm (Fig. 6). Thus, the maximum diameters of NMEs shown on supine MRI tended to be smaller than the maximum diameter of NMCs shown by pathology. Patients’ age, BMI, width of safety margin, or BPE shown on prone/supine MRI did not differ between the two groups (P > 0.05 for all).

Fig. 4.

There were significant differences in the maximum diameter of nonmass components shown by pathology (a, P ＝0.002), non mass enhancement shown on supine MRI (b, P = 0.028), or discrepancy of diameters between non mass component shown by pathology and non mass enhancement shown by supine MRI in the direction of the shortest lateral margin (c, P = 0.010), between cases with negative margins and those with close or positive margins. N, negative; C, close (≤ 2 mm); P, positive (> 2 mm) of surgical margin from cancerous cells.

Fig. 5.

Receiver operating characteristics (ROC) analysis was used to determine the threshold of MRI variables which predicted close or positive margins. The cut off value of 40 mm in nonmass enhancement provides high specificity of 91.7 %.

Fig. 6.

Bland–Altman analysis of agreements of diameters between nonmass enhancement (NMEs) shown on supine MRI and nonmass components (NMCs) shown by pathology in the direction of the shortest lateral margin. The mean difference is 6.2 mm, which indicates that the diameters of NMEs shown by MRI tend to be smaller than those of NMCs shown by pathology. Blue dots, negative margins; red dots, close or positive margins.

4. Discussion

This study demonstrated that the surgical margins delineated by the MRI-PMS were free of cancerous cells in almost all patients; that is, they were negative in 24 (58.5 %) of the 41 patients and close in 15 patients (36.6 %). Only two patients (4.9 %) had positive margins. This frequency of positive margins was quite low in the present study, compared with previous reports [3], [4], [10], [12], [25], [26], [27], [28], [29], and additional resection was performed in these patients. The patients with close margin underwent the addition of boost radiation to reduce the local recurrence rate [21], [24], [30].

The maximum diameters of NMCs shown by pathology, and those of NMEs shown on supine MRI, were significantly larger in the patients with close or positive margins than those with negative margins. This result was consistent with several reports [4], [10], [15], [26], [31]. We also observed that larger discrepancy of diameters between NMC and NME was associated with close or positive margins following BCS, and the diameters were underestimated by MRI compared with pathology [32]. This finding may be attributed to the low tumor density of DCIS or ILC [33]. Among these variables, only the maximum diameter of NME can be evaluated before the surgery. Even using MRI-PM, radiologists should recommend that surgeons apply sufficient safety margins or consider mastectomy to avoid close or positive margins in breast cancer with a large NME on MRI (e.g., > 40 mm indicated by ROC analysis), whereas there is a trade-off between breast appearance after BCS and a sufficient amount of resected breast tissue. On the other hand, intraoperative pathological diagnosis is still inevitable because of low sensitivity of the threshold for NME (i.e., 47.1 %). If T2-weighted, diffusion-weighted imaging or radiomics generated from artificial intelligence (AI) visualized DCIS more accurately than contrast-enhanced imaging [34], [35], the discrepancy between NMCs and NMEs could be reduced.

There are several prior studies to determine the surgical lines of BCS. The placement of a couple of markers in the tumor bed is invasive and does not cover the entire circumference of the tumor [36]. MRI-PMS exhibited the cancer contour noninvasively, which may lead to quite low rate of the positive margin. RVS is used to identify tumor location in the operating room [16]. Compared with RVS, MRI-PMS visualized the cancer contour comprehensively, especially the extent of NME reflecting DCIS or ILC, which might be difficult to be detected by ultrasonography. 3D printing, although in small quantities, raises environmental concerns regarding plastic waste [17]. Medical innovations utilizing AI have advanced in the field of breast cancer, including the AI-based diagnosis and treatment of breast cancer [37], [38]. However, to the best of our knowledge, there are no studies that have applied AI to direct localization of breast cancer when performing BCS. It is possible to scan supine images following a routine prone breast MRI [39]. However, we often experienced that the BPE in the delayed phase (e.g., about 13 min after contrast) obscured the spread of the NME. If AI was able to transform prone MRI to supine MR, we would reduce the dose of gadolinium and its cost necessary for MRI-PMS. This may overcome one disadvantage of MRI-PMS, that is, contrast-enhanced supine MRI should be added one day before BCS.

This study has some limitations. First, the sample size was small. Second, MRI overestimation [40], that is, the diameters in the direction of the shortest margin shown on supine MRI were larger than those shown by pathology, was not investigated, even though overresection may cause a lack of breast appearance. Third, this study did not evaluate some outcomes, such as local recurrence. We plan to conduct follow-up observations to evaluate local recurrence. Fourth, dye-guided cylindrical resection of breast cancer [22] was performed as BCS in our study, whereas BCS using a hook wire or iodine-125 seeding was performed in some earlier studies [15], [41], [42]. MRI-PMS is a great match for the cylindrical resection in delineating the extent of breast cancer and determining the resection lines. Last, because the present MRI-PMS was a prototype, we did not conduct multi-institutional study. Commercialization and mass production of MRI-PMS should be required for its distribution to various institutions. Recent use of augmented reality (AR) devices with tablet terminals can be alternatives to PMS [43], whereas supine contrast-enhanced MRI is still inevitable. Therefore, the present result, which indicates the MRI variables linked to the close or positive margin, may be applicable for AR-assisted BCS.

5. Conclusion

Women with breast cancer presenting NMEs may carry the risk of a positive margin during BCS. However, the MRI-PMS led to a low rate of positive margins during BCS in patients with breast cancer and NMEs shown on MRI. Nonetheless, several patients with close or positive margins had NMEs larger than 40 mm on supine MRI.

Video 1.

Download video file ^{(10.9MB, mp4)}

video of scenes from PM to staining the resection lines.

Ethical Statement

This retrospective study was approved by our Institutional Review Board (approval number: 20240302).

CRediT authorship contribution statement

Yumi Koyama: Methodology, Data curation. Xiaoyan Tang: Methodology, Data curation. Fumi Nozaki: Methodology, Data curation. Mayumi Tani: Methodology, Data curation. Yasuo Amano: Writing – review & editing, Supervision. Maki Amano: Writing – original draft, Validation, Project administration, Funding acquisition, Formal analysis. Jun Ozeki: Methodology, Data curation.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.