Annals of Surgical Oncology 14(3):1129–1135

DOI: 10.1245/s10434-006-9252-0

Lymph Node Involvement in Locally Advanced Cervical Cancer Patients Administered Preoperative Chemoradiation versus Chemotherapy Gabriella Ferrandina, MD,1,3 Mariagrazia Distefano, MD,1 Manuela Ludovisi, MD,1 Alessio Morganti, MD,3 Daniela Smaniotto, MD,2 Giuseppe DÕAgostino, MD,2 Francesco Fanfani, MD,4 and Giovanni Scambia, MD4

1

Gynecologic Oncology Unit, Catholic University of Rome, L.go A. Gemelli 8, 00168 Rome, Italy 2 Division of Radiotherapy, Catholic University of Rome, Rome, Italy 3 Division of Radiotherapy, Catholic University of Campobasso, Campobasso, Italy 4 Department of Oncology, Catholic University of Campobasso, Campobasso, Italy

Background: A retrospective study was planned in 127 locally advanced cervical cancer (LACC) to investigate: (1) the rate and pattern of metastatic lymphnode involvement in patients administered preoperative chemoradiation (CT/RT) versus neoadjuvant chemotherapy (NACT), and (2) the profile of clinico-pathological parameters predictive of metastatic lymph node involvement in these two clinical settings. Finally, we investigated whether the pathologically assessed status of lower pelvic nodes (LPN) was able to predict the pathologically assessed status of upper pelvic nodes (UPN) and parametrium in cases administered CT/RT. Methods: Patients were selected including LACC patients who were administered concomitant CT/RT (n = 87) or NACT (n = 40), before radical surgery. Results: Metastatic pelvic lymphnode involvement was significantly lower in cases administered CT/RT (11.5%) compared to cases administered NACT (30.0%) (P value = 0.009). In the CT/RT group, only MRI-assessed pelvic node status (both at staging and post-treatment evaluation) was associated with pathologic pelvic node status. In patients administered CT/ RT, the status of LPN appeared associated with the status of UPN. Conclusions: (1) Preoperative CT/RT treatment is associated with a lower rate of pelvic node disease in LACC patients compared to NACT; (2) there is no association between the preoperative extent of residual cervical disease after CT/RT and pathologically assessed pelvic node status; (3) the pathological status of LPN is predictive of the pathological status of UPN and parametrium. Key Words: Locally advanced cervical cancer—Preoperative chemoradiation—Lymphnode involvement.

Lymph node status represents a crucial issue in the prognostic characterization and management of cervical carcinoma.1,2 In early cervical cancer patients undergoing radical surgery the rate of metastatic

pelvic node involvement ranges from 13 to 28%, while it rose up to 50% in advanced cervical cancer1,3,4 depending on the severity of stage. As far as paraaortic node metastasis are concerned, in a GOG study on 621 cervical cancer patients undergoing surgical staging, the rate of metastatic involvement is 5% in stage I, 16% in stage II, and 25% in stage III disease.5 Exclusive radiotherapy has represented for years the mainstay of treatment, and currently concurrent chemoradiation has been widely recognized as the

Received June 17, 2006; accepted September 21, 2006; published online January 6, 2007. Address correspondence and reprint requests to: Gabriella Ferrandina, MD; E-mail: [email protected] Published by Springer Science+Business Media, Inc.
2007 The Society of Surgical Oncology, Inc.

1129

1130

G. FERRANDINA ET AL.

golden standard for the management of bulky stage IB and locally advanced cervical cancer (LACC).6 Multimodal investigational treatments including radical surgery after neoadjuvant chemotherapy (NACT) or chemoradiation (CT/RT) have also been explored7–11although no definitive conclusion on survival improvement has been reported: while performing surgery after neoadjuvant therapy allows to remove potential chemoresistant foci, and to assess pathological response, on the other hand the impact of multimodal approaches on rate and pattern of toxicity has to be taken into account. These multimodal approaches also allow to conduct the pathological evaluation of the effect of neoadjuvant treatments on lymph node status: indeed, based on the observations that the incidence of node metastasis detected after NACT is lower than expected7,12 it has been suggested that preoperative chemotherapy could be even more active on lymph node disease than on primary tumor. Conversely, very scanty data has been provided about the pattern of metastatic involvement in LACC patients undergoing preoperative chemoradiation. Houvenaeghel et al.13 recently reported that metastatic pelvic node involvement persists in approximately 16% of LACC cases undergoing preoperative chemoradiation. Indeed, it is conceivable that the control of pelvic disease achieved with concomitant chemoradiation may lead to difference in the rate and pattern of lymphnode involvement compared to NACT. Moreover, it cannot be excluded that also the profile of clinico-pathological variables predictive of lymphnode involvement could vary according to type of preoperative treatment. Finally, anatomo-pathological studies on systematic pelvic and para-aortic lymphadenectomy in patients with cervical cancer, showed that external iliac, interiliac, and obturator lymphnodes are the most frequently involved lymphnodes, and can be considered the primary nodes draining the cervix.12 We and others recently reported that the pathologically assessed status of external iliac, interiliac, and obturator nodes, called Primary Node Group3,14 or lower pelvic nodes (LPN)15,16 is strictly associated with the pathological status of parametria and also of lymphnode stations, such as internal iliac, common iliac, and presacral nodes, that we called upper pelvic nodes (UPN).15,16 This finding has been reported in squamous as well as not squamous cervical carcinoma and in LACC patients submitted to NACT.14–16 thus suggesting that intraoperative assessment of LPN status might be helpful in tailoring the extent of radicality of lymphadenectomy and parametrectomy. Currently, no data are available Ann. Surg. Oncol. Vol. 14, No. 3, 2007

about the ability of LPN to predict parametrial and/ or UPN status in LACC patients administered preoperative chemoradiation. The aim of this study was to investigate (1) the rate and pattern of metastatic lymphnode involvement in patients administered preoperative chemoradiation versus chemotherapy, and (2) the profile of clinicopathological parameters predictive of metastatic lymph node involvement in these two clinical settings. Moreover, we chose to study whether the ability of pathologically assessed LPN status to predict parametrial and UPN status differs according to type of preoperative treatment. To this purpose, a retrospective study was performed in a single institutional series of LACC patients undergoing radical surgery after concomitant CT/RT versus NACT.

PATIENTS AND METHODS The medical records were reviewed for all patients with cervical carcinoma who underwent surgery at the Gynecologic Oncology Unit, Catholic University of Rome, and Campobasso between April 1995 and April 2005. The study was approved by our Ethical Committee. One hundred twenty seven patients were selected on the basis of the following criteria: squamous cell (FIGO stage IIB–IIIB) cervical cancer with no evidence of para-aortic node involvement as assessed by MRI, who were administered concomitant CT/ RT (n = 87) or NACT (n = 40), before radical surgery. Staging was performed according to FIGO classification. Pretreatment evaluation consisted of a history and physical examination, biopsy and gynecologic examination under general anesthesia, abdominal pelvic MRI, pelvic ultrasonography, and chest X-ray. Cystoscopy and sigmoidoscopy were performed when indicated. The medical records were reviewed to obtain clinical, pathological as well as follow up data. Matching between patients administered preoperative CT/RT versus NACT was performed patient by patient (individual matching), i.e. each patient who had received NACT (cases) was individually matched (linked) with at least two controls (CT/RT) making sure that controls had at least all but one matching factor (age, stage, tumor diameter, grading, MRIassessed pelvic lymphnode status at staging) in values equal to the values in cases. This methodology17 led to reach the agreement between control–case pairs with respect to each factor, and to obtain a controlto-case subject ratio of 2.

1131

LYMPHNODES AFTER PREOPERATIVE CHEMORADIATION

TABLE 1. Distribution of clinico-pathological characteristics of locally advanced cervical cancer (LACC) patients according to neoadjuvant treatment Characteristics All cases Age (years), median (range) FIGO stage IIB III Tumor diameter (cm) <4 ‡4 Grade 1–2 3 Not available MRI-assessed pelvic node status Negative Positive Radical hysterectomy Type II Type III Type IV Type V a b c

CT/RT No. (%)

NACT No. (%)

P valuea

87 52 (25–72)

40 54 (27–78)

0.2b

70 (80.4) 17 (19.6)

35 (87.5) 5 (12.5)

0.4

20 (23.9) 67 (77.0)

16 (40.0) 24 (60.0)

0.08

25 (40.3) 37 (59.7) 25

18 (52.9) 16 (47.1) 6

0.3

53 (60.9) 34 (39.1)

24 (60.0) 16 (40.0)

0.8

12 49 24 2

5 25 9 1

0.8c

(13.8) (56.3) (27.6) ( 2.3)

(12.5) (62.5) (22.5) ( 2.5)

Calculated by FisherÕs exact or v2 test. Calculated by Kruskall–Wallis non-parametric test. Type II–III versus IV–V.

Neoadjuvant chemoradiation was performed as previously described.10 Briefly, radiotherapy was administered to the whole pelvic region in 22 fractions (1.8 Gy/day, totaling 39.6 Gy) in combination with cisplatin (2 h intravenous infusion of 20 mg/m2) plus 5-fluorouracil (1,000 mg/m2) both on days 1–4 and 27–30. NACT (two to three cycles) was performed by using cisplatin-based chemotherapy.18 Response Assessment and Surgery Four weeks after the end of neoadjuvant treatments, patients were reassessed by following the same clinical and imaging procedures described above and response was recorded according to WHO criteria.19 Patients achieving complete or partial response to treatment underwent surgery while patients experiencing no change or progression of disease were considered for salvage chemotherapy. Surgery consisted of Type II (n = 17), Type III (n = 74), Type IV (n = 33), or Type V (n = 3) radical hysterectomy according to Piver classification,20 with bilateral systematic pelvic lymphadenectomy. If pelvic nodes were intraoperatively defined as positive for tumor metastasis, para-aortic lymphadenectomy up to inferior mesenteric artery was carried out. Statistical Analysis The distribution of clinico-pathological features and lymphnode status according to type of neoadju-

vant treatment was analyzed by FisherÕs exact test or v2 test. The difference in the distribution of positive lymphnodes and rate of metastatic node involvement were examined with Kruskall–Wallis non-parametric test and FisherÕs exact test or v2 test. Pearson correlation test was used to analyze the correlation between the number of lymphnodes removed and the number of metastatic lymphnodes. Negative predictive value (NPV) and positive predictive value (PPV) of LPN status versus parametrial and UPN status were calculated. Statistical analysis was performed by using SOLO (BMDP Statistical Softwares, Los Angeles, CA, USA) and Crunch Interactive Statistical package (Crunch Software Corporation, San Francisco, CA, USA). RESULTS Rate and Pattern of Metastatic Lymphnode Involvement Table 1 shows the characteristics of the patients in the overall series, and in the two treatment groups, and confirms the adequacy of matching since CT/RT and NACT groups appear well balanced in terms of age, FIGO stage, grade, and MRI-assessed pelvic lymphnode involvement; a slightly higher percentage of cases with tumor diameter ‡4 cm, was present in CT/RT compared to NACT group, although the difference did not reach the statistical significance. Ann. Surg. Oncol. Vol. 14, No. 3, 2007

1132

G. FERRANDINA ET AL.

TABLE 2. Number of removed lymphnodes and rate of metastatic node involvement according to node stations in squamous LACC patients administered preoperative chemoradiation (CT/RT) versus neoadjuvant chemotherapy (NACT) Lymphnode stations Para-aortic Pelvic Parametrial a b

No. No. No. No. No. No.

of nodes removed, median (range) (%) of cases with metastatic node involvement nodes removed, median (range) (%) of cases with metastatic node involvement nodes removed, median (range) (%) of cases with metastatic node involvement

CT/RT

NACT

9 4 37 10 3 1

14 3 46 12 5 5

(2–46) (12.1) (13–73) (11.5) (1–16) (1.5)

(7–28) (13.6) (5–87) (30.0) (1–16) (12.5)

P valuea 0.2 0.8b 0.0003 0.009b 0.002 0.011b

Calculated by Kruskall–Wallis non-parametric test. Calculated by FisherÕs exact test for proportion.

As far as pathological response is concerned, in CT/RT group complete pathological response/persistence of microscopic tumor foci (microPR) was documented in 66 cases (75.9%), while macroscopic partial response/stable disease were observed in 21 cases (24.1%); in NACT group complete pathological response/microPR was documented in 9 cases (22.5%), while macroscopic partial response/stable disease were observed in 31 cases (77.5%) (data not shown). As shown in Table 2, the overall number of removed para-aortic lymphnodes did not differ according to type of neoadjuvant treatment (median 9, range 2–46 in cases administered preoperative CT/ RT vs. median 14, range 7–28 in cases administered NACT, P value = 0.2), and there was no difference in the rate of metastatic involvement at para-aortic level in cases administered CT/RT (12.1%) versus NACT (13.6%, P value = 0.8). On the other hand, the number of removed pelvic lymphnodes resulted lower in cases administered preoperative CT/RT (median 37, range 13–73) versus NACT (median 46, range = 5–87, P value = 0.0003). The percentage of cases with metastatic pelvic node involvement was significantly lower in cases administered CT/RT (11.5%) compared to cases administered NACT (30.0%) (P value = 0.009). In addition, the number of metastatic pelvic lymphnodes was lower in cases undergoing preoperative CT/RT (median 1, range 1–4), versus NACT (median 3, range 1–10) (P value = 0.026). Moreover, in patients administered CT/RT, who showed metastatic pelvic lymphnode involvement (n = 10), the overall number of metastatic lymphnodes was 16 out of 460 lymphnodes removed, giving a percentage of metastatic pelvic lymphnodes of 3.5%. In patients administered NACT, who showed metastatic pelvic lymphnodes (n = 12), the overall number of metastatic nodes was 75 out of 511 lymphnodes removed, giving a percentage of metastatic nodes of 14.7%. Therefore, in patients administered CT/RT versus Ann. Surg. Oncol. Vol. 14, No. 3, 2007

NACT, besides a reduction of the percentage of cases with metastatic pelvic lymphnodes, there was also a reduction by 76.2%, of the percentage of lymphnodes involved. Finally, we investigated the correlation between the number of pelvic lymphnodes removed and the number of metastatic pelvic lymphnodes for each patient in the overall series, as well as in each treatment group, obtaining the same results, consisting in the absence of any correlation between the number of lymphnodes removed and the number of metastatic lymphnodes (data not shown). The number of lymphnodes detected in the parametrium was significantly lower in cases administered CT/RT (median 3, range 1–16) than NACT (median 5, range 1–16) (P value = 0.002). In the group of preoperative CT/RT metastatic parametrial involvement was observed in 1/66 (1.5%) cases. In one case a direct invasion of only parametrial tissue was documented. Metastatic involvement of parametrial nodes was lower in cases administered CT/ RT than NACT (1.5 vs. 12.5%, P value = 0.011). Prediction of Metastatic Pelvic Lymph Node Involvement Table 3 shows the pelvic lymphnode status according to different clinico-pathological parameters. In the group of cases administered CT/RT, stage of disease, tumor size, and MRI-assessed extent of residual cervical disease after CT/RT failed to predict lymphnode status, while MRI-assessed pelvic lymphnode status, both at staging work-up and at response evaluation was associated with pathologic pelvic lymphnode status; the significance of the negative as well as the positive predictive values have to be considered with caution given the relatively low number of cases with positive pelvic lymphnodes. As expected, cases with pathologically documented absent or only microscopic residual cervical disease after chemoradiation showed a lower percentage of

1133

LYMPHNODES AFTER PREOPERATIVE CHEMORADIATION

TABLE 3. Prediction of pathologically assessed pelvic lymphnode status on the basis of clinical characteristics CT/RT Characteristics

Pathological pelvic node status

Negative Positive FIGO stage IIB 63 7 III 15 2 Tumor diameter (cm) <4 18 2 ‡4 60 7 MRI-assessed pelvic node status (at staging) Negative 51 2 Positive 27 7 MRI-assessed pelvic node status (post-treatment) Negative 72 6 Positive 6 3 MRI-assessed residual cervical disease Absent 42 5 Present 36 4 a

NACT P valuea

Pathological pelvic node status

P valuea

Negative

Positive

0.8

25 3

10 2

0.9

0.7

13 15

3 9

0.3

0.025

22 6

2 10

0.047

27 1

8 4

0.02

0.7

8 20

0 12

0.04

0.0004

Calculated by FisherÕs exact test for proportion.

metastatic pelvic lymphnodes compared to cases with macroscopic residual cervical disease regardless of type of neoadjuvant treatment (data not shown). Interestingly enough, in the group of cases administered NACT, not only MRI-assessed pelvic lymphnode status, but also the extent of residual cervical disease after treatment proved to be associated with pathologic pelvic lymphnode status. We were then prompted at investigating whether the pathologically assessed status of LPN was able to predict the status of parametria and UPN at final histopathologic diagnosis in cases administered CT/ RT versus NACT. As shown in Table 4, in patients administered CT/ RT, in case of histologically negative LPN (n = 81), 80 cases had no metastatic parametrial involvement (NPV = 98.7%), while of 6 cases with histologically positive LPN, only 1 had metastatic parametrial involvement (PPV = 16.7%). In patients administered NACT, the NPV and PPV of pathologically assessed LPN compared to parametrial involvement were 93.5 and 33.3%, respectively. As shown in Table 5, in patients administered CT/ RT, of 81 cases with histologically negative LPN, 79 showed histologically negative UPN (NPV = 97.5%), while of 6 cases with histologically positive LPN, only 1 resulted positive at UPN level (PPV = 16.7%). In patients administered NACT, of 31 cases with histologically negative LPN, 30 showed histologically negative UPN (NPV = 96.8%), while of 9 cases with positive LPN, 5 resulted positive at UPN (PPV = 55.5%).

Prediction of Metastatic Para-aortic Lymph Node Involvement It has to be noted that, although the overall population was selected on the basis of absence of lymphoadenopathies in the para-aortic area as assessed by MRI, 7 out of 55 (12.7%) cases undergoing paraaortic lymphadenectomy showed metastatic involvement of para-aortic nodes.

DISCUSSION The present data demonstrated that the rate of metastatic pelvic lymphnode involvement in LACC patients administered CT/RT is significantly lower compared to NACT group and well matches with recently published results in a similar clinical setting13: in particular, CT/RT seems to lessen more than 60% the chances of finding cases with positive pelvic lymphnodes compared to patients administered NACT, and this difference is even emphasized when considering that the percentage of positive nodes among the retrieved ones, was reduced by 76.2% in CT/RT-treated cases compared to NACTtreated cases. This observation could result not surprising given the synergistic activity of platinumbased chemotherapy and radiation21 which would in turn translate into the achievement of a better pelvic control, and supports the observation that the rate of metastatic involvement of lymph nodes in the paraaortic area, clearly outside the irradiation field, is Ann. Surg. Oncol. Vol. 14, No. 3, 2007

1134

G. FERRANDINA ET AL.

TABLE 4. Predictive role of pathologically assessed lower pelvic node (LPN) versus parametrium in squamous LACC patients administered preoperative CT/RT versus NACT

Metastatic LPN Negative Positive NPV (%) PPV (%)

CT/RT

NACT

Metastatic parametrium

Metastatic parametrium

Negative no.

Positive no.

80 5 98.7 16.7

1 1

Negative no.

Positive no.

29 2 93.5 33.3

2 3

TABLE 5. Predictive role of pathologically assessed LPN versus upper pelvic node (UPN) status in squamous LACC patients administered preoperative CT/RT versus NACT

Metastatic LPN Negative Positive NPV (%) PPV (%)

Negative no.

CT/RT

NACT

Metastatic UPN

Metastatic UPN Positive no.

79 5 97.5 16.7

similar in RT/CT and NACT administered patients. The lower number of pelvic lymphnode metastases in CT/RT-treated cases is unlikely to be related to a less radical lymphadenectomy leading to retrieve a lower number of pelvic nodes: indeed (1) it has been reported that radiotherapy by itself reduces the number of detectable lymphnodes;22 (2) moreover, we found that there was no correlation between the number of lymphnodes removed and those affected by tumor; (3) we have to consider that also in the parametrium the number of retrieved nodes resulted lower with respect to NACT-treated cases, although the extent of radical hysterectomy was comparable between the two groups, and iv)finally, there was no difference in the number of lymphnodes removed in CT/RT- versus NACT-treated cases in non-irradiated para-aortic areas, thus confirming that the lower percentage of metastatic parametrial as well as pelvic nodes is related to the effects of CT/RT rather than to a less extended surgery. Since one of the greatest issues in the management of cervical cancer remains the control of lymph node disease4, the availability of pre-operative and/or intra-operative predictors of lymph node involvement would be of utmost clinical relevance in order to individualize treatment and try selecting cases to be eventually triaged to less radical surgical approaches. Differently from NACT-treated group, in which both Ann. Surg. Oncol. Vol. 14, No. 3, 2007

2 1

Negative no. 30 4 96.8 55.5

Positive no. 1 5

post-treatment MRI-assessed pelvic node status and extent of residual cervical disease resulted associated with pathological pelvic node status, in CT/RTtreated group MRI-assessed pelvic lymphnode status after treatment resulted correlated with pathological findings, even though these data have to be taken with caution given the relatively small sample series. Although our data and also very recent findings13 showed that residual pelvic node disease is observed in a lower percentage of cases with complete or microscopic pathologic response than in cases with macroscopic residual tumor in the cervix, our current observations also underline the need not to underestimate the risk of pelvic node disease in case of presumptive preoperative complete response on primary tumor in this clinical setting. We also analyzed for the first time in CT/RT group, the relationship between the pathological status of the node stations considered the sentinel nodes in cervical cancer (LPN)3,14–16 and other lymphnode stations located on the truck of lymphatic diffusion in cervical cancer (UPN). Similarly to what reported for early and NACT-treated LACC patients,15,16 it seems that the pathological status of LPN could be predictive of the pathological status of UPN. However, given the small sample series, the possibility that radicality of surgery can be tailored according to the status of LPN has to be considered

LYMPHNODES AFTER PREOPERATIVE CHEMORADIATION

with caution and verified on large numbers before any definitive conclusion can be drawn. In conclusion, we showed that (1) preoperative CT/ RT induces a higher control of pelvic node disease in LACC patients compared to NACT; (2) there is no association between the extent of clinically assessed residual cervical disease after CT/RT and pelvic node status, thus underlining the need to be cautious in estimating the risk of pelvic node disease even in case of clinical complete response on primary tumor; (3) similarly to what demonstrated in NACT-treated patients, the status of LPN has a high NPV with respect to parametrial and UPN status. It remains to be verified whether the high rate of pelvic control reflects on the rate and pattern of recurrence and, most important, on patient clinical outcome. A major concern remains for the assessment of para-aortic lymphnode status and, above all, for the management of cases with high risk of para-aortic node involvement, since the chances of finding positive para-aortic nodes are not modified in CT/RT with respect to NACT-treated cases. In this context, the wider application of more sophisticated imaging techniques for detection of node disease, such as 2[18]-fluoro-2-deoxy-d-glucose (FDG) positron emission tomography (PET) would hopefully improve the discrimination of metastatic lymphnodes23. Moreover, one could also take advantage of the improved performances of laparoscopy-based approaches, which have raised novel interest in the surgical staging of cervical cancer.23

REFERENCES 1. Goff B, Muntz HG, Paley PJ, Tamimi HK, et al. Impact of surgical staging in women with locally advanced cervical cancer. Gynecol Oncol 1999; 74:436–42. 2. Cosin JA, Fowler JM, Chen MD, Paley PJ, et al. Pretreatment surgical staging of patients with cervical carcinoma: the case for lymph node debulking. Cancer 1998; 82:2241–8. 3. Benedetti Panici P, Cutillo G, Angioli R. Modulation of surgery in early invasive cervical cancer. Crit Rev Oncol/Hematol 2003; 48:263–270. 4. Morice P, Castaigne D, Pautier P, et al. Interest of pelvic and paraaortic lymphadenectomy in patients with stage IB and II cervical carcinoma. Gynecol Oncol 1999; 73:106–10. 5. Berman ML, Keys H, Creasman W, et al. Survival and patterns of recurrence in cervical cancer metastatic to periaortic nodes (a Gynecologic Oncology group Study). Gynecol Oncol 1984; 19:8–16.

1135

6. Green J, Kirwan J, Tierney J, et al. Concomitant chemotherapy and radiation therapy for cancer of the uterine cervix. Cochrane Database System Review 3: 2005 (CD002225). 7. Benedetti Panici P, Greggi S, Scambia G, et al. Long-term survival following neoadjuvant chemotherapy and radical surgery in locally advanced cervical cancer. Eur J Cancer 1998; 34(3):341–6. 8. Benedetti Panici P, Greggi S, Colombo A, et al. Neoadjuvant chemotherapy and radical surgery versus exclusive radiotherapy in locally advanced squamous cell cervical cancer: results from the Italian multicenter randomized study. J Clin Oncol 2001; 20:179–8. 9. Jurado M, Martinez-Monge R, Garcia-Foncillas J, et al. A Pilot study of concurrent cisplatin, 5-fluorouracil, and external beam radiotherapy prior to radiotherapy in locally advanced cervical cancer. Gynecol Oncol 1999; 74:30–7. 10. Distefano M, Fagotti A, Ferrandina G, et al. Preoperative chemoradiotherapy in locally advanced cervical cancer: long term outcome and complications. Gynecol Oncol 2005; 1:S166– 70. 11. Houvenaegel G, Lelievre L, Gonzague-Casabianca L, et al. Long-term survival after concomitant chemoradiotherapy prior to surgery in advanced cervical carcinoma. Gynecol Oncol 2006; 100:338–43. 12. Benedetti Panici P, Maneschi F, Scambia G, et al. Lymphatic spread of cervical cancer: an anatomical and pathological study based on 225 radical hysterectomies with systematic pelvic and aortic lymphadenectomy. Gynecol Oncol 1996; 62:19–24. 13. Houvenaeghel G, Lelievre L, Rigouard AL, et al. Residual pelvic lymphnode involvement after concomitant chemoradiation for locally advanced cervical cancer. Gynecol Oncol 2006; 102(1):74–9. 14. Benedetti Panici P, Angioli R, Palaia I, et al. Tailoring the parametrectomy in stages IA2-IB1 cervical carcinoma: is it feasible and safe?. Gynecol Oncol 2005; 96:792–8. 15. Ferrandina G, Distefano M, Ludovisi M, et al. Assessment of pathological status of lower pelvic nodes to tailor surgery in not squamous early cervical carcinoma. Gynecol Oncol 2005; 99(1):246–7. 16. Scambia G, Ferrandina G, Distefano M, et al. Is there a place for a less extensive radical surgery in locally advanced cervical cancer patients? Gynecol Oncol 2001; 83:319–24. 17. Katz MH. Study design and statistical analysis: a practical guide for clinicians. Cambridge University press 2006, 27–28. 18. DÕAgostino G, Ferrandina G, Greggi S, et al. Neoadjuvant treatment of locally advanced carcinoma of the uterine cervix with epirubicin, paclitaxel, and cisplatin. Cancer Chemother Pharmacol 2002; 49(3):256–60. 19. World Health Organization WHO handbook of cancer treatment Geneva: WHO offset publication; 1979 p 48. 20. Piver MS, Rutledge F, Smith JP. Five classes of extended hysterectomy. Obstet Gynecol 1974; 44:265–72. 21. Vokes EE, Weichselbaum RR. Concomitant chemoradiotherapy: rationale and clinical experience in patients with solid tumors. J Clin Oncol 1990; 8:911–34. 22. Lagasse LD, Smith ML, Moore JG, et al. The effect of radiation therapy on pelvic lymph node involvement in stage I carcinoma of the cervix. Am J Obstet Gynecol 1974; 119(3):328–34. 23. Schneider A, Hertel H. Surgical and radiographic staging in patients with cervical cancer. Curr Opin Obstet Gynecol 2004; 16:11–8.

Ann. Surg. Oncol. Vol. 14, No. 3, 2007