New approaches in the surgical management of early stage cervical cancer Marie Plante and Michel Roy

Transcription

1 New approaches in the surgical management of early stage cervical cancer Marie Plante and Michel Roy Quality of life has become a very important issue in deciding the extent of surgical procedures for patients affected with a variety of cancers. In recent years, more attention has been given to preserving organ function, cosmesis, and now reproductive function. As cancer treatment has improved survival in several neoplasias, cancer survivors are becoming more interested in preserving their fertility potential.we focus here on new surgical advances in the management of early stage cervical cancer, with emphasis on the radical trachelectomy technique to preserve fertility potential and the Saling procedure to prevent premature labor. We also discuss the issue of the sentinel node localization. Curr Opin Obstet Gynecol 13:41±46. # 2001 Lippincott Williams & Wilkins. Gynecologic Oncology Service, L`Hoà tel-dieu de Que bec, Centre Hospitalier Universitaire de Que bec (CHUQ), Laval University, Quebec City, Canada Correspondence to Dr Marie Plante, L`Hoà tel-dieu de Que bec, Gynecology Service, 11 Coà te du Palais, Quebec City, Canada G1R-2J6 Tel: ; fax: ; marie.plante@crhdq.ulaval.ca Current Opinion in Obstetrics and Gynecology 2001, 13:41±46 Abbreviations VSI vascular space invasion SGO Society of Gynecologic Oncologists GOG Gynecologic Oncology Group # 2001 Lippincott Williams & Wilkins X Introduction Cure is undoubtedly the ultimate aim of cancer treatment. However, cancer treatments can lead to a variety of long-term effects that can have dramatic permanent impacts on quality of life: impaired sexual function and/or reproduction, cosmetic appearance, loss or signi cantly reduced organ function, etc. In recent years, we have witnessed `radical' changes in the surgical management of some cancers in order to reduce and/or minimize long-term consequences of cancer treatments. In gynecologic oncology, the most striking advances have probably been in treatment of vulvar cancer, where nowadays we hardly ever see the dis guring en-bloc radical vulvectomy with groin node dissection. In earlystage epithelial and germ-cell ovarian cancer, the extent of surgery has also decreased in order to preserve fertility while maintaining excellent outcome with adjuvant chemotherapy when indicated. In the area of cervical cancer there have not been any major changes in the management of early-stage cervical cancer since the introduction of the Wertheim radical hysterectomy technique developed by Meig about 60 years ago [1]. The introduction of invasive laparoscopic procedures in gynecologic oncology in the early 1990s has revived the potential for the vaginal radical hysterectomy or Schauta±Amreich operation [2]. Soon after this, Professor D. Dargent developed the radical trachelectomy procedure, a modi ed version of the Schauta operation [3]. The procedure implies removing only the cervix and parametrium to preserve the body of the uterus and thus the reproductive function. Because of the rate of premature labor and delivery in some patients who became pregnant after a radical trachelectomy, Dargent recently began using the Saling procedure initially developed by Saling for patients with habitual abortions [4]. Also, the area of node mapping is gaining more acceptance in the management of a number of malignancies such as melanoma, breast, and vulvar cancer. Dargent has begun using this technique to identify the sentinel node in early stage cervical cancer to reduce the extent of the lymph node dissection in this low-risk group [5]. These three recent advances in the surgical management of early stage cervical cancer are discussed below. 41

2 42 Gynecologic oncology and pathology Radical trachelectomy It is estimated that 10±15% of cervical cancers are diagnosed in childbearing years [6]. In western countries, women often delay childbearing until the mid to late thirties. So, mathematically, a number of women will be diagnosed with cervical cancer prior to having completed their family. Although women can be offered alternative reproductive options such as in-vitro fertilization with embryo cryopreservation and gestational surrogacy, those approaches remain fairly complex and are not ethically acceptable or affordable to many [7]. Indeed, Duska et al. raise several important technical problems and ethical issues with regard to the use of these technologies speci cally in the context of women with cervical cancer [7]. Therefore, the radical trachelectomy procedure, which allows preservation of the body of the uterus and thus preserves reproductive function, emerges as a true breakthrough in the management of young women with early-stage cervical cancer. The surgical procedure was rst described by Dargent in the French literature in 1994 [3]. Initially received with skepticism when presented before the Society of Gynecologic Oncologists (SGO) in 1994 [8], the concept of preserving the body of the uterus is now gradually gaining recognition and acceptance. Brie y, the procedure begins with a laparoscopic pelvic lymphadenectomy. Nodes are sent for frozen section evaluation and, if negative, the trachelectomy procedure is then performed. The specimen is also sent for frozen section analysis of the endocervical margin. A clear margin of 5± 8 mm is considered adequate, otherwise more endocervix is removed, or a complete vaginal radical hysterectomy is performed. Details of the surgical technique have been described in length elsewhere [9,10]. The results are, interestingly, quite comparable. Table 1 summarizes patients' and tumor characteristics of the four most recent published series for a total of 130 cases [11..,12.,13..,14]. The majority of lesions are squamous, stage IB1 or less, without vascular space invasion (VSI) and measure less than 2 cm. Only Dargent has done the procedure on stage IIB, however those were `pathological' stage IIB, that is, invasion of the parametrium discovered on nal pathology. As expected, his data indicate that these patients have a much higher risk of extrauterine spread [11.. ]. Table 2 summarizes intraand postoperative data. Overall, blood loss, transfusion rate, and hospital stay are shorter compared with the standard radical hysterectomy, although operating time is longer (except for Dargent). Intraoperative complication rate can be divided into those attributed to the laparoscopic lymphadenectomy and those related to the trachelectomy itself. In the former group, complications were usually related to a few vascular traumas repaired at the time of surgery, although Dargent reported a 10% rate of reoperation [11.. ]. Complications related to the trachelectomy were most commonly iatrogenic cystotomies repaired without consequences. Of note, there have been no ureteral injury nor trocar site recurrences reported so far. Table 3 summarizes oncologic outcome. The overall recurrence rate of the four published series is 3.1% (4/ 130), which is no different from the rate reported after radical hysterectomy. Indeed, Covens compared his series of trachelectomies with two groups of matched patients treated with standard radical hysterectomy and showed that the recurrence-free survival is not statistically signi cantly different among the three groups. The 2-year actuarial survival for his trachelectomy patients Table 1. Patients and tumor characteristics Age: median (range) N/A 32 (22±42) 31 (25±46) 32 (24±43) Stage IA1 5 (11%) 2 (5%) 10 (32%) 0 IA2 13 (28%) 12 (29%) 11 (34%) 0 IB1 25 (53%) 25 (61%) 11 (34%) 10 (100%) IB IIA 1 (2%) 2 (5%) 0 0 IIB* 3 (6%) Histology Squamous 39 (83%) 27 (65%) 19 (59%) 8 (80%) Adenocarcinoma 8 (17%) 14 (35%) 13 (41%) 2 (20%) VSI No 34 (72%) 37 (90%) 18 (56%) N/A Yes 13 (28%) 4 (10%) 14 (44%) N/A Size 52 cm 40 (85%) 39 (95%) 31 (97%) N/A 42 cm 7 (15%) 2 (5%) 1 (3%) N/A *Pathologic stage IIB.

3 Surgery of early stage cervical cancer Plante and Roy 43 Table 2. Intra- and postoperative data Operating room time (h) (range) 2.1 (N/A) 4 (2.5±7.5) 3 (2.5±5.0) N/A Blood loss (ml) (range) N/A 200 (50±1200) 400 (200±1400) N/A Blood transfusion (%) 6% 6% 4% N/A Intraoperative complications a 2% 11% 25% N/A Postoperative complications a 17% 17% 6% N/A Days hospital stay (range) 7 (N/A) 3 (2±9) 1 (1±7) N/A a May be inaccurate as some authors only reported major complications. Table 3. Oncologic outcome Follow-up (months) (range) 52 (7±123) 32 (1±99) 23 (1±61) 2 (1±35) Pelvic node count N/A 32 (11±107) N/A N/A Positive nodes 2 (4.2%) 1 (2.4%) 0 3 (30%) Recurrences 2 (4.2%) 1 (2.4%)* 1 (3.1%) 0 *There was also one patient with a 1 cm neuroendocrine small cell tumor who recurred and died despite aggressive adjuvant chemotherapy (negative margins and negative nodes). was 95% [13.. ]. Three of the four reported recurrences have been parametrial/sidewall recurrences, suggesting that these may be `trachelectomy failures'. Although patients selected for trachelectomy mostly have small early stage cancers, there is a 2±5% rate of positive nodes, which emphasizes the need for a pelvic lymphadenectomy in those cases. Interestingly, Dargent reported cases of paracervical spread and parametrial node metastasis that were not contiguous to the primary tumor itself [11.. ]. This emphasizes the fact that patients should truly have a `radical' trachelectomy with adequate parametrial length and not just a `cervicectomy'. A recent article from an Italian group using the giant-section technique indicates that subclinical parametrial spread occurs in 30±60% of patients with stage IB±IIA cancers [15.. ]. They further noted that all the patients with positive pelvic nodes also had parametrial involvement. This study also supports the importance of the parametrectomy along with the trachelectomy [15.. ]. The size of the lesion seems to be an important predictor of recurrence. In Dargent's series, 28% of cases with extrauterine spread on nal pathology were in patients with lesions measuring greater than 2 cm, whereas this occurred in 13% of stage IB lesions measuring less than 2 cm and was absent in IA lesions [11.. ]. Moreover, one of our patients with a recurrence and the two of Dargent's were inpatients with lesions measuring more than 2 cm who also had VSI. It is now our policy to limit the trachelectomy procedure to patients with lesions measuring less than 2 cm or with stage Ia1 with VSI, IA2, and IB1 cancers. In terms of obstetric outcome, the data are also quite encouraging (Table 4). Altogether, 49 pregnancies in 36 women have been reported, resulting in 26 live births. In their series, Covens et al. reported that all women became pregnant within 12 months of attempting to conceive, giving a conception rate of 37% at 1 year [13.. ]. Most women became pregnant `naturally', without assisted reproductive help, even though there were initial concerns about potential infertility problems related to the shortened cervix (i.e. inadequate or hostile mucus production). Interestingly, of the women who attempted to conceive in Coven's series, 5 had a prior history of infertility and yet 3 became pregnant spontaneously [13.. ]. This would emphasize the fact that a `history' of infertility should not necessarily exclude patients from this procedure. However, in Dargent's series, 8 women have been unable to conceive, including 5 women with prior infertility problems, for which 3 had even tried in-vitro fertilization unsuccessfully [11.. ]. Ethically, it is thus dif cult to decide where to draw the line. Most women had term cesarean section, although in our series, there were 3 cases of premature labor and delivery at 25, 28, and 34 weeks gestation [16]. Luckily children did not suffer long-term complications. Our patient who delivered at 25 weeks became pregnant again and delivered at 20 weeks, despite placement of another cervical cerclage at 14 weeks. She recently became pregnant again. She had another cerclage placed at 14 weeks of gestation and the Saling procedure was also performed. She did well until 28 weeks when she began contracting and required a cesarean section. The series

4 44 Gynecologic oncology and pathology Table 4. Obstetric outcome Total No. of pregnancies 25 (18 women) 17 (12 women) 5 (4 women) 7 (4 women) Live births Neonatal death Currently pregnant N/A 3 N/A N/A 1st trimester miscarriage Voluntary abortion nd trimester miscarriage Unable to conceive a Including 5 ongoing pregnancies at the time of the trachelectomy. b Including 1 ongoing pregnancy at the time of the trachelectomy. of events in this case raises the possibility of an underlying idiopathic premature labor problem rather than a `cervical' problem. We also had a patient who delivered prematurely at 31 weeks after premature rupture of membranes. The baby was born with excellent Apgar score but died of E. coli sepsis within 36 hours. Dargent also reported 8 deliveries prior to 37 weeks (cf. Saling procedure section below). Can we do less than a radical trachelectomy? For instance, can we do just a simple conization with or without lymphadenectomy in very early lesions? A recent case report of a recurrent adenosquamous carcinoma widely metastatic to the body of the uterus and ovary after a polypectomy followed by a conization (with negative margins) and a lymphadenectomy (with negative nodes) cautions us against the danger of being too conservative [17]. As mentioned by those authors, we may see unusual patterns of recurrence following conservative treatment approaches in the future [17]. We had a similar case of a patient with a squamous cell cancer con ned to a polyp which had been excised. There was no residual disease on the trachelectomy specimen, yet we found 5 positive microscopic lymph nodes on nal pathology (frozen section of the nodes had been reported as negative). Data from the Italian group mentioned above would also strongly argue against a conization alone [15.. ]. In addition, our group and Dargent's group each had four cases of young women with seemingly early cancers based on clinical evaluation that had spread to the endometrium and myometrium, discovered at the time of surgery. In our cases, the trachelectomies were immediately followed by a radical vaginal hysterectomy and adjuvant chemo-radiation therapy [12. ]. To avoid those problems it may be useful to determine the degree of endocervical extension of the lesion using magnetic resonance imaging [18.. ]. So, overall, the oncologic and obstetric results are favorable and the morbidity of the procedure appears to be low. However, numbers are still small. An international collaboration of all the groups performing this procedure to create a registry to more rapidly accumulate data on this procedure is in preparation. Proper training to perform this surgery adequately is another important issue. Unfortunately animal models are not very suitable to teach vaginal surgery. Although unpleasant, using cadavers may be the best way to teach this procedure. Saling procedure Second trimester abortion and prematurity can be a major problem after a radical trachelectomy. The inevitable shortening of the cervix after this procedure seems to prevent the formation of an ef cacious mucus plug. The mucus plug is thought to be a physiological barrier between the vaginal ora and the membranes to prevent ascending infections. In order to avoid chorioamnionitis, which is most likely responsible for premature rupture of membranes and premature labor following the radical trachelectomy procedure, Dargent has proposed a complete cervical closure of the cervix during pregnancy, a technique described by Saling in 1981 for patients with habitual abortions [4]. The Saling technique of cervical closure is simple. The procedure is ideally performed at 12±14 weeks of pregnancy under general or regional anesthesia. The vaginal tissue just around the external os is super cially injected with a saline solution in order to separate the mucosa from the underlying mucosal layers. An area of cervico-vaginal mucosa 1.5 cm wide is then removed 360 degrees around the external os. The defect is closed with a mono lament resorbable suture in two layers: the deep layer includes the cervical stroma, taking care not to go too deep in order to avoid rupturing the membranes, and the second layer includes the vaginal mucosa. Restoration of the permeability of the cervix is accomplished at the time of the planned cesarean section (at approximately 38 weeks), by digital perforation of this reversible vaginal closure. Only 1 of the 6 patients operated by Dargent with this technique had premature labor, compared to 6 out of 21 before the routine use of the Saling procedure (personal communication, Dargent

5 Surgery of early stage cervical cancer Plante and Roy 45 D). Thus, this technique warrants further evaluation as it appears promising to prevent premature labor and delivery in patients with shortened cervices following radical trachelectomy. However, larger series will be necessary to prove the necessity of the procedure and determine its indications. Sentinel node Data on the sentinel node localization have been very useful and promising in some malignancies. For instance, in melanoma, the area of node mapping is gaining more acceptance in the eld, as it substantially decreases the morbidity of the complete node dissection [19. ]. The technique has also allowed the recognition of `aberrant' lymph node drainage in some cases which may explain some unusual patterns of recurrences. Another good example is in breast cancer, where identi cation of the sentinel node is performed increasingly often to limit the extent of the axillary node dissection [20]. In gynecologic oncology, the issue of node mapping in vulvar cancer is also gaining acceptance as a means to further decrease the permanent side-effects of lymphedema of the lower limbs [21. ]. The GOG is currently conducting a study looking at the intraoperative lymphatic mapping in patients with stage I and II squamous carcinoma of the vulva (GOG-173). So far, very little has been done in cervical cancer in the area of node mapping. In patients with very early smallvolume cervical cancer the yield of lymph node metastasis is low (probably less than 5%). It is thus a concern to remove 30±50 normal and benign lymph nodes in the vast majority of cases. Recently, Dargent has presented very interesting data at the 2000 Society of Gynecologic Oncologists meeting in San Diego on the laparoscopic identi cation of the sentinel node using intracervical injection of Patent blue dye [5]. In his preliminary series of 23 patients, the sentinel node was identi ed 76.5% of the time. As the intracervical injection technique of the dye improved, identi cation of the sentinel node approached 90%. The positive and negative predictive value of the sentinel node was 100%, i.e. when the sentinel node was negative the rest of the complete lymphadenectomy did not reveal any metastatic nodes. Conversely, in all the cases where positive nodes were identi ed the sentinel node had also been positive. There were thus no false negative cases, which is encouraging [5]. Currently, the technology does not yet allow the use of lymphoscintigraphy localization, performed in nuclear medicine, and combined with the intraoperative g-probe detection because the probes do not t the size of a laparoscopy port. If this technical problem can be circumvented, it may be potentially useful to combine both approaches to increase the yield of identifying the sentinel nodes, particularly when it is located in an unusual place. In the future, the next step towards the conservative management of very early cervical cancer may be the combined use of the laparoscopic sentinel node localization technique followed by a radical trachelectomy. The obvious advantage of such a technique would be to limit the extent of the lymphadenectomy in this low risk subgroup and further reduce the related side effects of the procedure. This approach however is experimental and remains unproved at the present time but deserves further evaluation. Conclusion The Halstead paradigm of en-bloc removal of the primary tumor along with its lymphatic drainage is now being questioned [22.. ]. Indeed, surgery for several cancers has become less aggressive in order to preserve quality of life for patients (i.e. function, cosmesis, and reproduction). As we may be seeing an increase in the incidence of cervical cancer in young women as a result of detected cases from screening [14], the radical trachelectomy procedure is without question one of the most exciting new developments in the eld of conservative surgery in gynecologic oncology in the last decade. These women are likely to present with early small-volume disease and are thus highly curable. In that context, radical trachelectomy is becoming an important treatment option to offer to those young women who wish to preserve their fertility potential. Long-established concepts in the management of early stage cervical cancer are now being challenged. Selection of patients for conservative treatment is critical as well as proper training in laparoscopic and vaginal surgery. Data are accumulating around the world to suggest that in well selected cases the procedure is safe and the morbidity is low. The oncologic outcome is reassuring with a low recurrence rate (55%). The obstetric outcome is also exciting as most women have been able to conceive normally after the procedure and most delivered near term by elective cesarean section. Closure of the cervical os at 14 weeks of pregnancy (Saling procedure) may further decrease the rate of second trimester miscarriage, most likely initiated by subclinical chorioamnionitis from an inadequate mucus plug. In the future we may also be in a position to be even more conservative and limit the extent of the lymphadenectomy in patients with very early lesions by using the laparoscopic intraoperative lymph node mapping technique to identify the sentinel node.

6 46 Gynecologic oncology and pathology References and recommended reading Papers of particular interest, published within the annual period of review, have been highlighted as:. of special interest.. of outstanding interest 1 Meigs J. Carcinoma of the cervix: The Wertheim operation. Surg Gynecol Obstet 1944; 78:195± Dargent D. A new future for the Schauta's operations through presurgical retroperitoneal pelviscopy. Eur J Gynecol Oncol 1987; 8:292± Dargent D, Brun J-L, Roy M, et al. La tracheâ lectomie eâ largie (T.E.). Une alternative aá l`hysteâ rectomie radicale dans le traitement des cancers infiltrants deâ veloppeâ s sur la face externe du col uteâ rin. J Obstet Gynecol 1994; 2:285± Saling E. Early total occlusion of os uteri to prevent habitual abortion and premature deliveries [in German]. Z Geburtshilfe Perinatol 1981; 185:259± Dargent D, Martin X, Roy M, et al. Identification of a sentinel node with laparoscopy in cervical cancer. Gynecol Oncol 2000; 76: Van der Vange N, Weverling G, Ketting B, et al. The prognosis of cervical cancer associated with pregnancy: a matched cohort study. Obstet Gynecol 1995; 85:1022± Duska L, Toth T, Goodman A. Fertility options for patients with stages IA2 and IB cervical cancer: presentation of two cases and discussion of technical and ethical issues. Obstet Gynecol 1998; 92:656± Dargent D, Brun JL, Roy M, et al. Pregnancies following radical trachelectomy for invasive cervical cancer. Gynecol Oncol 1994; 52: Plante M, Roy M. Radical trachelectomy. In: Operative technique in gynecologic surgery. Gershenson D, Fowler M (editors). Philadelphia: W.B. Saunders; Vol 2(3); pp. 187± Roy M, Plante M. Radical vaginal trachelectomy. In: Laparoscopic surgery in gynaecologic oncology. Querleu D, Childers JM, Dargent D (editors). Oxford: Blackwell Science; pp. 78± Dargent D, Martin X, Sacchetoni A, et al. Laparoscopic vaginal radical.. trachelectomy: a treatment to preserve the fertility of cervical carcinoma patients. Cancer 2000; 88:1877±1882. Largest experience with radical trachelectomies with emphasis on oncologic and obstetric outcome. 12 Plante M, Roy M. Pregnancies following radical trachelectomy for the treatment. of early-stage cervical cancer. In: 7th Biennial Meeting of the International Gynecologic Cancer Society.: Pecorelli S, Atlante A, Benedetti Panici P, et al. (editors). Bologna: Monduzzi Editore; pp. 449±453. This presentation reviews the authors' experience with this procedure. 13 Covens A,Shaw P, Murphy J,et al. Is radical trachelectomy a safe alternative to.. a radical hysterectomy for patients with stage IA-B carcinoma of the cervix? Cancer 1999; 86:2273±2279. This paper compares surgical data and survival of patients treated with a radical trachelectomy and standard radical hysterectomy. 14 Shepherd J, Crawford R, Oram D. Radical trachelectomy: a way to preserve fertility in the treatment of early cervical cancer. Br J Obstet Gynaecol 1998; 105:912± Benedetti-Panici P, Maneschi F, D'Andrea G, et al. Early cervical carcinoma... The natural history of lymph node involvement redefined on the basis of thorough parametrectomy and giant section study. Cancer 2000; 88:2267±2274. This article demonstrates the high rate of subclinical parametrial spread in patients with stage IB-IIA cervical cancers using the giant section technique. 16 Roy M, Plante M. Pregnancies following vaginal radical trachelectomy for early-stage cervical cancer. Am J Obstet Gynecol 1998; 179:1491± Zanetta G, Gabriele A, Vecchione F, et al. Unusual recurrence of cervical adenosquamous carcinoma after conservative surgery. Gynecol Oncol 2000; 76:409± Peppercorn P, Jeyarajah A, Woolas R, et al. Role of MR imaging in the selection.. of patients with early cervical carcinoma for fertility-preserving surgery: initial experience. Radiology 2000; 212:395±399. This review shows that MR imaging is very accurate in detecting myometrial invasion from cervical cancer and in determining the extent of the cancer in relation to the internal os, which may be useful in selecting patients for radical trachelectomy. 19 Gershenwald J, Thompson W, Mansfield P, et al. Multi-institutional melanoma. lymphatic mapping experience: the prognostic value of sentinel lymph node status in 612 stage I or II melanoma patients. J Clin Oncol 1999; 17:976±983. This paper shows that sentinel node biopsy is highly accurate to identify regional lymph node metastasis and this information can be useful to identify patients who need adjuvant treatment. 20 Alazraki N, Styblo T, Grant S, et al. Sentinel node staging of early breast cancer using lymphoscintigraphy and the intraoperative gamma-detecting probe. Semin Nucl Med 2000; 30:56± Terada K, Shimizu D, Wong J. Sentinel node dissection and ultrastaging in. squamous cell cancer of the vulva. Gynecol Oncol 2000; 76:40±44. This study shows the excellent sensitivity of the sentinel node dissection to identify micrometastasis. 22 Fisher B. From Halstead to prevention and beyond: advances in the.. management of breast cancer during the twentieth century. Eur J Cancer 1999; 35:1963±1973. This review describes the evolution in the management of breast cancer over the last 30 years.