European Journal of Cancer (2015) 51, 734– 741
Available at www.sciencedirect.com
ScienceDirect journal homepage: www.ejcancer.com
A new method of surgical margin assuring for abdominal radical trachelectomy in frozen section Dandan Zhang a,c,d, Huijuan Ge b,d, Jin Li a,d, Xiaohua Wu a,d,⇑ a
Department of Gynecologyic Oncology, Fudan University Shanghai Cancer Center, Shanghai, 200032, PR China Department of Pathology, Fudan University Shanghai Cancer Center, Shanghai, 200032, PR China c Department of Obstetrics & Gynecology, Shengjing Hospital of China Medical University, Shenyang, 110004, PR China d Department of Oncology, Shanghai Medical College, Fudan University, Shanghai, 200032, PR China b
Received 4 November 2014; received in revised form 19 January 2015; accepted 28 January 2015 Available online 19 February 2015
KEYWORDS Abdominal radical trachelectomy Cervical cancer Surgical section Frozen section
Abstract Objectives: The aim of this study was to introduce a new method of assuring surgical margins for abdominal radical trachelectomy (ART) and report our experience using the method. Methods: We combined transverse and perpendicular sections to assess surgical margins of specimens from RT. All surgeries from 1st August 2012 to 1st October 2013 were performed by one surgeon. The frozen section (FS) was consistently performed by a group of gynaecologic pathologists according to the detailed protocol described in this article. All cases were prepared by the same pathologist, and the slides were reviewed by two pathologists. Results: There were 53 patients treated using the new method in our institution. The patient ages ranged from 20 to 41 years old (median 32). The surgeries were performed for clinical stage IA (n = 11) with LVSI and IB (n = 42) tumours (40 squamous cell carcinoma, 11 adenocarcinoma, two adenosquamous and two others). In 20 (37.74%) cases, no residual tumour of the ART specimen on frozen section was observed in the specimens as it was cleared by the preceding loop electrical excision procedure (LEEP) or conization. The margins were initially reported as negative in 45 cases and positive in nine cases. In those nine cases, a second slice of cervix was removed and negative in six cases and positive again in two cases, the other one with positive nodes. The results of frozen sections were concordant with the final paraffin-embedded sections. There were no false negative intraoperative assessments. There were no recurrences after a median follow-up of 15.4 months (range, 6–21 months). Conclusions: Combining transverse and perpendicular sections to assess surgical margins of specimens from RT makes the protocol simple, reliable and produces accurate results. Ó 2015 Elsevier Ltd. All rights reserved.
⇑ Corresponding author at: Department of Gynecologic Oncology, Fudan University Shanghai Cancer Center, 270 Dong’an Road, Shanghai 200032, PR China. Tel.: +86 21 6417 5590x81006; fax: +86 21 64220677. E-mail address: [email protected] (X. Wu).
http://dx.doi.org/10.1016/j.ejca.2015.01.062 0959-8049/Ó 2015 Elsevier Ltd. All rights reserved.
D. Zhang et al. / European Journal of Cancer 51 (2015) 734–741
1. Introduction Cervical cancer is the second most common cancer in women in developing countries, and more than 500,000 new cases are diagnosed worldwide every year [1]. Early-stage cervical cancer has a very good prognosis and excellent overall survival rates. Most patients suffering from early-stage cervical cancer are diagnosed at a median age of 40 years old. Due to the increased number of delayed pregnancies, many patients are diagnosed during reproductive ages. This result leads to the issue of whether it is possible to preserve the uterus for future pregnancy without increasing the risk of recurrence. In 1994, Dargent [2] reported the utility of vaginal radical trachelectomy (VRT) with laparoscopic lymph node dissection for the treatment of early-stage cervical cancer. Smith et al. [3] reported abdominal radical trachelectomy (ART) in 1997, and ART is considered another radical surgical procedure for early cervical cancer. Many studies investigating radical trachelectomy have been published worldwide in the last decade. Most of the publications have focused on the procedural clinical aspects, and there is little emphasis on the specimen handling in the pathology suite. Corpus uteri invasion is determined by the operative margin and is confirmed by intraoperative frozen section analysis. The lower uterine segment (LUS) or the upper endocervix (EC) surgical margin is the boundary of the trachelectomy specimen. The presence of tumour invasion is a very important indicator for cancer recurrence or adjuvant therapy indications, especially for patients with an endogenous tumour in the cervical canal or cervical stromal invasion. Several published reports of different techniques demonstrate a lack of guidance in the pathology community regarding the accepted guidelines for the most reliable and feasible approach. There is no existing agreement on the best method of sampling a margin of a frozen section. The selection of an appropriate method is dictated by the information required by the surgeon. This study was performed in our centre and presents a new method of frozen section analysis for radical trachelectomy. The new method has the following advantages: (1) it is simple, convenient and less time consuming; (2) no professional gynaecological pathologist is required so the method is easier to popularise; and (3) it ensures margins without residual tumour. 2. Materials and methods A prospectively maintained database of early-stage cervical cancer patients who were scheduled to undergo abdominal radical trachelectomy was analysed. All of the operations were performed in the Department of Gynecologic Oncology at Fudan University Shanghai Cancer Center from 1st August 2012 to 1st October
735
2013 by the same surgeon. The initial diagnosis was made on cervical biopsies, loop electrical excision procedure (LEEP) specimens, or cone biopsy specimens. The criteria for preserving the corpus uteri and part of the cervix were as follows: (1) desire to preserve fertility and no clinical evidence of impaired fertility; (2) histologic diagnosis of squamous cell carcinoma, adenocarcinoma, or adenosquamous carcinoma; (3) FIGO stage IA1 disease with lymphovascular space invasion, or positive surgical margin and distorted cervicovaginal anatomy after conisation, and stage IA2 or IB1 disease; (4) tumour size less than 4 cm; (5) no evidence of parametrial involvement and confirmation of tumour limited to the cervix; and (6) no evidence of pelvic lymph node metastases [4]. There are different approaches for submitting sections of the trachelectomy margin for histologic evaluation. However, there is no defined consensus on the best sampling method. Surgeons may prefer transverse (en face) sections [5], while others might prefer a longitudinal (perpendicular) section [6]. Others only take a perpendicular margin if there is a grossly visible tumour [7]. At our institution, we applied both transverse and perpendicular methods. The new combination method is briefly described in the following four steps. Step 1: When the specimen was sliced from the corpus uteri, the endocervical canal was opened through a longitudinal anterior section at the 12:00 location (Fig 1A). The shape, size and location of the tumour were examined by an unaided eye, described and recorded in the final pathological report. The tumour form in ART specimens was graded into three categories by the gross appearance: (1) no grossly visible lesion, such that the tumour could not be clearly located due to preoperative conisation or LEEP; (2) a nonspecific lesion, which was defined as a granular, irregular, or ulcerated zone that may correspond to the previous biopsy site and in which a gross diagnosis of residual cancer was uncertain; and (3) a grossly visible tumour for which a tumour was clearly identifiable. Step 2: If the tumour was previously visible or coned, the pathologist then cut off the cervical specimen 8 mm from the margin and cut away a 2-mm-thick en face slice. This requires making a transverse section 10 mm from the end of the specimen for the frozen section (Fig 1B). Step 3: The section was divided into two slices, which were marked 0–6 o’clock and 6–12 o’clock. Each slice was checked from the mucosal surface to the serosa using a microscope (Fig 1C). The most important aspect of the method is examining the whole endocervical stroma including endocervical clefts deep within the stroma. If no residual tumour was found in the frozen section, then there was a satisfactory margin with no invasive tumour at least 10 mm from the tumour.
736
D. Zhang et al. / European Journal of Cancer 51 (2015) 734–741
A
B 12 o’clock
8mm
0 o’clock
10mm
12 o’clock
C 6-12 o’clock
0-6 o’clock 6-12 o’clock
0-6 o’clock
2mm frozen section 12 o’clock
0 o’clock
D The additionally excised tissue 10mm The second en face The first en face
Fig. 1. Schematic representation of how to submit trachelectomy margin at the time of frozen-section. (A) Longitudinal open the endocervical canal at 12 o’clock. (B) Transverse section, approximately 2 mm in thickness, 8 mm distal to surgical margin. (C) Divide the section into two slices, marked 0–6 and 6–12 o’clock respectively. (D) If residual tumour is found, the margin is revised, and transverse section is submitted 8 mm distal to revised margin.
Step 4: If the frozen section presented tumour tissue and the tumour clearance is less than 10 mm, then the LUS/EC margin had to be revised. The thickness of additional excised tissue was more than 5 mm. Then, a 2-mm en face cut of the cervical canal was performed, which ensured that the distance from the second transverse section to the additional proximal resection margin was 10 mm (Fig 1D). The en face tissue was examined by frozen section analysis using the same protocol. If the second transverse section showed tumour infiltration or the tumour-margin range was still unsatisfactory, the ART was abandoned and the patient received a radical hysterectomy. The surgeons had informed the patients of this risk during the pre-operative period. The final evaluations of paraffin-embedded tissue and the ART specimens were submitted again after formalin fixation. We improved the first step of the procedure to ensure the tumour-margin distance. Before slicing the specimen from the uteri, the surgeon quarantined the cervix from the surgical sterile region and cut the endocervical canal at the 12:00 location through a longitudinal anterior section (Fig 2). The cervix was then sliced from the corpus uteri at the estimated position, which was determined by
the gross 10 mm separation in the patulous cervix. The divided specimen was also checked, and en face slices were prepared as described in the second step of the original protocol. The patients with positive pelvic lymph nodes on final pathology after ART were recommended to undergo adjuvant radiation with concurrent platinum-based chemotherapy. The patients who had deep stromal infiltration (DSI) or lymphovascular space invasion (LVSI) on final pathology underwent adjuvant chemotherapy. The specific treatment regimens have been published previously. 3. Results Between August 2012 and October 2013, a total of 53 patients underwent a laparotomy for a planned fertilitysparing ART and pelvic lymphadenectomy. All of the patients received a preoperative MRI to exclude endocervical canal involvement and pelvic metastasis. The procedure was converted to an abdominal radical hysterectomy (ARH) in three of the 53 patients who were considered for ART. There was one patient with positive pelvic lymph nodes, and two patients included in
D. Zhang et al. / European Journal of Cancer 51 (2015) 734–741
737
Tumor
Tumor
Internal Cerivcal os
Fig. 2. Open the endocervical canal before divied from the uteri.
the study underwent an ART followed by an ARH because of a positive margin found on the second intraoperative frozen section. None of these patients received radiotherapy or chemotherapy before surgery. The patients in this study were 20–41 (median, 32) years of age. The histological subtypes were as follows: 40 squamous cell carcinomas, 11 adenocarcinomas and two adenosquamous carcinomas. There were 4 (7.55%) patients with Stage IA1 disease with LVSI, 7 (13.21%) with stage IA2 and 42 (79.25%) patients with stage IB1 disease. The pre-operative diagnoses were obtained by cervical biopsies and/or LEEP or conisation. Of the 53 included patients, 20 (37.74%) were treated with preoperative cervical conisation that showed no residual lesion; 11 (20.75%) had only nonspecific lesions and 22 (41.51%) had grossly visible lesions. All of the patient specimens were evaluated by our new combination method. The median time to frozen section was approximately 15 min (12–24 min). The LUS/EC margin was reported intraoperatively as negative in 45 cases and positive in eight cases for the first frozen section (Table 1). Of the nine positive cervical margin patients, two showed residual tumour in the second slice of cervix, who had to accept ARH and abandon any preservation of fertility? One patient accepted ARH for positive lymph node in frozen section. The other six patients had a satisfactory margin after supplementary resection that reserved the corpus uteri (Fig 3). The success rate of ART is 94.33%, (50/53). In the nine cases with positive margin, there was one patient with no visible tumour, four patients with a nonspecific lesion and four patients with grossly visible tumours. There were 15 (15/53, 28.30%) patients who received postoperative adjuvant chemotherapy. The postoperative chemotherapy or radiotherapy was given to eight of the nine positive margin patients due to parametrial lymph node metastasis and lymph vascular space
invasion with or without deeper stromal invasion. The postoperative adjuvant treatment was based on a previously published protocol [8]. There were no recurrences at a median follow-up of 14.5 months (range, 6–21 months). 4. Discussion In this study, we describe our experience with a particular protocol that was developed locally when radical trachelectomy was first implemented. The protocol is simple, reliable and produces accurate results. It is critical to ensure that there is no residual tumour of the LUS/EC margin by providing a satisfactory tumour-margin distance. Additionally, the method has been consistently followed by a group of gynaecologic pathologists for 14.5 months. We also attempt to evaluate whether this methodology might have impacted patient outcomes. There is also an urgent need for pathologists to establish guidelines and homogenous protocols regarding the handling of VRT/ART specimens. The earliest published data for margin assessment of radical trachelectomy were from Dargent’s study [5]. In their opinion, 5 mm clearance from the tumour margin was satisfactory and had a low risk of recurrence. Thus, the pathologist made an en face section 5 mm from the surgical margin for frozen section analysis that was performed tangential to the endocervical margin. This section allowed the examination of the entire margin surface. An earlier report of 61 trachelectomy specimens by Tanguay et al. [7] showed that surgical pathologists had the option of rendering their intraoperative consultation by gross inspection only or by performing a frozen section analysis, depending on whether a tumour was grossly visible [7]. If the cases had a grossly visible lesion, then a longitudinal frozen section analysis was performed. If
738
D. Zhang et al. / European Journal of Cancer 51 (2015) 734–741
Table 1 ART patients with positive endocervical margin at the first transverse section. No. Age Gross description
Size Final Lymph node Histological metastasis diagnosis
1
31
Ulcer, small
13
2
29
3
Lymph Vascular Space Invasion
Depth Additionally Treatment of resection stromal adequacy invasion
ASC
+(Parametrial) +
Yes
Irregular firm 40
AC
+(Parametrial) +
Yes
35
Endogenous 25
SCC
4
32
19
SCC
+(Parametrial) +
5
31
40
SCC
+
6 7
30 33
30 25
SCC SCC
8
20
Endogenous, ulcer Exophytic, nodular Ulcer Exophytic, firm Ulcer
30
SCC
9
23
Irregular firm 25
SCC
2/3
2/3
Yes
PTX + CBP 6 cycles ART + PLA PTX + CBP 6 cycles ART + PLA PTX + CBP 3 cycles ART + PLA + OS Radiotherapy
Yes
ART + PLA
Yes
Yes +(Pelvic) 2/3 +
ART + PLA
Adjuvant therapy
Follow Result up (month) 13
NR
13
NR
13
NR
9
NR
PTX + DDP 5 6 cycles ART + PLA 9 ARH + PLA+OS Radiotherapy 6
NR NR NR
ARH + PLA
NR
ARH + PLA
PTX + CBP 10 3 cycles PTX + CBP 5 3 cycles
NR
SCC = squamous cell carcinoma; ASC = adenosquamous carcinoma; AC = indicates adenocarcinoma; ARH = abdominal radical hysterectomy; PLA = pelvic lymphadenectomy; ART = abdominal radical trachelectomy; OS = Ovarian suspension; NR = no recurrence.
53 patients PLN examing in frozen section 52 patients PLN(-) frozen section of cerivcal margin 44 patients without tumor residual
8 patients with tumor residual second time of margin frozen section
ART performed 6 patients without tumor residual ART performed
2 patients with tumor residual converted to ARH
1 patient with PLN(+)* converted to ARH
Fig. 3. Frozen section outcomes. *The patient was found positive margin and lymph node both in frozen section, which means she had to receive abdominal radical hysterectomy (ARH) without additionally margin resection.
there was no grossly visible lesion or nonspecific lesion, then the longitudinal and transverse sections were cut perpendicular or parallel to the LUS/EC at the pathologist’s discretion. The transverse section was also taken 5 mm from the endocervical margin for frozen section analysis. A united section protocol that combined Dargent’s [5] and Tanguary’s [7] methods was presented by Park [9] in 2008. The pathologist applied both transverse
and perpendicular methods. The combined method starts by submitting an en face 2-mm-thick section of the margin to allow for evaluation of its entire circumference. This section analysis was used to examine the entire mucosal surface of the proximal margin. If this margin was positive for carcinoma, then no other frozen section analysis was performed on the specimen. Otherwise, then the cervical canal was opened and the endocervical canal was examined grossly. When a grossly visible tumour was present, its distance to the proximal end plus 2 mm (for the shave that was already taken) was reported as the margin clearance. When a grossly visible lesion was present but the nature of the lesion was uncertain (benign versus malignant), a perpendicular section including the lesion and the margin was taken for frozen-section examination. If a microscopic lesion was found the perpendicular section, the distance to the proximal edge was measured and 2 mm was added for a final margin distance. When no visible lesion was present within the canal, a second transverse cross section was taken 8 mm distal to the margin (for a total of 10 mm from the original margin) and examined. Both margins were clear, then they were reported as safe. If a positive margin was found by frozen section analysis intraoperatively (the total distance less than 10 mm), the surgeon made a wider uterine resection or performed a hysterectomy. The initial transverse sample of the endocervical margin allows evaluation of the entire surface of the endocervix so the pathologist can reliably measure the distance to the margin using additional longitudinal or transversal samples after surgery. Ismiil [6] improved the protocol that Park [9] reported in 2009 and proposed a more detailed evaluation
D. Zhang et al. / European Journal of Cancer 51 (2015) 734–741
method. Their local frozen section protocol involved examining the entire circumference of the LUS/EC margin through perpendicular serial sections in all cases. After the specimen was received in the pathology suite, the proximal 1-cm segment was cut off of the rest of the specimen and was opened to display its mucosa at 12 o’clock. It was grossly examined for the presence of any residual tumour and was then serially sliced into 10–12 sections that were 3–5-mm thick. Each of these sections was then submitted for frozen section examination. If no residual tumour was found in the frozen section analysis, the pathologist considered that the invasive tumour was more than 10 mm from the endocervical margin in the frozen section and there was a safe margin. However, if the frozen section harboured tumour tissue, the distance from the most superior neoplastic in situ or invasive cell to the LUS/EC margin was measured and reported in millimetres. In sections with a tumour clearance less than 5 mm, the LUS margin was revised and the additionally excised tissue could be examined by frozen section as requested. This method could assess the surgical margin comprehensively and accurately. However, their entire frozen section protocol required 20–30 min and a dedicated medical laboratory assistant, a pathologist assistant and a laboratory technologist to cut the frozen sections. Additionally, a gynaecologic pathologist was needed to examine and report the results. Thus, the complications of this protocol limited its promotion and application in other cancer centres. In Tanguay’s study [7], frozen section analysis was performed and 13 of the 17 VRT case patients displayed no grossly visible lesion. The residual tumour was found in one case where a longitudinal and a transverse section were taken for frozen section analysis. The incidence of positive margins was 5.88%. Similarly, in 27 cases with nonspecific lesions, the positive margin rate was 33.3%. Chenevert [10] reported a study of 53 patients with VRT cervical specimens processed by longitudinal section identical to Tanguay’s [7] and also categorised them into three different groups by their gross appearance. There was one case that displayed residual cancer in 15 patients with grossly normal specimens. This discovery is similar to results published earlier in Tanguay’s [7] study where there were no cases that showed any residual disease among 16 grossly normal VRT specimens. Using our protocol, there was one unsatisfactory margin in the first frozen section in 20 patients with no grossly visible lesions. Therefore, the incidence of positive margin was 5.00%. In 11 cases with nonspecific lesions, the positive margin rate was 36.36%, which was consistent with Tanguay’s [7] results. Thus, gross inspection alone yielded imperfect results. There was one patient with no visible tumour but positive margins, so the gross inspection is too subjective to judge in situ disease because it may result in residual tumour.
739
In Tanguay [7]’s series, residual tumour located within 5 mm of the endocervical margin was found in nine cases. The study also found one case in which the distance was 12 mm on the frozen section. After paraffin embedding, the tumour was less than 5 mm from the endocervical margin. This error was probably a result of tissue retraction after fixation. Chenevert’s 24 VRT specimens revealed a nonspecific lesion, and frozen section was performed on five cases [10]. There were only four frozen sections that were negative. One of these cases revealed SCC 8 mm from the endocervical margin in frozen section, but evaluation of paraffin-embedded tissue confirmed the presence of SCC (final distance to the endocervical margin was 4.5 mm). For this patient, the surgeon chose to intraoperatively perform an additional resection of the remaining cervix, which did not contain residual carcinoma. Thus, the final margin was satisfactory. There were no frozen sections performed on the remaining 19 VRT specimens with a nonspecific lesion. After evaluation of paraffin-embedded tissue, three of the 19 VRT specimens (15.8%) showed residual lesions. These data also verified Tanguay’s result that gross evaluation is too subjective to increase false-negatives. Park [9] proposed that the distance between the tumour and the free endocervical margin as measured on a longitudinal frozen section or transverse section was 10 mm to ensure the range and significantly reduce tumour recurrence risk. This particular report only summarised the author’s experience with endocervical adenocarcinoma in 19 patients. Compared to earlier methods, this protocol has the advantage of screening the entire circumference by the initial en face section. The evaluation scheme by gross appearance for grossly visible tumour is also too subjective to increase the risk of false-positive results. In 132 cases in Ismiil’s [6] series, there were six cases with a positive result in frozen section. There were three cases with suspicious positive results, and the frozen section versus permanent correlation was concordant in these cases. The margin was revised in 16 cases because the intraoperative assessment was positive in six cases, suspicious in two cases and negative but less than 5 mm in eight cases. The revised margin in these cases occurred during the same surgery by resecting another 5–10-mm portion of the lower uterine segment. Pre-trachelectomy LEEP or cone changed the cervical morphology and made the residual tumour indistinguishable. This caused an incorrect assessment by gross appearance, and residual tumour was always found in the cases where at least one of the margins of the prior LEEP or cone was positive. A residual tumour was found in 54 of our 132 audited cases (40.9%) compared with 39/56 (69.6%) and 7/19 (36.8%) cases in the series reported by Tanguay [7] and Park [9], respectively. The advantages of using the entire circumference of the LUS/EC margin by perpendicular section in all cases
740
D. Zhang et al. / European Journal of Cancer 51 (2015) 734–741
Table 2 Summarise data from the previously published series. Author
Year No. of FIGO patient stage
Type of surgery
Gross Safety Time appearance distance spending
Immunohis- Disadvantage tochemistry
Dargent
1993 –
VRT
No
5 mm
–
N
Tanguay 2004 56
IA1-IB1 VRT
Yes
5 mm
–
Y
Park
IA2-IB1 ART/VRT Yes
10 mm –
Y
Chevevert 2009 53
IA1-IB1 VRT
5 mm
Ismiil
IA1-IB1 ART/VRT Yes
2008 19
2009 132
–
Yes
15–20 min N
10 mm 20–30 min N
1. Insufficient tumour-margin distance 2. Lack of comparison between paraffin and frozen section 1. Insufficient tumour-margin distance 2. Too subjective to judge critical situation by gross inspection Evaluate the invasion of cervix insufficient
Advantage –
–
10 mm distance from tumour to margin determined Prove the adequacy of 10 mm distance
Too subjective to judge critical situation by gross inspection 1. Time-consuming process Evaluate the invasion 2. Professional pathologist of cervix needed comprehensive
ART = abdominal radical trachelectomy; VRT = vaginal radical trachelectomy.
described by Tanguay et al. [7] or Park et al. [9] should be considered. We only prepared two frozen sections and examined them directly for surgical margin analysis. We did not need a dedicated team of gynaecological pathologists who managed the entire intraoperative consultation service as in the study by Ismiil [11]. In our 53 audited cases, a residual tumour was found in 9 (16.98%) cases compared with 54/132 (40.9%) and 39/ 56 (69.6%) in previous reports [6,7]. The discrepancy in the tumour-residual rate varied due to the method of identifying residual tumour in the trachelectomy specimen and also depended on the clinical practice, patient selection and the amount of cervix routinely removed by the surgeon. In most cases (20/49, 40.82%) where the pre-operation LEEP or cone was available, there was no residual tumour found due to tumour excision in the initial operation. It is extremely helpful for the pathologist to perform the intraoperative consultation compared with the initial diagnosis, which was defined by pre-trachelectomy cervical biopsy. The histological sections of patients in published series are different regarding the proportions of cases with adenocarcinoma. In Tanguay’s series [7], 29% of cases had adenocarcinoma compared to 41% in the series of Park et al. [9]. The adenocarcinoma rate was 48% in Ismiil’s series [6] but only 18.37% (9/49) in our series. In their protocol, the lower uterine segment endometrium could be examined on frozen section by perpendicular section to exclude endocervical adenocarcinoma. We compared our new combination method with published methods (Table 2). We believe the adjusted technique of our protocol that examines the 10-mm circumference by parallel
section is a safer practice and is probably the most cost-effective for clinics because no professional gynaecological pathologist is required. We are hoping that this report on our experience summarises the discussion around uniformity regarding the intraoperative margins defined across institutions. 5. Conclusion In summary, we report the accuracy of our institutional intraoperative protocol with examining the 10mm circumference distance from the LUS/EC margin through an en face section in all cases of abdominal radical trachelectomy. This procedure provides the most straightforward intraoperative pathology consultation service with minor adjustments in the frozen section suite. Conflict of interest statement None declared. References [1] Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D. Global cancer statistics. CA Cancer J Clin 2011;61:69–90. [2] Dargent D. Laparoscopic surgery and gynecologic cancer. Curr Opin Obstet Gynecol 1993;5:294–300. [3] Smith J, Boyle D, Corless D, Ungar L, Lawson A, Priore GD, et al. Abdominal radical trachelectomy: a new surgical technique for the conservative management of cervical carcinoma. BJOG 1997;104:1196–200. [4] Rob L, Skapa P, Robova H. Fertility-sparing surgery in patients with cervical cancer. Lancet Oncol 2011;12:192–200.
D. Zhang et al. / European Journal of Cancer 51 (2015) 734–741 [5] Dargent D, Martin X, Sacchetoni A, Mathevet P. Laparoscopic vaginal radical trachelectomy. Cancer 2000;88:1877–82. [6] Ismiil N, Ghorab Z, Covens A, Nofech-Mozes S, Saad R, Dube´ V, et al. Intraoperative margin assessment of the radical trachelectomy specimen. Gynecol Oncol 2009;113:42–6. [7] Tanguay C, Plante M, Renaud M-C, Roy M, Teˆtu B. Vaginal radical trachelectomy in the treatment of cervical cancer: the role of frozen section. Int J Gynecol Pathol 2004;23:170–5. [8] Li J, Wu X, Li X, Ju X. Abdominal radical trachelectomy: Is it safe for IB1 cervical cancer with tumours P 2cm? Gynecol Oncol 2013;131:87–92.
741
[9] Park KJ, Soslow RA, Sonoda Y, Barakat RR, Abu-Rustum NR. Frozen-section evaluation of cervical adenocarcinoma at time of radical trachelectomy: pathologic pitfalls and the application of an objective scoring system. Gynecol Oncol 2008;110:316–23. [10] Cheˆnevert J, Teˆtu B, Plante M, Roy M, Renaud M- C, Gregoire J, et al. Indication and method of frozen section in vaginal radical trachelectomy. Int J Gynecol Pathol 2009;28:480–8. [11] Ismiil N, Ghorab Z, Nofech-Mozes S, Plotkin A, Covens A, Osborne R, et al. Intraoperative consultation in gynecologic pathology: a 6-year audit at a tertiary care medical center. Int J Gynecol Cancer 2009;19:152–7.
Available at www.sciencedirect.com
ScienceDirect journal homepage: www.ejcancer.com
A new method of surgical margin assuring for abdominal radical trachelectomy in frozen section Dandan Zhang a,c,d, Huijuan Ge b,d, Jin Li a,d, Xiaohua Wu a,d,⇑ a
Department of Gynecologyic Oncology, Fudan University Shanghai Cancer Center, Shanghai, 200032, PR China Department of Pathology, Fudan University Shanghai Cancer Center, Shanghai, 200032, PR China c Department of Obstetrics & Gynecology, Shengjing Hospital of China Medical University, Shenyang, 110004, PR China d Department of Oncology, Shanghai Medical College, Fudan University, Shanghai, 200032, PR China b
Received 4 November 2014; received in revised form 19 January 2015; accepted 28 January 2015 Available online 19 February 2015
KEYWORDS Abdominal radical trachelectomy Cervical cancer Surgical section Frozen section
Abstract Objectives: The aim of this study was to introduce a new method of assuring surgical margins for abdominal radical trachelectomy (ART) and report our experience using the method. Methods: We combined transverse and perpendicular sections to assess surgical margins of specimens from RT. All surgeries from 1st August 2012 to 1st October 2013 were performed by one surgeon. The frozen section (FS) was consistently performed by a group of gynaecologic pathologists according to the detailed protocol described in this article. All cases were prepared by the same pathologist, and the slides were reviewed by two pathologists. Results: There were 53 patients treated using the new method in our institution. The patient ages ranged from 20 to 41 years old (median 32). The surgeries were performed for clinical stage IA (n = 11) with LVSI and IB (n = 42) tumours (40 squamous cell carcinoma, 11 adenocarcinoma, two adenosquamous and two others). In 20 (37.74%) cases, no residual tumour of the ART specimen on frozen section was observed in the specimens as it was cleared by the preceding loop electrical excision procedure (LEEP) or conization. The margins were initially reported as negative in 45 cases and positive in nine cases. In those nine cases, a second slice of cervix was removed and negative in six cases and positive again in two cases, the other one with positive nodes. The results of frozen sections were concordant with the final paraffin-embedded sections. There were no false negative intraoperative assessments. There were no recurrences after a median follow-up of 15.4 months (range, 6–21 months). Conclusions: Combining transverse and perpendicular sections to assess surgical margins of specimens from RT makes the protocol simple, reliable and produces accurate results. Ó 2015 Elsevier Ltd. All rights reserved.
⇑ Corresponding author at: Department of Gynecologic Oncology, Fudan University Shanghai Cancer Center, 270 Dong’an Road, Shanghai 200032, PR China. Tel.: +86 21 6417 5590x81006; fax: +86 21 64220677. E-mail address: [email protected] (X. Wu).
http://dx.doi.org/10.1016/j.ejca.2015.01.062 0959-8049/Ó 2015 Elsevier Ltd. All rights reserved.
D. Zhang et al. / European Journal of Cancer 51 (2015) 734–741
1. Introduction Cervical cancer is the second most common cancer in women in developing countries, and more than 500,000 new cases are diagnosed worldwide every year [1]. Early-stage cervical cancer has a very good prognosis and excellent overall survival rates. Most patients suffering from early-stage cervical cancer are diagnosed at a median age of 40 years old. Due to the increased number of delayed pregnancies, many patients are diagnosed during reproductive ages. This result leads to the issue of whether it is possible to preserve the uterus for future pregnancy without increasing the risk of recurrence. In 1994, Dargent [2] reported the utility of vaginal radical trachelectomy (VRT) with laparoscopic lymph node dissection for the treatment of early-stage cervical cancer. Smith et al. [3] reported abdominal radical trachelectomy (ART) in 1997, and ART is considered another radical surgical procedure for early cervical cancer. Many studies investigating radical trachelectomy have been published worldwide in the last decade. Most of the publications have focused on the procedural clinical aspects, and there is little emphasis on the specimen handling in the pathology suite. Corpus uteri invasion is determined by the operative margin and is confirmed by intraoperative frozen section analysis. The lower uterine segment (LUS) or the upper endocervix (EC) surgical margin is the boundary of the trachelectomy specimen. The presence of tumour invasion is a very important indicator for cancer recurrence or adjuvant therapy indications, especially for patients with an endogenous tumour in the cervical canal or cervical stromal invasion. Several published reports of different techniques demonstrate a lack of guidance in the pathology community regarding the accepted guidelines for the most reliable and feasible approach. There is no existing agreement on the best method of sampling a margin of a frozen section. The selection of an appropriate method is dictated by the information required by the surgeon. This study was performed in our centre and presents a new method of frozen section analysis for radical trachelectomy. The new method has the following advantages: (1) it is simple, convenient and less time consuming; (2) no professional gynaecological pathologist is required so the method is easier to popularise; and (3) it ensures margins without residual tumour. 2. Materials and methods A prospectively maintained database of early-stage cervical cancer patients who were scheduled to undergo abdominal radical trachelectomy was analysed. All of the operations were performed in the Department of Gynecologic Oncology at Fudan University Shanghai Cancer Center from 1st August 2012 to 1st October
735
2013 by the same surgeon. The initial diagnosis was made on cervical biopsies, loop electrical excision procedure (LEEP) specimens, or cone biopsy specimens. The criteria for preserving the corpus uteri and part of the cervix were as follows: (1) desire to preserve fertility and no clinical evidence of impaired fertility; (2) histologic diagnosis of squamous cell carcinoma, adenocarcinoma, or adenosquamous carcinoma; (3) FIGO stage IA1 disease with lymphovascular space invasion, or positive surgical margin and distorted cervicovaginal anatomy after conisation, and stage IA2 or IB1 disease; (4) tumour size less than 4 cm; (5) no evidence of parametrial involvement and confirmation of tumour limited to the cervix; and (6) no evidence of pelvic lymph node metastases [4]. There are different approaches for submitting sections of the trachelectomy margin for histologic evaluation. However, there is no defined consensus on the best sampling method. Surgeons may prefer transverse (en face) sections [5], while others might prefer a longitudinal (perpendicular) section [6]. Others only take a perpendicular margin if there is a grossly visible tumour [7]. At our institution, we applied both transverse and perpendicular methods. The new combination method is briefly described in the following four steps. Step 1: When the specimen was sliced from the corpus uteri, the endocervical canal was opened through a longitudinal anterior section at the 12:00 location (Fig 1A). The shape, size and location of the tumour were examined by an unaided eye, described and recorded in the final pathological report. The tumour form in ART specimens was graded into three categories by the gross appearance: (1) no grossly visible lesion, such that the tumour could not be clearly located due to preoperative conisation or LEEP; (2) a nonspecific lesion, which was defined as a granular, irregular, or ulcerated zone that may correspond to the previous biopsy site and in which a gross diagnosis of residual cancer was uncertain; and (3) a grossly visible tumour for which a tumour was clearly identifiable. Step 2: If the tumour was previously visible or coned, the pathologist then cut off the cervical specimen 8 mm from the margin and cut away a 2-mm-thick en face slice. This requires making a transverse section 10 mm from the end of the specimen for the frozen section (Fig 1B). Step 3: The section was divided into two slices, which were marked 0–6 o’clock and 6–12 o’clock. Each slice was checked from the mucosal surface to the serosa using a microscope (Fig 1C). The most important aspect of the method is examining the whole endocervical stroma including endocervical clefts deep within the stroma. If no residual tumour was found in the frozen section, then there was a satisfactory margin with no invasive tumour at least 10 mm from the tumour.
736
D. Zhang et al. / European Journal of Cancer 51 (2015) 734–741
A
B 12 o’clock
8mm
0 o’clock
10mm
12 o’clock
C 6-12 o’clock
0-6 o’clock 6-12 o’clock
0-6 o’clock
2mm frozen section 12 o’clock
0 o’clock
D The additionally excised tissue 10mm The second en face The first en face
Fig. 1. Schematic representation of how to submit trachelectomy margin at the time of frozen-section. (A) Longitudinal open the endocervical canal at 12 o’clock. (B) Transverse section, approximately 2 mm in thickness, 8 mm distal to surgical margin. (C) Divide the section into two slices, marked 0–6 and 6–12 o’clock respectively. (D) If residual tumour is found, the margin is revised, and transverse section is submitted 8 mm distal to revised margin.
Step 4: If the frozen section presented tumour tissue and the tumour clearance is less than 10 mm, then the LUS/EC margin had to be revised. The thickness of additional excised tissue was more than 5 mm. Then, a 2-mm en face cut of the cervical canal was performed, which ensured that the distance from the second transverse section to the additional proximal resection margin was 10 mm (Fig 1D). The en face tissue was examined by frozen section analysis using the same protocol. If the second transverse section showed tumour infiltration or the tumour-margin range was still unsatisfactory, the ART was abandoned and the patient received a radical hysterectomy. The surgeons had informed the patients of this risk during the pre-operative period. The final evaluations of paraffin-embedded tissue and the ART specimens were submitted again after formalin fixation. We improved the first step of the procedure to ensure the tumour-margin distance. Before slicing the specimen from the uteri, the surgeon quarantined the cervix from the surgical sterile region and cut the endocervical canal at the 12:00 location through a longitudinal anterior section (Fig 2). The cervix was then sliced from the corpus uteri at the estimated position, which was determined by
the gross 10 mm separation in the patulous cervix. The divided specimen was also checked, and en face slices were prepared as described in the second step of the original protocol. The patients with positive pelvic lymph nodes on final pathology after ART were recommended to undergo adjuvant radiation with concurrent platinum-based chemotherapy. The patients who had deep stromal infiltration (DSI) or lymphovascular space invasion (LVSI) on final pathology underwent adjuvant chemotherapy. The specific treatment regimens have been published previously. 3. Results Between August 2012 and October 2013, a total of 53 patients underwent a laparotomy for a planned fertilitysparing ART and pelvic lymphadenectomy. All of the patients received a preoperative MRI to exclude endocervical canal involvement and pelvic metastasis. The procedure was converted to an abdominal radical hysterectomy (ARH) in three of the 53 patients who were considered for ART. There was one patient with positive pelvic lymph nodes, and two patients included in
D. Zhang et al. / European Journal of Cancer 51 (2015) 734–741
737
Tumor
Tumor
Internal Cerivcal os
Fig. 2. Open the endocervical canal before divied from the uteri.
the study underwent an ART followed by an ARH because of a positive margin found on the second intraoperative frozen section. None of these patients received radiotherapy or chemotherapy before surgery. The patients in this study were 20–41 (median, 32) years of age. The histological subtypes were as follows: 40 squamous cell carcinomas, 11 adenocarcinomas and two adenosquamous carcinomas. There were 4 (7.55%) patients with Stage IA1 disease with LVSI, 7 (13.21%) with stage IA2 and 42 (79.25%) patients with stage IB1 disease. The pre-operative diagnoses were obtained by cervical biopsies and/or LEEP or conisation. Of the 53 included patients, 20 (37.74%) were treated with preoperative cervical conisation that showed no residual lesion; 11 (20.75%) had only nonspecific lesions and 22 (41.51%) had grossly visible lesions. All of the patient specimens were evaluated by our new combination method. The median time to frozen section was approximately 15 min (12–24 min). The LUS/EC margin was reported intraoperatively as negative in 45 cases and positive in eight cases for the first frozen section (Table 1). Of the nine positive cervical margin patients, two showed residual tumour in the second slice of cervix, who had to accept ARH and abandon any preservation of fertility? One patient accepted ARH for positive lymph node in frozen section. The other six patients had a satisfactory margin after supplementary resection that reserved the corpus uteri (Fig 3). The success rate of ART is 94.33%, (50/53). In the nine cases with positive margin, there was one patient with no visible tumour, four patients with a nonspecific lesion and four patients with grossly visible tumours. There were 15 (15/53, 28.30%) patients who received postoperative adjuvant chemotherapy. The postoperative chemotherapy or radiotherapy was given to eight of the nine positive margin patients due to parametrial lymph node metastasis and lymph vascular space
invasion with or without deeper stromal invasion. The postoperative adjuvant treatment was based on a previously published protocol [8]. There were no recurrences at a median follow-up of 14.5 months (range, 6–21 months). 4. Discussion In this study, we describe our experience with a particular protocol that was developed locally when radical trachelectomy was first implemented. The protocol is simple, reliable and produces accurate results. It is critical to ensure that there is no residual tumour of the LUS/EC margin by providing a satisfactory tumour-margin distance. Additionally, the method has been consistently followed by a group of gynaecologic pathologists for 14.5 months. We also attempt to evaluate whether this methodology might have impacted patient outcomes. There is also an urgent need for pathologists to establish guidelines and homogenous protocols regarding the handling of VRT/ART specimens. The earliest published data for margin assessment of radical trachelectomy were from Dargent’s study [5]. In their opinion, 5 mm clearance from the tumour margin was satisfactory and had a low risk of recurrence. Thus, the pathologist made an en face section 5 mm from the surgical margin for frozen section analysis that was performed tangential to the endocervical margin. This section allowed the examination of the entire margin surface. An earlier report of 61 trachelectomy specimens by Tanguay et al. [7] showed that surgical pathologists had the option of rendering their intraoperative consultation by gross inspection only or by performing a frozen section analysis, depending on whether a tumour was grossly visible [7]. If the cases had a grossly visible lesion, then a longitudinal frozen section analysis was performed. If
738
D. Zhang et al. / European Journal of Cancer 51 (2015) 734–741
Table 1 ART patients with positive endocervical margin at the first transverse section. No. Age Gross description
Size Final Lymph node Histological metastasis diagnosis
1
31
Ulcer, small
13
2
29
3
Lymph Vascular Space Invasion
Depth Additionally Treatment of resection stromal adequacy invasion
ASC
+(Parametrial) +
Yes
Irregular firm 40
AC
+(Parametrial) +
Yes
35
Endogenous 25
SCC
4
32
19
SCC
+(Parametrial) +
5
31
40
SCC
+
6 7
30 33
30 25
SCC SCC
8
20
Endogenous, ulcer Exophytic, nodular Ulcer Exophytic, firm Ulcer
30
SCC
9
23
Irregular firm 25
SCC
2/3
2/3
Yes
PTX + CBP 6 cycles ART + PLA PTX + CBP 6 cycles ART + PLA PTX + CBP 3 cycles ART + PLA + OS Radiotherapy
Yes
ART + PLA
Yes
Yes +(Pelvic) 2/3 +
ART + PLA
Adjuvant therapy
Follow Result up (month) 13
NR
13
NR
13
NR
9
NR
PTX + DDP 5 6 cycles ART + PLA 9 ARH + PLA+OS Radiotherapy 6
NR NR NR
ARH + PLA
NR
ARH + PLA
PTX + CBP 10 3 cycles PTX + CBP 5 3 cycles
NR
SCC = squamous cell carcinoma; ASC = adenosquamous carcinoma; AC = indicates adenocarcinoma; ARH = abdominal radical hysterectomy; PLA = pelvic lymphadenectomy; ART = abdominal radical trachelectomy; OS = Ovarian suspension; NR = no recurrence.
53 patients PLN examing in frozen section 52 patients PLN(-) frozen section of cerivcal margin 44 patients without tumor residual
8 patients with tumor residual second time of margin frozen section
ART performed 6 patients without tumor residual ART performed
2 patients with tumor residual converted to ARH
1 patient with PLN(+)* converted to ARH
Fig. 3. Frozen section outcomes. *The patient was found positive margin and lymph node both in frozen section, which means she had to receive abdominal radical hysterectomy (ARH) without additionally margin resection.
there was no grossly visible lesion or nonspecific lesion, then the longitudinal and transverse sections were cut perpendicular or parallel to the LUS/EC at the pathologist’s discretion. The transverse section was also taken 5 mm from the endocervical margin for frozen section analysis. A united section protocol that combined Dargent’s [5] and Tanguary’s [7] methods was presented by Park [9] in 2008. The pathologist applied both transverse
and perpendicular methods. The combined method starts by submitting an en face 2-mm-thick section of the margin to allow for evaluation of its entire circumference. This section analysis was used to examine the entire mucosal surface of the proximal margin. If this margin was positive for carcinoma, then no other frozen section analysis was performed on the specimen. Otherwise, then the cervical canal was opened and the endocervical canal was examined grossly. When a grossly visible tumour was present, its distance to the proximal end plus 2 mm (for the shave that was already taken) was reported as the margin clearance. When a grossly visible lesion was present but the nature of the lesion was uncertain (benign versus malignant), a perpendicular section including the lesion and the margin was taken for frozen-section examination. If a microscopic lesion was found the perpendicular section, the distance to the proximal edge was measured and 2 mm was added for a final margin distance. When no visible lesion was present within the canal, a second transverse cross section was taken 8 mm distal to the margin (for a total of 10 mm from the original margin) and examined. Both margins were clear, then they were reported as safe. If a positive margin was found by frozen section analysis intraoperatively (the total distance less than 10 mm), the surgeon made a wider uterine resection or performed a hysterectomy. The initial transverse sample of the endocervical margin allows evaluation of the entire surface of the endocervix so the pathologist can reliably measure the distance to the margin using additional longitudinal or transversal samples after surgery. Ismiil [6] improved the protocol that Park [9] reported in 2009 and proposed a more detailed evaluation
D. Zhang et al. / European Journal of Cancer 51 (2015) 734–741
method. Their local frozen section protocol involved examining the entire circumference of the LUS/EC margin through perpendicular serial sections in all cases. After the specimen was received in the pathology suite, the proximal 1-cm segment was cut off of the rest of the specimen and was opened to display its mucosa at 12 o’clock. It was grossly examined for the presence of any residual tumour and was then serially sliced into 10–12 sections that were 3–5-mm thick. Each of these sections was then submitted for frozen section examination. If no residual tumour was found in the frozen section analysis, the pathologist considered that the invasive tumour was more than 10 mm from the endocervical margin in the frozen section and there was a safe margin. However, if the frozen section harboured tumour tissue, the distance from the most superior neoplastic in situ or invasive cell to the LUS/EC margin was measured and reported in millimetres. In sections with a tumour clearance less than 5 mm, the LUS margin was revised and the additionally excised tissue could be examined by frozen section as requested. This method could assess the surgical margin comprehensively and accurately. However, their entire frozen section protocol required 20–30 min and a dedicated medical laboratory assistant, a pathologist assistant and a laboratory technologist to cut the frozen sections. Additionally, a gynaecologic pathologist was needed to examine and report the results. Thus, the complications of this protocol limited its promotion and application in other cancer centres. In Tanguay’s study [7], frozen section analysis was performed and 13 of the 17 VRT case patients displayed no grossly visible lesion. The residual tumour was found in one case where a longitudinal and a transverse section were taken for frozen section analysis. The incidence of positive margins was 5.88%. Similarly, in 27 cases with nonspecific lesions, the positive margin rate was 33.3%. Chenevert [10] reported a study of 53 patients with VRT cervical specimens processed by longitudinal section identical to Tanguay’s [7] and also categorised them into three different groups by their gross appearance. There was one case that displayed residual cancer in 15 patients with grossly normal specimens. This discovery is similar to results published earlier in Tanguay’s [7] study where there were no cases that showed any residual disease among 16 grossly normal VRT specimens. Using our protocol, there was one unsatisfactory margin in the first frozen section in 20 patients with no grossly visible lesions. Therefore, the incidence of positive margin was 5.00%. In 11 cases with nonspecific lesions, the positive margin rate was 36.36%, which was consistent with Tanguay’s [7] results. Thus, gross inspection alone yielded imperfect results. There was one patient with no visible tumour but positive margins, so the gross inspection is too subjective to judge in situ disease because it may result in residual tumour.
739
In Tanguay [7]’s series, residual tumour located within 5 mm of the endocervical margin was found in nine cases. The study also found one case in which the distance was 12 mm on the frozen section. After paraffin embedding, the tumour was less than 5 mm from the endocervical margin. This error was probably a result of tissue retraction after fixation. Chenevert’s 24 VRT specimens revealed a nonspecific lesion, and frozen section was performed on five cases [10]. There were only four frozen sections that were negative. One of these cases revealed SCC 8 mm from the endocervical margin in frozen section, but evaluation of paraffin-embedded tissue confirmed the presence of SCC (final distance to the endocervical margin was 4.5 mm). For this patient, the surgeon chose to intraoperatively perform an additional resection of the remaining cervix, which did not contain residual carcinoma. Thus, the final margin was satisfactory. There were no frozen sections performed on the remaining 19 VRT specimens with a nonspecific lesion. After evaluation of paraffin-embedded tissue, three of the 19 VRT specimens (15.8%) showed residual lesions. These data also verified Tanguay’s result that gross evaluation is too subjective to increase false-negatives. Park [9] proposed that the distance between the tumour and the free endocervical margin as measured on a longitudinal frozen section or transverse section was 10 mm to ensure the range and significantly reduce tumour recurrence risk. This particular report only summarised the author’s experience with endocervical adenocarcinoma in 19 patients. Compared to earlier methods, this protocol has the advantage of screening the entire circumference by the initial en face section. The evaluation scheme by gross appearance for grossly visible tumour is also too subjective to increase the risk of false-positive results. In 132 cases in Ismiil’s [6] series, there were six cases with a positive result in frozen section. There were three cases with suspicious positive results, and the frozen section versus permanent correlation was concordant in these cases. The margin was revised in 16 cases because the intraoperative assessment was positive in six cases, suspicious in two cases and negative but less than 5 mm in eight cases. The revised margin in these cases occurred during the same surgery by resecting another 5–10-mm portion of the lower uterine segment. Pre-trachelectomy LEEP or cone changed the cervical morphology and made the residual tumour indistinguishable. This caused an incorrect assessment by gross appearance, and residual tumour was always found in the cases where at least one of the margins of the prior LEEP or cone was positive. A residual tumour was found in 54 of our 132 audited cases (40.9%) compared with 39/56 (69.6%) and 7/19 (36.8%) cases in the series reported by Tanguay [7] and Park [9], respectively. The advantages of using the entire circumference of the LUS/EC margin by perpendicular section in all cases
740
D. Zhang et al. / European Journal of Cancer 51 (2015) 734–741
Table 2 Summarise data from the previously published series. Author
Year No. of FIGO patient stage
Type of surgery
Gross Safety Time appearance distance spending
Immunohis- Disadvantage tochemistry
Dargent
1993 –
VRT
No
5 mm
–
N
Tanguay 2004 56
IA1-IB1 VRT
Yes
5 mm
–
Y
Park
IA2-IB1 ART/VRT Yes
10 mm –
Y
Chevevert 2009 53
IA1-IB1 VRT
5 mm
Ismiil
IA1-IB1 ART/VRT Yes
2008 19
2009 132
–
Yes
15–20 min N
10 mm 20–30 min N
1. Insufficient tumour-margin distance 2. Lack of comparison between paraffin and frozen section 1. Insufficient tumour-margin distance 2. Too subjective to judge critical situation by gross inspection Evaluate the invasion of cervix insufficient
Advantage –
–
10 mm distance from tumour to margin determined Prove the adequacy of 10 mm distance
Too subjective to judge critical situation by gross inspection 1. Time-consuming process Evaluate the invasion 2. Professional pathologist of cervix needed comprehensive
ART = abdominal radical trachelectomy; VRT = vaginal radical trachelectomy.
described by Tanguay et al. [7] or Park et al. [9] should be considered. We only prepared two frozen sections and examined them directly for surgical margin analysis. We did not need a dedicated team of gynaecological pathologists who managed the entire intraoperative consultation service as in the study by Ismiil [11]. In our 53 audited cases, a residual tumour was found in 9 (16.98%) cases compared with 54/132 (40.9%) and 39/ 56 (69.6%) in previous reports [6,7]. The discrepancy in the tumour-residual rate varied due to the method of identifying residual tumour in the trachelectomy specimen and also depended on the clinical practice, patient selection and the amount of cervix routinely removed by the surgeon. In most cases (20/49, 40.82%) where the pre-operation LEEP or cone was available, there was no residual tumour found due to tumour excision in the initial operation. It is extremely helpful for the pathologist to perform the intraoperative consultation compared with the initial diagnosis, which was defined by pre-trachelectomy cervical biopsy. The histological sections of patients in published series are different regarding the proportions of cases with adenocarcinoma. In Tanguay’s series [7], 29% of cases had adenocarcinoma compared to 41% in the series of Park et al. [9]. The adenocarcinoma rate was 48% in Ismiil’s series [6] but only 18.37% (9/49) in our series. In their protocol, the lower uterine segment endometrium could be examined on frozen section by perpendicular section to exclude endocervical adenocarcinoma. We compared our new combination method with published methods (Table 2). We believe the adjusted technique of our protocol that examines the 10-mm circumference by parallel
section is a safer practice and is probably the most cost-effective for clinics because no professional gynaecological pathologist is required. We are hoping that this report on our experience summarises the discussion around uniformity regarding the intraoperative margins defined across institutions. 5. Conclusion In summary, we report the accuracy of our institutional intraoperative protocol with examining the 10mm circumference distance from the LUS/EC margin through an en face section in all cases of abdominal radical trachelectomy. This procedure provides the most straightforward intraoperative pathology consultation service with minor adjustments in the frozen section suite. Conflict of interest statement None declared. References [1] Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D. Global cancer statistics. CA Cancer J Clin 2011;61:69–90. [2] Dargent D. Laparoscopic surgery and gynecologic cancer. Curr Opin Obstet Gynecol 1993;5:294–300. [3] Smith J, Boyle D, Corless D, Ungar L, Lawson A, Priore GD, et al. Abdominal radical trachelectomy: a new surgical technique for the conservative management of cervical carcinoma. BJOG 1997;104:1196–200. [4] Rob L, Skapa P, Robova H. Fertility-sparing surgery in patients with cervical cancer. Lancet Oncol 2011;12:192–200.
D. Zhang et al. / European Journal of Cancer 51 (2015) 734–741 [5] Dargent D, Martin X, Sacchetoni A, Mathevet P. Laparoscopic vaginal radical trachelectomy. Cancer 2000;88:1877–82. [6] Ismiil N, Ghorab Z, Covens A, Nofech-Mozes S, Saad R, Dube´ V, et al. Intraoperative margin assessment of the radical trachelectomy specimen. Gynecol Oncol 2009;113:42–6. [7] Tanguay C, Plante M, Renaud M-C, Roy M, Teˆtu B. Vaginal radical trachelectomy in the treatment of cervical cancer: the role of frozen section. Int J Gynecol Pathol 2004;23:170–5. [8] Li J, Wu X, Li X, Ju X. Abdominal radical trachelectomy: Is it safe for IB1 cervical cancer with tumours P 2cm? Gynecol Oncol 2013;131:87–92.
741
[9] Park KJ, Soslow RA, Sonoda Y, Barakat RR, Abu-Rustum NR. Frozen-section evaluation of cervical adenocarcinoma at time of radical trachelectomy: pathologic pitfalls and the application of an objective scoring system. Gynecol Oncol 2008;110:316–23. [10] Cheˆnevert J, Teˆtu B, Plante M, Roy M, Renaud M- C, Gregoire J, et al. Indication and method of frozen section in vaginal radical trachelectomy. Int J Gynecol Pathol 2009;28:480–8. [11] Ismiil N, Ghorab Z, Nofech-Mozes S, Plotkin A, Covens A, Osborne R, et al. Intraoperative consultation in gynecologic pathology: a 6-year audit at a tertiary care medical center. Int J Gynecol Cancer 2009;19:152–7.
