Article Text

other Versions

Download PDFPDF
Tailored radical hysterectomy for locally advanced cervical cancer
  1. Noriaki Sakuragi1,2,
  2. Masanori Kaneuchi2,
  3. Tatsuya Kato3,
  4. Chisa Shimada4,
  5. Yukiharu Todo4,
  6. Kei Ihira3,
  7. Ayako Nozaki3,
  8. Takeshi Umazume3,
  9. Yosuke Konno1,
  10. Takashi Mitamura3,
  11. Noriko Kobayashi3,
  12. Gen Murakami5 and
  13. Hidemichi Watari1
  1. 1Obstetrics and Gynecology, Hokkaido University Graduate School of Medicine, Sapporo, Japan
  2. 2Gynecology, Otaru General Hospital, Otaru, Japan
  3. 3Obstetrics and Gynecology, Hokkaido University Hospital, Sapporo, Japan
  4. 4Gynecologic Oncology, National Hospital Organisation Hokkaido Cancer Center, Sapporo, Japan
  5. 5Anatomy II, Sapporo Medical University, Sapporo, Japan
  1. Correspondence to Professor Noriaki Sakuragi, Gynecology, Otaru General Hospital, Otaru 047-8550, Japan; nsakuragi{at}jcom.home.ne.jp

Abstract

Objective The survival and prognostic factors for locally advanced cervical cancer treated with nerve-sparing Okabayashi–Kobayashi radical hysterectomy have not been elucidated. We aimed to evaluate the oncological outcomes of those patients after radical hysterectomy with adjuvant chemotherapy.

Methods This retrospective cohort study was conducted from January 2002 to December 2011. Treatment was conducted at a single tertiary center in northern Japan. We used the Okabayashi–Kobayashi radical hysterectomy with lymphadenectomy. We applied unilateral nerve preservation for stage IIA/IIB cancer if there was a one-sided extension of the disease outside the cervix. Indication for adjuvant therapy was based on Sedlis criteria. High-risk was defined as evidence of lymph node metastasis, pathological parametrial invasion, and a positive/close surgical margin. The choice of adjuvant therapy was chemotherapy which consisted of paclitaxel and cisplatin.

Results The study included 76 early-stage IB1 (≤4 cm) and IIA1 cervical cancer and 45 locally advanced stage IB2 (>4 cm), IIA2, and IIB disease treated consecutively. The median follow-up was 106 (range: 6-203) months. There were 18 (15%) patients with recurrence, with five of 76 in the early-stage (7%) and 13 of 45 in the locally advanced disease (29%) (P<0.001). For locally advanced cervical cancer, pT classification (P<0.001), lymph node metastasis (P=0.007), and histology (P=0.05) were associated with locoregional recurrence. The five-year locoregional recurrence rate in the locally advanced disease was 20% and 5% in the early-stage disease (P=0.01). The five-year disease-free survival in the locally advanced cervical cancer was 71% and 93% in the early-stage disease (P<0.001). The overall survival in locally advanced disease depended on the adeno-type histology and lymph node metastasis.

Conclusion The tailored use of nerve-sparing Okabayashi–Kobayashi radical hysterectomy with adjuvant chemotherapy based on tumor histology and lymph node metastasis may be a possible option as a treatment of locally advanced cervical cancer in selected patients.

  • adenocarcinoma
  • cervical cancer
  • cervix uteri
  • surgical procedures, operative

Statistics from Altmetric.com

Request Permissions

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.

Highlights

  • The five-year disease-free survival in the early stage cervical carcinoma was 93% after nerve-sparing radical hysterectomy.

  • Okabayashi–Kobayashi radical hysterectomy showed a satisfactory curative effect even in locally advanced disease in stages IB2–IIB.

  • Radical hysterectomy is less likely to be effective for stage IB2–IIB node-positive adenocarcinoma/adenosquamous carcinoma.

Introduction

Treatment for cervical cancer is planned according to the International Federation of Gynecology and Obstetrics (FIGO) stage. Locally advanced cervical cancer traditionally includes patients with stage IIB–IVA. Many oncologists now include stage IB2 and IIA2 disease in this category. We use the term locally advanced cervical cancer to denote stage IB2/IIA2/IIB in this article. Concurrent chemoradiotherapy is the treatment of choice for locally advanced cervical cancer according to the National Comprehensive Cancer Network (NCCN) Clinical Practice Guidelines.1 In Japan, patients with not only early-stage IB1 and IIA1, but also locally advanced cervical cancer are treated with radical hysterectomy.2 In 2015, the treatment used for 1775 stage II cervical cancer registered to the Japan Society of Obstetrics and Gynecology included upfront surgery (47%), primary radiotherapy (52%), and others (1%).3 The critical aspects of surgical oncology comprise tumor biology, patient selection, and surgical maneuvers.4 One needs to evaluate the applicability of a treatment modality for a specific condition of cervical cancer before selecting the type of treatment. The selection of patients who would benefit from surgical treatment is essential. We found numerous reports on the prognostic significance of adenocarcinoma histology and lymph node metastasis in locally advanced cervical cancer. Nevertheless, the patient selection for radical hysterectomy in locally advanced cervical cancer has been scarcely reported.

We recently reported the results of nerve-sparing Okabayashi–Kobayashi radical hysterectomy with adjuvant chemotherapy for stage IB1–IIB cervical cancer, which demonstrated survival rates which were not inferior to radiotherapy in the published literature.5 In this article, we focus on locally advanced cervical cancer and discuss the patterns of recurrence, prognostic factors, and aspects of patient selection for this surgery.

Methods

The study included invasive cervical cancer treated with radical hysterectomy using a laparotomic approach. The 2008 FIGO staging was used in assigning the stage. The patients underwent Okabayashi–Kobayashi radical hysterectomy from January 1, 2002, to December 31, 2011, at Hokkaido University Hospital, Sapporo, Japan. There was no selection of patients who may have been more likely to respond to surgery and adjuvant chemotherapy.

Patients were followed every 3 months for the first 3 years, every 6 months during years 4–5, and at 12-month intervals after this time. We used chest X-rays and either CT, MRI, or positron emission tomography scans once a year and on any suspicion of recurrence according to bimanual examinations, symptoms, and increases in serum tumor markers that were elevated preoperatively. This study was approved by the local ethics committee (IRB) of Hokkaido University Hospital, and written informed consent was waived because of the retrospective design. We used the opt-out method on the hospital's website for obtaining consent. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline were followed.6

The characteristics of patients with early-stage cervical cancer and locally advanced cervical cancer included age, 2008 FIGO stage, histology (squamous cell carcinoma vs adenocarcinoma/adenosquamous carcinoma), pT classification (pT1 vs pT2), lymph node metastasis (no vs yes), lymphovascular space invasion (LVSI: no vs yes), recurrence risk (low vs intermediate vs high), postoperative adjuvant therapy (no vs yes), and recurrence (no vs yes).

Radical hysterectomy and postoperative adjuvant therapy

Wertheim first described radical hysterectomy in 1912, which was later modified worldwide.7–9 The Meigs (Piver class III) operation is widely used in Western countries. The Okabayashi operation was developed in Japan, and Kobayashi later modified the method to preserve the pelvic nerves.10 11 We used the Okabayashi–Kobayashi method combined with systematic nerve-sparing procedures.12 13 A Online supplementary video file is provided with this article. We applied bilateral nerve preservation for stage IB1/IB2 disease and unilateral nerve preservation for stage IIA/IIB cancer if there was one-side extension of the disease outside the uterine cervix. We did not offer radical hysterectomy to patients with stage IIA/IIB disease with a bilateral extension of the tumor, which would require the extirpation of the bilateral pelvic nerves. We employed a systematic pelvic lymphadenectomy for regional control, and para-aortic lymphadenectomy was used for tumors in stage IB2 or more stages for prognostic purposes based on our study.14

Postoperative adjuvant therapy

We used criteria for adjuvant treatment similar to Sedlis’ definition.1 We defined any of the lymphovascular space invasion, deep cervical stromal invasion (>2/3), or large tumor size (>4 cm) as intermediate risk. High-risk criteria was defined as evidence of lymph node metastasis, pathological parametrial invasion, and a positive/close surgical margin. The choice of adjuvant therapy was chemotherapy consisting of paclitaxel and cisplatin every 3 weeks for four to six courses. The final decision for adjuvant treatment and its modality was made by a discussion between a patient and an attending physician.

Statistical analysis

The primary outcome measures were 5-year disease-free survival, 5-year overall survival, and 5-year disease-specific survival. Overall survival included death from any cause. Disease-specific survival included death from cervical cancer. Patients known to be alive or those lost to follow-up at the time of analysis were censored at their last follow-up. The follow-up period was defined as the time from surgery until recurrence or death and the time from surgery until the last confirmation. The site of recurrence was categorized as local (relapse in the vaginal stump or paravaginal area), regional (relapse in the pelvic sidewall of the pelvic lymph node area), or distant (relapse in areas outside the pelvis including the paraaortic lymph nodes, distant organs, and peritoneal cavity). Disease-free survival was defined as the time from surgery to recurrence. Locoregional recurrence was defined as a relapse of tumor in the local or regional area after the initial treatment. We used Gray’s method to obtain the cumulative incidence curves for locoregional recurrence.15 We used the Fine–Gray proportional subdistribution hazards regression to evaluate the hazard ratio (HR) between the groups.16

The patient and tumor characteristics were compared between the groups using Fisher's exact test. Survival curves were calculated with the Kaplan–Meier method and compared with the log-rank test. We used Cox regression to evaluate the effects of the variables on survival. It is generally assumed that ten outcome events per independent variable are necessary to obtain a Cox regression model. Still, other rules for variables in a prediction model exist. In this study, we employed a rule of five to nine events per variable as discussed by Vittinghoff and McCulloch.17 We performed a univariable Cox regression analysis for overall survival, and then obtained a two-variable regression model based on the number of death events observed. The significance level was set at P<0.05. For multiple comparisons, we used the Bonferroni correction to set the significance P-value level.

These statistical analyses were performed using EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan), which is a graphical user interface for R software, version 3.6.1 (The R Foundation for Statistical Computing, Vienna, Austria).18

Results

The study included a total of 121 patients, with 76 patients diagnosed with early cervical carcinoma and 45 patients diagnosed with locally-advanced cervical carcinoma who were consecutively treated with radical hysterectomy using an open approach. The demographics of the patients are provided in Table 1. There was no difference in age, histologic type, or patterns of recurrence between the early-stage cervical cancer and locally advanced cervical cancer groups. Among 76 patients with intermediate-risk or high-risk recurrence, 68 (89.4%) patients received adjuvant therapy. The percentage of patients who received adjuvant chemotherapy in the early-stage cervical cancer and locally advanced cervical cancer groups was 34% and 82%, respectively. The percentage of patients who received adjuvant radiotherapy in the early-stage cervical cancer and locally advanced cervical cancer groups was 3% and 7%, respectively, with no significant difference (P=0.28). The number of early-stage cervical cancer patients and the number of patients with an intermediate risk or high risk of recurrence were coincidentally the same: 76 patients. Eight patients were at high risk with more than or equal to five positive nodes with pathological parametrial invasion. Three received radiotherapy, and the remaining five patients received chemotherapy.

Table 1

Clinical and demographic features of patients with early-stage and locally advanced cervical cancer treated with radical hysterectomyAdeno(squamous)carcinoma: adenocarcinoma/adenosquamous carcinoma.

The median follow-up was 106 months (range: 6-203 months, interquartile range: 70.5; 125 months). None of the patients were lost to follow-up before 60 months after surgery. Fifteen patients died of cervical cancer and one patient died of intercurrent disease after surgery, and no patient died from any other causes. There were 18 (15%) patients with recurrence, with five of 76 in the early-stage disease (7%) and 13 of 45 in the locally advanced disease (29%) (P<0.001). For patients with locally advanced cervical cancer, pT classification (P<0.001), lymph node metastasis (P=0.007), and histology (P=0.05) were associated with locoregional recurrence, but LVSI was not associated with locoregional recurrence. The five-year locoregional recurrence rate in the locally advanced cervical cancer was 20%, which was significantly higher than 5% in the early-stage cervical cancer (P=0.01, Figure 1). The HR for locoregional recurrence between the locally advanced cervical cancer and early-stage cervical cancer was 4.2 (95% confidence interval [CI]: 1.3 to 13.6, P=0.02). We observed three locally advanced cervical cancer patients (one with pT2a1N1M0 and two with pT2bN1M0) and one early-stage cervical cancer patient (pT2a1N1M0) with local recurrence in the paracolpium on the same side as the nerve-sparing procedure. The five-year disease-free survival for stages IB1–IIB combined was 85%. The five-year disease-free survival in the locally advanced cervical cancer was 71% (95% CI: 56% to 82%), which was significantly lower than 93% in the early-stage cervical cancer (95% CI: 85% to 97%) (P<0.001). Histology, pT classification, and lymph node metastasis were associated with disease-free survival of patients with locally advanced cervical cancer.

Figure 1

Cumulative incidence of locoregional recurrence in locally advanced cervical cancer and early-stage cervical cancer after Okabayashi–Kobayashi radical hysterectomy and adjuvant chemotherapyThe cumulative incidence curve for locoregioal recurrence was obtained using Gray’s method. Comparison of the cumulative incidence curves was achieved by the Fine–Gray proportional subdistribution hazards regression. Recurrence in the distant organs and all-cause death were competing risks for locoregional recurrence. If the locoregional recurrence occurred synchronously with distant metastasis, we designated it as the locoregional recurrence in the analysis.

There were 16 (13%) deaths: four in the early-stage cervical cancer and 12 in the locally advanced cervical cancer. Five-year overall survival and disease-specific survival were worse in the locally advanced cervical cancer than in the early-stage cervical cancer (82% vs. 95%, P=0.001). We obtained a two-variable regression model for overall survival in locally advanced cervical cancer that included histology and lymph node metastasis (Table 2). Histology of adenocarcinoma/adenosquamous carcinoma (adeno [squamous] carcinoma) and lymph node metastasis were independent predictor of decreased survival in locally advanced cervical cancer.

Table 2

Cox regression analysis for overall survival using univariable and two-variable models in patients with locally advanced cervical cancer after Okabayashi–Kobayashi radical hysterectomyUnivariable Cox regression analysis for each variable was performed, and then a two-variable regression model based on the number of observed events was obtained. Adeno(squamous)carcinoma: adenocarcinoma/adenosquamous carcinoma.

The 5-year disease-specific survival in the locally advanced cervical cancer patients with adeno(squamous)carcinoma and positive lymph node metastasis (38%) was less than that in patients with other combinations of histologic type and lymph node metastasis: 100% for patients with squamous cell carcinoma without lymph node metastasis (P<0.001), 85% for patients with squamous cell carcinoma with lymph node metastasis (P=0.002), and 88% for patients with adeno(squamous)carcinoma without lymph node metastasis (P=0.01). We set the P-value for significance at 0.05/3=0.017 according to the Bonferroni correction (Figure 2).

Figure 2

Disease-specific survival in patients with locally advanced cervical cancer treated with Okabayashi–Kobayashi radical hysterectomy according to histologic type and lymph node metastasisSCC: squamous cell carcinoma. Adeno/Adsq: adeno(squamous)carcinoma. LN: lymph node metastasis. DSS curves were obtained using the Kaplan–Meier method and compared with the log-rank test. The HR was calculated using Cox regression. The cut-off value for significance was P<0.017 according to the Bonferroni correction.

Discussion

We obtained favorable locoregional control and survival in patients with locally advanced cervical cancer after radical hysterectomy with adjuvant chemotherapy. It was suggested that the prognosis of some group of patients with locally advanced cervical cancer, as characterized by squamous cell carcinoma with or without lymph node metastasis, and adeno(squamous)carcinoma without lymph node metastasis was excellent after this treatment modality. We observed some patients with a pT2a or pT2b tumor had local recurrence in the paracolpium area on the same side of nerve-sparing procedure, which suggested that pathological vaginal and parametrial invasion may be associated with the nervous system involvement of cancer. Radical hysterectomy and radiotherapy are two radical treatments that are offered with curative intent. Administering dual radical therapy to patients has been argued because of the concern regarding the increased morbidity associated with dual therapy. Recently, comparisons between concurrent chemoradiotherapy and neoadjuvant chemotherapy followed by surgery have been conducted.19 20 We need to evaluate further the results of the two RCTs regarding survival and morbidities. If adjuvant radiotherapy was given to patients after neoadjuvant chemotherapy + surgery, the patients received tri-modal therapy, which raises concerns regarding the treatment-associated morbidity. Concurrent chemoradiotherapy is considered a standard of care in high-risk patients after surgery. We wanted to avoid long-lasting complications caused by dual radical therapy. We reported that adjuvant chemotherapy had the equivalent therapeutic effect as radiotherapy with fewer postoperative complications.21–23 A randomized controlled trial was launched by the Japan Gynecologic Oncology Group (JGOG) in May 2019, which compares radiotherapy and chemotherapy as an adjuvant treatment for high-risk patients with stage IB–IIB treated with radical hysterectomy (JGOG 1082, AFTER Trial).

We could not find a randomized controlled trial for locally advanced cervical cancer comparing upfront radical hysterectomy with concurrent chemoradiotherapy. One population-based study obtained from the Surveillance, Epidemiology, and End Results (SEER) database that included early-stage (IB1 and IIA1) disease with squamous cell histology, suggested that cancer-specific and overall survivals were similar with the two modalities.24 In Japan, radical hysterectomy was used in 46.6% of patients with stage II (15%: IIA1, 85%: IIA2 and IIB) cervical cancer in 2015.3 The treatment results for locally advanced cervical cancer are provided from the SEER database,25 and the FIGO Annual Report.26 The 5-year overall survival rate for stages IB2, IIA, and IIB in the SEER database was 74%, 62%, and 64%, respectively. The 5-year overall survival rate for stage IB2, IIA, and IIB in the FIGO Annual Report was 76%, 73%, and 66%, respectively. According to the Annual Report of the Committee on Gynecologic Oncology from the Japan Society of Obstetrics and Gynecology, the values were 76%, 81%, and 72%, respectively.27 The results for stages IIA and IIB in Japan seem not inferior to the findings from the SEER database and the FIGO report. It has been suggested that grade 3–4 late adverse events in patients with stage IIB disease are more frequent with radical hysterectomy plus adjuvant radiotherapy than with concurrent chemoradiotherapy.28 However, the type of adjuvant therapy is related to treatment-associated morbidity. A nationwide cohort study in Japan revealed that compared with adjuvant chemotherapy, adjuvant radiotherapy increased grade 3–4 adverse events after radical hysterectomy.29

The staging of patients in this study was made by clinical bimanual examination and with the aid of MRI for assessing tumor diameter. In our locally advanced cervical cancer cohort, 32 patients had stage IIA2 (n=5) or IIB disease (n=27), and 17 (53%) had tumor pathologically proven bulky vaginal invasion (n=6) or parametrial invasion (n=11). Our results seem to correspond to the previous reports that stage IIA disease had pathological parametrial invasion at risk higher than or equivalent to stage IIB,30 and the accordance of the preoperative diagnosis of parametrial invasion with pathologically confirmed parametrial invasion is as low as 21% to 55%.31 Some authors have discussed the role of surgery for locally advanced cervical cancer from the difficulty in the preoperative diagnosis of stage IIB.32

We need to take into account tumor biology, in addition to the FIGO stage. Numerous reports on cervical cancer revealed that adenocarcinoma histology33 34 and lymph node metastasis33 35 are associated with patients’ poor survival. We found locally advanced cervical cancer patients with squamous cell carcinoma irrespective of lymph node metastasis and node-negative adeno(squamous)carcinoma had excellent survival without using adjuvant radiotherapy. Sufficient removal of paracervical tissues and the vagina/paracolpium is required in the treatment of locally advanced cervical cancer. The aggressive biology of the tumor, represented by adeno-type histology and lymph node metastasis, is the most critical factor that determines survival in patients with locally advanced cervical cancer treated with radical hysterectomy. The blood vessel invasion was more frequent in adenocarcinoma than in squamous cell carcinoma, and it increased as the lymphatic invasion became more prominent.36 A node-positive cervical adenocarcinoma may have a propensity of hematogenous spread resulting in systemic disease. Cady’s notion about the importance of tumor biology and patient selection seems to have critical significance here,3 while the method of radical hysterectomy also has an essential role as well.

This retrospective study was performed in a single tertiary care center, and it included a small number of patients without comparison with primary chemoradiotherapy, which limits the generalizability of the study. Another limitation is omitting patients with bilateral vaginal/parametrial invasion, although previous publication showed that the presence of bilateral parametrial invasion did not alter the disease-free survival in stage IIB patients.37 Nevertheless, the surgery performed was well-standardized, systematic, nerve-sparing Okabayashi–Kobayashi radical hysterectomy. Additionally, an intensive effort was made to ensure that the quality of follow-up was high, with no patients lost to follow-up before 60 months after surgery.

In addition to chemoradiotherapy, the tailored use of Okabayashi–Kobayashi radical hysterectomy based on tumor histology of adeno(squamous)carcinoma and lymph node metastasis may be an option in Japan as a treatment of choice for locally advanced cervical cancer.

References

Footnotes

  • Collaborators Gen Murakami, MD, PhD., Department of Anatomy II, Sapporo Medical University.

  • Contributors Study concept: NS, MK, YT, HW. Study design: NS, MK. Data acquisition: TK, CS, AN, KI, YK, TM. Quality control of data and algorithms: NS, HW. Data analysis and interpretation: NS, MK. Statistical analysis: NS, MK, TU. Manuscript preparation: NS. Manuscript editing: NS. Manuscript review: MK, YT, NK, HW.

  • 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 available upon reasonable request. Deidentified participant data may be available upon request for research purposes and upon approval by the IRB of Hokkaido University.