Background The need for radical surgery followed by adjuvant chemoradiation may be reduced by abandoning radical surgery in patients in whom lymph node involvement is detected intra-operatively.
Objectives To analyze, in a retrospective cohort study, the efficacy of the algorithm using intra-operative pathological assessment of sentinel lymph nodes.
Methods A retrospective single-institution study was carried out, which analyzed data from all consecutive patients with cervical cancer who were referred for primary surgical treatment between May 2005 and December 2015. Inclusion criteria were as follows: (1) TNM stage T1a1 with lymphovascular space invasion, T1a2, T1b, T2a, and selected T2b with incipient parametrial invasion; (2) adenocarcinoma, squamous cell carcinoma, or adenosquamous carcinoma; (3) no evidence of enlarged suspicious nodes or distant metastases on pre-operative imaging; (4) primary surgery with curative intent; (5) successful detection of sentinel lymph node, at least, unilaterally. All patients had at least one sentinel lymph node detected and submitted for frozen section evaluation. When sentinel lymph node involvement was detected intra-operatively, the cervical procedure was abandoned and the patient was referred for definitive chemoradiation. Radical surgery was completed in patients with intra-operative negative sentinel lymph nodes. The reliability of intra-operative sentinel lymph node assessment was evaluated by calculating the sensitivity, specificity, positive predictive value, negative predictive value, positive likelihood ratio, and negative likelihood ratio.
Results The study included a total of 309 patients. Sentinel lymph nodes were detected bilaterally in 86% of the patients. Lymph node positivity was detected intra-operatively in 18 (6%) patients in whom the cervical procedure was abandoned. Adjuvant radiotherapy after completed radical surgery was given to 29 (9%) patients, including 20 patients with macrometastases (8) or micrometastases (12) reported from the final histology, eight patients with positive parametria (all ≤3 mm), and one patient with a positive vaginal resection margin. The sensitivity, specificity, positive predictive value, and negative predictive value for the intra-operative detection of lymph node positivity (macrometastases or micrometastases) was 47% (95% CI 31% to 64%), 100%, 100%, and 93% (95% CI 90% to 96%), respectively. A total of 18 (6%) patients were spared combined treatment owing to the intra-operative sentinel lymph node triage; 29 patients (9%) received combined treatment with both radical surgery and adjuvant radiotherapy
Conclusions Of 47 patients with high-risk prognostic risk factors (lymph node, parametria, or surgical margin involvement), combined treatment was successfully avoided in 18 (38%). Despite an effort to triage the patients intra-operatively, 9% received a combination of cervical procedure and adjuvant chemoradiation, mostly owing to the low sensitivity of the frozen section in the detection of micrometastases and macrometastases.
- cervical cancer
- postoperative complications
- SLN and lympadenectomy
Statistics from Altmetric.com
Half of lymph node-positive patients are spared from combination of surgery and radiotherapy due to intraoperative sentinel lymph node assessment.
Combined treatment was administered mostly due to the low sensitivity of sentinel lymph node frozen section for the detection of micrometastases.
The second reason was the low accuracy of preoperative imaging in the detection of microscopic parametrial involvement.
Lymph node metastasis has long been identified as a major negative prognostic factor in early-stage cervical cancer.1 It represents, together with the involvement of parametria or vaginal resection margin, the most common indication for adjuvant radiotherapy after radical surgery. Radical surgery and radiation treatment are associated with different adverse events, and undoubtedly their combination results in an increased risk of complications compared with either treatment modality alone. Some of these, such as urinary fistulas or stenoses, are mostly associated with combined treatment when ureters and urinary bladder are further exposed to radiation.2
Implementation of a 'one-step protocol' of sentinel lymph node identification and intra-operative assessment opens the possibility of receiving information about pelvic nodal status before the cervical procedure (simple/radical hysterectomy or fertility-sparing surgery, such as conization or simple/radical trachelectomy) is performed. If sentinel lymph node positivity is detected intra-operatively, radical surgery may be abandoned, and the patient then referred for primary chemoradiation and spared from receiving combined treatment.3
The aim of this retrospective study was to analyze data from a single institution where the algorithm using intra-operative one-step sentinel lymph node triage has been used in the majority of treated patients with early-stage cervical cancer since 2004. We assessed the number of patients who were spared combined treatment and who, despite this effort, received it.
We conducted a retrospective single-institution study that analyzed data from all consecutive patients with cervical cancer who were referred for primary surgical treatment between May 2005 and December 2015. The inclusion criteria were as follows: (1) TNM stage T1a1 with lymphovascular space invasion, T1a2, T1b, T2a, and selected T2b with incipient parametrial invasion; (2) adenocarcinoma, squamous cell carcinoma, or adenosquamous carcinoma; (3) no evidence of enlarged suspicious nodes or distant metastases on pre-operative imaging; (4) primary surgery with curative intent; (5) successful detection of sentinel lymph node, at least, unilaterally. Patients who received neoadjuvant chemotherapy were also included if lymph node staging was performed as the first management step, before the administration of chemotherapy.
Local clinical staging consisted of a physical examination and expert ultrasound examination in combination with pelvic MRI in selected cases. PET/CT or CT was performed as a distant staging in patients with locally advanced tumors or when suspicious pelvic lymph nodes had been found in local staging. The study protocol was approved by the local ethics committee.
Sentinel Lymph Node Detection
A dual tracer technique was used for sentinel lymph node identification. Patients received an injection of radioactive colloid (9mTc; SentiScint, MediRadiopharma Ltd, Hungary; 4×20 MBq) the day before the surgery. Blue dye (Patent blue V, Byk Gueden, Germany or Blue Patente V sodium Guerbet, Roissy, France) was applied immediately before the surgery. Both radio-colloid and blue dye were injected at the four quadrants of the cervix.4 In larger tumors, the technique of tracer application was modified as described previously5; and the site and depth of the injection were adjusted according to the tumor topography verified by ultrasound. All sentinel lymph nodes were sent for intra-operative assessment.
All surgeries were performed by open laparotomy. Sentinel lymph nodes were identified after the opening of the retro-peritoneum at the beginning of each surgery. All blue nodes and/or radioactive nodes were considered to be sentinel lymph nodes and were submitted for frozen section assessment. If macrometastases or micrometastases were detected intra-operatively, the cervical procedure was abandoned and para-aortic lymph node dissection was completed. The remaining patients underwent the cervical procedure, including conization or trachelectomy (29), simple hysterectomy (14), radical trachelectomy (11), or radical hysterectomy (237). Systematic pelvic lymphadenectomy was performed in 296 patients (96%) while sentinel lymph node biopsy was the only lymph node staging in 13 patients. Of these 13 patients, 10 had a T1a tumor (complete pelvic lymph node dissection was not indicated), and three with a T1b1 tumor were enrolled in the prospective study on sentinel lymph node biopsy without pelvic retro-peritoneum (SENTIX).6
Fresh tissue labeled sentinel lymph node was delivered without fixation for intra-operative evaluation. The specimens were examined to determine the size and number of individual nodes. Grossly metastatic nodes were sectioned. The sentinel lymph nodes that appeared normal were cut perpendicularly to their long axis, and one half of each node was then examined in frozen section on one level after staining with hematoxylin and eosin (H&E). The other half of the node was immediately fixed. Finally, both sentinel lymph node halves were sectioned at 2 mm intervals, and the entire sentinel lymph node was submitted for routine processing and H&E staining. Any sentinel lymph node found to be negative on routine examination was further examined by an ultrastaging protocol.
This protocol consisted of two consecutive sections (4 μm thick) obtained in regular 150 µm intervals, which were cut from each paraffin block in four levels. The first section was stained with H&E and the second section was examined immunohistochemically with antibody against cytokeratins (AE1/AE3; 1:50 dilution, Dako, Glostrup, Denmark). An immunohistochemical examination was performed using the avidin-biotin complex method (Ventana ES autostainer, Ventana, Medical Systems, Tucson, Arizona, USA). In patients enrolled in the prospective SENTIX trial, sentinel lymph nodes were processed completely in regular 150 µm intervals according to the study protocol.6
Metastases were classified according to their maximum diameter as macrometastases (tumor deposit greater than 2 mm), micrometastases (tumor deposit greater than 0.2 and up to 2 mm), and isolated tumor cells, as individual tumor cells or small clusters of cells <0.2 mm in diameter. Macrometastases and micrometastases were further reported as pN1 (pN1(mi) for micrometastases) and isolated tumor cells, as pN0.7
Further Clinical Management
Patients with positive lymph nodes, infiltration of vaginal margin, or infiltration of parametria were referred for adjuvant chemoradiation. Extended-field radiotherapy was administered in cases with positive para-aortic or common iliac lymph node. In our institution, a combination of intermediate risk factors1 was not considered to be an indication for adjuvant treatment.
Standard measures of summary statistics were used to describe primary data: relative and absolute frequencies and arithmetic mean supplied with the SD of mean. The reliability of intra-operative sentinel lymph node assessment was evaluated by calculating the sensitivity, specificity, positive predictive value, negative predictive value, positive likelihood ratio, and negative likelihood ratio The result of the frozen section was used as a unit of analysis, and the final pathology result (of both sentinel lymph node and pelvic lymph node) as a reference standard method. All 95% confidence intervals (CIs) for proportions were estimated using the exact binomial distribution. A Χ2 test was used to compare the groups in the parametric categories and the Kruskal-Wallis test to compare the groups in the categories in which the continual variables were provided. A value of p=0.05 was used as the limit of statistical significance in all the analyses performed.
In all, 309 patients were retrospectively evaluated in this study. Although more than three-quarters (239 (77%)) of patients had early-stage disease (T1a and T1b1), 50 patients (16%) had tumors >4 cm (T1b2), six patients (2%) had involvement of the vaginal cuff (T2a), and 14 patients (5%) had infiltration of the parametria (T2b). The characteristics of the total population and sub-groups of patients with nodal metastases detected either intra-operatively or from the final histology are summarized in Table 1.
Sentinel lymph nodes were detected bilaterally in 86% of the patients. The metastatic spread was reported intra-operatively in 18 patients: 16 patients with macrometastases and two patients with micrometastases. The cervical procedure was abandoned in all of them. An additional 30 patients had metastasis in pelvic lymph nodes and/or sentinel lymph node reported from the final pathology: eight with macrometastases, 12 with micrometastases, and 10 with isolated tumor cells (see Table 2).
The sensitivity, specificity, positive predictive value, and negative predictive value for the intra-operative detection of lymph node positivity (macrometastases or micrometastases) were 47% (95% CI 31% to 64%), 100%, 100%, and 93% (95% CI 0% to 96%), respectively. The negative likelihood ratio was 0.53 (95% CI 0.03 to 0.98). The sensitivity, specificity, positive predictive value, and negative predictive value for the detection of macrometastases only were 67% (95% CI 45% to 84%), 100%, 100%, and 97% (95% CI 95% to 99%), respectively. The negative likelihood ratio was 0.33 (95% CI 0.03 to 0.98). Positive likelihood ratios were infinite because there was no false-positive cases.
All patients in whom the cervical procedure was intra-operatively abandoned received definitive chemoradiation. In the entire group of 309 patients, 18 (6%) patients were spared a combined treatment thanks to the intra-operative sentinel lymph node triage. Twenty-nine patients (9%) received combined treatment of both radical surgery and adjuvant radiotherapy, including 20 patients with negative intra-operative assessment of sentinel lymph node, but positive lymph node on final histology, eight patients with infiltration of parametria, and one patient with the involvement of vaginal resection margin (Online supplementary figure 1). In all cases with positive parametria the invasion was <3 mm.
In this retrospective cohort study of 309 patients treated by surgery for early-stage cervical cancer, 18 patients (6%) were spared combined treatment owing to the one-step sentinel lymph node intra-operative assessment protocol. Despite all efforts to triage patients to a single major treatment modality, 29 (9%) patients received adjuvant radiotherapy after radical surgery; the majority of these patients (20) received radiotherapy owing to a failure to detect lymph node involvement intra-operatively.
Adjuvant chemoradiation is considered in two cohorts of patients after radical surgery. It is widely accepted for patients with 'high-risk' prognostic factors, such as positive lymph nodes, positive parametria, and positive surgical margins, but it is not uniformly accepted for a cohort with 'intermediate risk' and a combination of tumor-related traditional risk factors, including tumor size, depth of stromal invasion, and presence of lymphovascular space invasion.8 According to the new European Society of Gynaecological Oncology, European Societyfor Radiotherapy and Oncology, and the European Society ofPathology Guidelines for the Management of Patients with Cervical Cancer, management of early stage cervical cancer should aim to avoid combining radical surgery and radiotherapy because of the highest morbidity after a combined treatment.3 Intra-operative pathologic assessment of lymph node involvement represents an opportunity to identify additional patients not detected pre-operatively by imaging. Radical parametrectomy may be abandoned, the patient referred for primary chemoradiation and spared combined therapy.
The feasibility of such a strategy has already been evaluated by several groups. Only 39 patients (9%) received combined treatment in a prospective study of 448 patients, in which sentinel lymph node biopsy was not performed, but all pelvic lymph nodes were sent for frozen section. The cervical procedure was abandoned when lymph node positivity was reported intra-operatively. Of those who received combined treatment, lymph node positivity was an indication, for adjuvant radiotherapy in 28% (sensitivity of frozen section in detection of macrometastases and micrometastases reached 83%), of other high-risk factors (parametrial invasion, positive surgical margins) in 36%, and of intermediate risk factors in 36% of patients.9 Even higher frozen section sensitivity (87%) for the detection of lymph node metastases was reported in a similarly designed trial published in 2016.10 It is important to emphasize that the high sensitivity of intra-operative lymph node assessment reported in these papers has been attributed to a low prevalence of micrometastases in the absence of sentinel lymph node ultrastaging. The probability of detecting small macrometastases and micrometastases rises substantially when a sentinel lymph node is detected and processed in the pathological ultrastaging protocol.11
In trials in which sentinel lymph node biopsy was included in the management, the sensitivity of the sentinel lymph node frozen section for the final lymph node status varied considerably, from 33% to 100%.12–15 In our earlier study, we reported 81% frozen section sensitivity for detection of macrometastases but only 63% for detection of macrometastases and micrometastases.16 Frozen section sensitivity for detection of macrometastases and micrometastases was only 38% in the French prospective study SENTICOL (analysis performed for each lymph node not for each patient).17 A low rate of intra-operative detection of micrometastases and isolated tumor cells was also reported by Sonoda et al in 2018, with 60% sensitivity of frozen section,18 In another prospective trial with 38 patients, the sensitivity of frozen section in combination with imprint cytology reached only 33% for macrometastases, micrometastases, and isolated tumor cells. This trial was, however, affected by a low bilateral sentinel lymph node detection rate (47%).19
In our study, patients with a combination of intermediate-risk factors were not referred for adjuvant treatment without the presence of other high-risk factors. We have recently shown that an excellent outcome, especially local control, can be achieved in this group of patients without adjuvant treatment.20 An intensive protocol for sentinel lymph node ultrastaging included complete processing of all sentinel lymph nodes at four levels in a 150 μm interval. The sensitivity of the frozen section for detection of macrometastases and micrometastases was only 47% and it represented the major reason for a failure to avoid combined treatment. Although primary chemoradiation is a standard of care in patients with T2b stage disease, 14 patients were included in our study, including three patients with missed parametrial involvement on pre-operative staging and 11 highly selected cases with an initial disruption of the peri-cervical ring, diagnosed by expert ultrasound examination. These patients preferred radical surgery over chemoradiation after discussing all management options.
The strength of our study is primarily its inclusion of all consecutive patients referred for radical surgery and the large cohort of patients with sentinel lymph node detection and intra-operative frozen section assessment. Patient management was consistent throughout the study, and the cervical procedure was abandoned in all patients with positive sentinel lymph nodes, with subsequent referral for primary chemoradiation. The main weakness is the retrospective nature of the study.
In conclusion, this retrospective trial showed that 6% of patients were spared from receiving combined treatment owing to implementation of the intra-operative sentinel lymph node triage protocol. In all, 47% of node-positive patients and 38% of all patients with high-risk parameters were referred for primary chemoradiation after abandoning the cervical procedure. Combined treatment was given only to 9% of patients, mostly due to lymph node metastases missed intra-operatively.
Correction notice The Highlgihts section of this article was erroneously altered. This section has now been corrected.
Contributors LDo: data curation; formal analysis; methodology; project administration; writing - original draft; writing - review and editing; investigation. JS, DF, RK, AG, FF, PD, and KN: investigation. LDu and JJ: data curation; formal analysis; methodology. DC: conceptualization; formal analysis; methodology; project administration; supervision; writing - review and editing; investigation.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Patient consent for publication Not required.
Provenance and peer review Not commissioned; externally peer reviewed.
Data availability statement Data are not in public repository and are available upon reasonable request.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.