Concomitant chemoradiation versus neoadjuvant chemotherapy in locally advanced cervical carcinoma: results from two consecutive phase II studies

Annals of Oncology 13: 1212–1219, 2002 DOI: 10.1093/annonc/mdf196

Original article

Concomitant chemoradiation versus neoadjuvant chemotherapy in locally advanced cervical carcinoma: results from two consecutive phase II studies A. Duenas-Gonzalez1,2*, C. Lopez-Graniel1, A. Gonzalez-Enciso1, A. Mohar1,2, L. Rivera1, A. Mota1, R. Guadarrama1, G. Chanona1 & J. de la Garza1 1

Departments of Medical Oncology, Gynecology, Radiotherapy, Pathology and Basic Research, Instituto Nacional de Cancerología, Mexico City; Department of Genomic Medicine and Environmental Toxicology, Instituto de Investigaciones Biomédicas, UNAM, Mexico City, Mexico

2

Received 3 September 2001; revised 14 December 2001; accepted 16 January 2002

Background: Randomized studies comparing induction chemotherapy followed by surgical resection with radiation alone found that the neoadjuvant approach produces better results. So far, this latter modality has not been compared with standard concomitant chemoradiation. The objective of this report was to compare the results of two consecutive phase II studies: neoadjuvant chemotherapy followed by surgery or chemoradiation for the unresectable cases versus standard cisplatin-based chemoradiation. Patients and methods: From February 1999 to July 1999, 41 patients with cervical carcinoma, stages IB2–IIIB, were treated with neoadjuvant chemotherapy. Treatment consisted of three 21-day courses of cisplatin 100 mg/m2 on day 1 and gemcitabine 1000 mg/m2 on days 1 and 8, followed by either surgery or concomitant chemoradiation for the non-operable cases. From August 1999 to December 1999, an equal number of patients having comparable clinicopathological characteristics were treated with six weekly courses of cisplatin 40 mg/m2 during standard pelvic radiation. Results: A total of 82 patients were analyzed. Both groups were similar with regard to age, histology, International Federation of Gynecology and Obstetrics (FIGO) stage, tumor size, pretreatment hemoglobin levels, parametrial infiltration and performance status. In the neoadjuvant arm the overall response rate to induction chemotherapy was 95% (95% confidence interval 88% to 100%). Twentythree patients had surgery and 14 underwent chemoradiation. In the definitive chemoradiation study, 38 patients completed treatment, the median number of cisplatin courses was six for a dose intensity of 33 mg/m2/week. Doses to points A and B were 85 Gy (range 68–95) and 55 Gy (range 51–65), respectively. Chemoradiation was delivered in 44.6 (range 28–113) days. Complete response rates after all treatment were similar: 97% and 87% in the neoadjuvant and chemoradiation groups, respectively. At a median follow-up of 28 (range 2–33) and 24 (range 3–30) months, respectively, there were no differences in overall survival. To date, 15 and 13 patients in the neoadjuvant and chemoradiation groups, respectively, have died of disease (P = 0.8567). Conclusions: The results of this non-randomized comparison suggest that induction chemotherapy followed by surgery or chemoradiation is at least as effective in terms of response and survival as standard cisplatin-based chemoradiation. A randomized study is needed to confirm these findings. Key words: cervical cancer, concomitant chemoradiation, neoadjuvant chemotherapy

Introduction Cervical carcinoma is the most frequent cause of death from cancer in women from developing countries, and most of these cases are locally advanced at diagnosis [1]. Almost a century *Corresponding author: Dr Alfonso Duenas-Gonzalez, Investigacion Basica, Instituto Nacional de Cancerologia, San Fernando 22, Tlalpan 14080, D.F. Mexico. Tel: +525-628-0424; Fax: +525-628-0432; E-mail: [email protected] © 2002 European Society for Medical Oncology

after its introduction as a treatment, radiation remains the main therapeutic modality for cervical carcinoma. Radiation treatment did not undergo any major modifications until the recent addition of chemotherapy in a concomitant setting. Five randomized studies accruing almost 2000 patients have demonstrated the superiority of the arms with cisplatin-based chemotherapy during pelvic radiation [2–6]. Additional evidence supporting these results has come from two additional phase III trials utilizing other radiosensitizers [7, 8]. These

1213 studies have confirmed that the benefits of chemoradiation are not limited to surgically staged patients and that the patients with more advanced FIGO stage (IIIB) benefit the most [8]. Adjuvant and neoadjuvant chemotherapy, however, have been tested in cervical carcinoma for many years without success. Several adjuvant trials with either chemotherapy alone or sequential radiation and chemotherapy for high-risk, surgically treated early stage patients have failed to prolong survival [9–11]. Moreover, at least eight phase III randomized trials comparing radiation alone with neoadjuvant chemotherapy followed by definitive radiation in locally advanced cervical cancer show no advantage in terms of response and overall survival [12–18]. Despite these results, investigators have attempted to improve the results of neoadjuvant chemotherapy incorporating radical surgery to consolidate the response to chemotherapy. This latter approach is based on the fact that surgery may bypass the cross-resistance between radiation and chemotherapy. Four randomized studies comparing this modality against radiation alone have yielded positive results. Sardi et al. [19–21], using ‘quick’ PVB (cisplatin, vincristine, bleomycin), reported a longer overall survival in three separate studies involving stage IB2 [19], IIB [20] and IIIB patients [21]. Another study in patients with stages IB2–IIIB [22] also reported better survival rates for the chemotherapy arm, using several combinations of chemotherapy based on cisplatin. Of note, in this latter study, the benefit derived only reached statistical significance for stage IB and IIB patients. At our institution, radiation has been the main modality of treatment for locally advanced disease but other modalities have also been tested. From February 1999 to July 1999, we conducted a study of neoadjuvant chemotherapy followed by surgery [23, 24], and from August 1999 to December 1999 we performed a consecutive study of cisplatin-based chemoradiation. The objective of this study was to analyze the results of treatment with these modalities in a comparable patient population.

Patients and methods A total of 82 newly diagnosed patients were entered into two consecutive phase II trials (41 per group) at the Instituto Nacional de Cancerología. All patients were previously untreated and had a histological diagnosis of squamous, adenosquamous or adenocarcinoma cervical carcinoma, and were stage IB2–IIIB according to the FIGO classification. The clinical evaluation for staging was under no anesthesia and performed by two examiners, and in the case of disagreement, a third examiner was required to reach a consensus. In all cases, a detailed drawing of the lesion was registered and intravenous pyelography, cystoscopy and proctosigmoidoscopy were performed as appropriate. The status of para-aortic lymph nodes was not assessed. In addition, patients required the following inclusion criteria in order to be analyzed: (i) measurable disease; (ii) age 18–70 years; (iii) Karnofsky performance status of ≥70%; (iv) normal hematological, renal and hepatic function as follows: hematological, Hb ≥10 g/l (however, patients could be transfused before any treatment to reach this level of hemoglobin); leukocytes >4000/mm3; platelets

>100 000/mm3; total bilirubin and transaminases <1.5× upper limit of normal and normal creatinine; (v) a normal chest X-ray; and (vi) informed consent. Exclusion ciriteria were as follows: (i) other histologies, such as neuroendocrine small cell carcinoma, lymphoma, sarcoma and other rare histologies; (ii) stages IA, IB1 and IV; (iii) non-measurable disease; (iv) severe or uncontrolled infection or other systemic diseases; (v) concomitant treatment with any experimental drug; (vi) pregnant or nursing women; (vii) mental illness and (viii) previous or concomitant malignant diseases other than non-melanoma skin cancer.

Neoadjuvant chemotherapy The details of treatment for this group of patients have already been reported [23]. Briefly, it consisted of intravenous gemcitabine at a dose of 1000 mg/m2 diluted in 500 ml normal saline administered over 30 min on days 1 and 8, followed by cisplatin 100 mg/m2 administered over 4 h on day 1. Conventional hydration and antiemetic treatments were used. A total of three 21-day courses were administered. Chemotherapy was held in case of prohibitive toxicity or progression of disease. In both cases, patients were taken for immediate locoregional treatment. After neoadjuvant chemotherapy patients were submitted to type III radical hysterectomy and bilateral pelvic lymph node dissection if the multidisciplinary team judged the disease could be resected obtaining margins free of disease. Those patients with partial response who were judged to have non-resectable disease underwent pelvic radiation (see below) with concomitant 6-weekly doses of cisplatin 40 mg/m2 during external radiation. Locoregional treatment, either surgery or chemoradiation, was performed within 3 weeks of finishing the third cycle of chemotherapy. Postoperative chemoradiation was used only in cases with positive surgical margins.

Chemoradiation Patients received external beam radiation using 60Co or lineal accelerator equipment with a minimum photon-beam energy of 4 MeV at a target or skin source distance of 80 cm to the whole pelvis for a total dose of 50 Gy (5 weeks, 200 cGy fractions from Monday to Friday) followed by one or two intracavitary cesium (low-dose rate) applications 2 weeks after finishing external radiation. The planned total dose to points A and B were 85 and 55 Gy, respectively. Patients were treated with the four-field box technique as follows: the irradiated volume was to include the whole uterus, the paracervical, parametrial and uterosacral regions, as well as the external iliac, hypogastric and obturator lymph nodes. Minimum margins were the upper margin of L-5, (superiorly) the midportion of the obturator foramen or the lowest extension of the disease (inferiorly), and 1 cm beyond the lateral margins of the bony pelvis and its widest plane (laterally). For the lateral fields, the anterior margin was the anterior edge of the symphysis or 3 cm in front of the sacral promontory. The posterior margin was the S2–S3 interspace. Cisplatin was administered for 6 weeks during external radiation, beginning with the first day of radiation. For logistic reasons the cisplatin infusion was administered 2 h either before or after the application of radiation. A dose of 40 mg/m2 was used and administered by a peripheral vein in an outpatient setting as follows: 1000 ml normal saline for 1 h followed by cisplatin diluted in 500 ml normal saline containing 62.5 ml of 20% mannitol for 1 h, followed by 500 ml normal saline for 30 min. Intravenous dexametasone 8 mg and ondansetron 8 mg were employed as antiemetic prophylaxis. Cisplatin was withheld in any case of grade 3 toxicity (except nausea/vomiting) until the toxicity regressed to less than

1214 grade 3. In patients with grade 3 toxicity that persisted beyond 2 weeks, chemotherapy was no longer administered.

Table 1. Clinical characteristics of patients Neoadjuvant [n (%)]

Chemoradiation [n (%)]

Number

41

41

Mean age, years (range)

48.3 (32–65)

43.8 (26–68)

38 (92)

37 (90)

3 (8)

4 (10)

Evaluation of response Response to chemoradiation was clinically and cytologically evaluated 3 months after the end of treatment. The response to neoadjuvant chemotherapy was clinically evaluated after the third course and the pathological response was assessed in those cases that went to surgery. In patients that received chemoradiation (in the neoadjuvant arm) instead of surgery, the assessment of clinical response was also carried out 3 months after completing chemoradiation. Complete response was registered when no clinical or cytological evidence of disease existed. All other cases (persistence or progression) were registered as no complete response. A clinical complete response to neoadjuvant chemotherapy was registered when no evidence of disease existed, partial response was registered when there was a ≥50% reduction in the product of the longest perpendicular diameters of the disease, stable disease with <50% reduction or an increase of <25% in the product of the longest perpendicular diameters of the disease, and progressive disease with a >25% increase or the appearance of new lesions. Complete pathological response was registered when there was no residual disease; microscopic response when foci or disease measuring <5 mm remained and macroscopic response when there was any measurable residual disease >5 mm in diameter.

Evaluation of toxicity

Histology Squamous Ad/Adenosq Stage IB2–IIA

6 (15)

6 (15)

IIB

20 (49)

17 (41)

IIIB

15 (36)

18 (44)

33.5

34.5

Unilateral

14 (40)

17 (49)

Bilateral

21 (60)

18 (51)

12.7 (7.5–15.8)

11.9 (6.4–16.8)

100

14 (34)

17 (41)

90

16 (39)

20 (49)

80

11 (27)

4 (10)

Mean tumor size, cm2 Parametrial infiltration

Hemoglobin, g/dl (range) Karnofsky performance status

The acute and chronic toxicities to chemoradiation were evaluated according to the Radiation Therapy Oncology Group (RTOG) toxicity criteria. Toxicity to neoadjuvant chemotherapy was evaluated using the World Health Organization criteria.

There were no significant differences for any of the parameters. Ad/Adenosq, adenocarcinoma/adenosquamous.

Survival

Neoadjuvant chemotherapy group

Patients were followed up with visits every 3 months in which both complete and pelvic examinations were performed, as well as blood counts, clinical chemistry and chest X-rays. Computed tomography scans, ultrasound and other studies were carried out when appropriate. Survival was analyzed on the basis of intention-to-treat and it was considered from the date of diagnosis until death or the patient’s last visit. Curves were constructed using the method of Kaplan–Meier [25] and the log-rank test [26] to assess differences between groups. The chi-square and t-tests were used when appropriate to compare patient characteristics, responses and toxicity.

The results of treatment of these patients have been reported previously [22]. Briefly, a total of 119 courses of chemotherapy were delivered. One patient developed renal failure during the first course of chemotherapy and received radiation alone, achieving complete response. Three patients abandoned treatment during the second (two patients) and third course (one patient) of neoadjuvant therapy, and were locoregionally treated elsewhere (one with incomplete radiation and the other with simple hysterectomy). Two of these patients went back to our institution with recurrent disease whereas the other patient was lost to follow-up. Thus, 37 patients completed their treatment, and of these, 23 underwent surgery and 14 received definitive chemoradiation (two of 14 patients were considered operable but had a medical contraindication to surgery) (Table 2).

Results Patient characteristics A total of 82 patients were analyzed. Neoadjuvant chemotherapy patients were treated from February 1999 to July 1999 and the chemoradiation patients from August 1999 to December 1999. Clinical characteristics of the two groups of patients are shown in Table 1. The clinicopathological characteristics were well-balanced and there were no differences in age, histology, stage, tumor size, parametrial infiltration, hemoglobin levels or performance status. In addition (data not shown), socioeconomic and demographic status were also similar in both groups of patients.

Chemoradiation group All but three patients completed chemoradiation treatment. Table 3 shows the details of treatment. The median number of cisplatin courses administered was six (range 3–6) for a dose intensity of 33 mg/m2/week. The mean doses to points A and B were 85 Gy (range 68–95) and 55 Gy (range 51–65), respectively. The mean treatment duration time was 44.6 (range 28–123) days.

1215 Table 2. Overall treatment of the neoadjuvant chemotherapy arm

Table 4. Clinical response at the end of treatment

No. of patients (n = 41)

Treatment

Complete

Persistence/progression

Locoregional treatment

37

Neoadjuvant (n = 37)

36 (97%)

1 (3%)

Surgerya

23

Chemoradiation (n = 38)

33 (87%)

5 (13%)

Chemoradiation

14

None or incomplete treatment

4

a One patient with positive surgical margins received postoperative chemoradiation.

Table 3. Treatment of the chemoradiation group No. of patients (n = 41)

Chemoradiation

Completed treatment

38

No. of courses (median)

6

6

67.5%

5

20.0%

4

7.5%

3

5.5%

Dose point A

8578 (range 6800–9526)

B

6064 (range 5100–6463)

Treatment time (days)

44.6 (range 28–123)

Clinical response In the neoadjuvant arm, 40 patients who received at least two courses of treatment were evaluated for clinical response to chemotherapy. Thirty-eight patients had an objective response for an overall response rate of 95% (7.5% partial and 87.5% complete). Among the 37 patients that completed locoregional treatment, 23 underwent surgery of these, six patients had pathological complete responses (26%), three had microscopic (13%) and 14 had macroscopic (61%) residual disease. One patient had positive surgical margins and received postoperative chemoradiation. In regard to the 14 non-resectable patients receiving chemoradiation, all but one had clinical complete response. The final response rates for the neoadjuvant group (37 patients that completed chemotherapy and locoregional treatment) and for the standard chemoradiation group (38 patients that completed treatment) are shown in Table 4. Thirty-six patients in the neoadjuvant group had complete response (97%) whereas 33 (87%) in the chemoradiation arm had complete response. This difference was not statistically significant (P = 0.2139).

Survival

The overall response rates for the 37 and 38 patients that completed all treatment, respectively; P = 0.2139.

for the neoadjuvant chemotherapy and chemoradiation groups, respectively, the median survival had not been reached and overall survival was 62% (95% CI 0.45–0.75) and 65% (95% CI 0.47–0.79) with 15 and 13 deaths per group, respectively (P = 0.8567) (Figure 1). With regard to the pattern of relapses, Table 5 shows that 17 (45%) and 13 (34%) patients relapsed in the neoadjuvant and chemoradiation groups, respectively, with pelvic/systemic failures being the most common type of relapse.

Toxicity Neoadjuvant chemotherapy was well tolerated; the most common side effects were nausea/vomiting and alopecia. Hematological toxicity was moderate, no episodes of bleeding or neutropenic fever were seen. This has been described previously [23]. The toxicities of the 14 patients receiving chemoradiation, as well as for those patients in the chemoradiation arm, are shown in Table 6. As expected, the hematological and gastrointestinal toxicities were higher than that seen in patients treated with radiation alone, although remarkably, no grade 4 toxicity was seen. Interestingly, among the patients in the neoadjuvant arm that were not operated and treated with chemoradiation, 85% of them presented grade 1/2 neurological toxicity, which can be explained by the total dose of cisplatin received during induction and consolidation. With regard to late toxicity in the neoadjuvant group, there were no late surgical complications in the 23 patients operated on, whereas in the 14 chemoradiation patients, there were two patients with grade 2 proctitis and two with grade 3 proctitis (28%). Among the chemoradiation group, grades 1, 2, 3 and 4 proctitis occurred in two, six, one and one patients, respectively (26%).

Table 5. Relapses and survival status Status

Neoadjuvant

CT/RT

At risk

37

38

Recurred

17

13

Local

6

2

Systemic

2

2

Combined

9

9

20

25

Disease-free

At a maximum follow-up time of 33 and 30 months, median follow-up time of 28 (range 2–33) and 24 (range 3–30) months

CT/RT, chemotherapy/radiotherapy.

1216

Figure 1. Overall survival (intention-to-treat analysis). Survival stands at 62% (95% CI 0.45–0.75) and 65% (95% CI 0.47–0.79).

Table 6. Acute toxicity RTOG criteria (% per patient) Neoadjuvanta Grade 1–2

CT/RT Grade 3–4

Grade 1–2

Grade 3–4

Skin

28

0

17

0

Upper GI

64

0

90

10b

Lower GI

71

0

53

2b

GU

21

0

WBC Neutrophils Hemoglobin

64

44

0

21

b

61

12b

7

b

61

5b

19u (0.4)c

–

64 4u (0.3)c

–

Platelets

14

0

0

0

Neuro/sensor

85

0

0

0

a

Toxicity for the subset of 14 patients who were not operated on and received CT/RT as definitive treatment in the neoadjuvant treatment arm. b All cases were grade 3. c Units of red blood cells transfused. CT/RT, chemotherapy/radiotherapy; RTOG, Radiation Therapy Oncology Group.

Discussion Pelvic failures with or without a systemic component is a frequent problem for patients with locally advanced cervical carcinoma, despite the recent encouraging results of concomitant chemoradiation based on cisplatin [2–6] or other radiosensitizing drugs [7, 8]. These results emphasize the need to test new treatment alternatives in order to improve survival. To our knowledge, this is the first study that has compared the new standard of treatment, chemoradiation, against neoadjuvant chemotherapy followed by surgery or chemoradiation in unresectable cases. Within the limitations of a non-

randomized study, our results suggest that this neoadjuvant approach is at least as effective as cisplatin-based chemoradiation and deserves further evaluation in phase III randomized studies. After many years of studying neoadjuvant chemotherapy its role in the treatment of locally advanced cervical carcinoma remains unclear. Several phase III randomized studies of this form of chemotherapy delivery failed when radiotherapy was used to consolidate the response to induction chemotherapy [12–18]. In fact, two of them not only failed but showed negative effects on the survival of patients assigned to this treatment as compared with radiation alone [15, 17]. These results are on

1217 the whole explained by the cross-resistance between both modalities of treatment delivered in a sequential fashion. The use of surgery after induction chemotherapy, however, has a sound biological basis due to the fact that surgical resection effectively removes residual tumors that otherwise could not be eradicated by radiation after chemotherapy. This hypothesis is supported by the results of several phase II studies of induction chemotherapy, which show that patients who are not able to undergo surgery and who receive radiation treatment have poorer survival rates as compared with those patients who undergo hysterectomy [27–31]. For instance, Fujiyama et al. [27] reported a 5-year disease-free survival of 85.3% and 31.3% for surgically and radiotherapy consolidated patients, respectively, and in the study of Minawa et al. [28], these figures were 72.2% and 25%, respectively. This has led to the suggestion that more effort should be put into operating on the patients with suboptimal responses to neoadjuvant chemotherapy instead of those who show a major response. In contrast to those studies where radiation was used to consolidate treatment, it is remarkable that the four randomized studies comparing neoadjuvant chemotherapy/surgery against radiation alone have been positive [19–22]. The main limitations of these studies were that the comparisons were made relative to radiation alone, which at present is not longer the standard treatment. Even though the length of follow-up in our study was short, the comparable responses and survival shown between the neoadjuvant chemotherapy/surgery group and the group treated with chemoradiation can be explained as follows. First, the use of a new drug combination highly active for advanced, pretreated cervical carcinoma [32, 33] and for untreated patients, which is able to achieve better clinical and pathological responses [23] as compared with older regimens such as cisplatin–vincristine–bleomycin, cisplatin–bleomycin, cisplatin–ifosfamide and many others [34]; as well as compared with newer combinations, such as cisplatin–navelbine [35], cisplatin–irinotecan [36] and cisplatin–ifosfamide–paclitaxel [37]. Secondly, the use of surgical resection, even in tumors that only showed partial responses to chemotherapy, and finally, the use of concomitant chemoradiation for those cases considered unresectable. This last issue seems to be important. The modality of neoadjuvant chemotherapy followed by concomitant chemoradiation has already demonstrated encouraging results in head and neck, esophagus and lung carcinomas [38–40]. To our knowledge, this is the first report on the use of chemoradiation after induction chemotherapy in the treatment of cervical cancer. It is important to mention the fact that the induction chemotherapy/surgery approach requires a highly motivated team of surgeons to attempt surgical resection in cases with no optimal responses to chemotherapy. In a subsequent trial of oxaliplatin– gemcitabine in a comparable population of patients [41], the resection rate was increased to 70% and in an ongoing protocol for comparable cases of locally advanced disease using carboplatin–paclitaxel, our resection rate has so far been

100% (25 cases). Finally, the results of the neoadjuvant approach might be further improved by adding postoperative chemoradiation to resected cases. In this trial, adjuvant chemoradiation was delivered only in those cases with positive surgical margins. Because none of the patients with pathological complete response have so far relapsed, in our subsequent protocols, chemoradiation was added to all cases with any residual disease. With regard to the cisplatin-based chemoradiation arm, our results are comparable, within the limitations of the differences in median follow-up (24 months compared with 35 and 43 months in Morris et al. [3] and Rose et al. [5], respectively). These authors reported an overall survival of 73% and 66/67% for the investigational arms, respectively; however, it should be mentioned that in our study, patients were not surgically staged and therefore an unknown proportion of them could have had positive para-aortic lymph nodes. Thus our chemoradiation arm is a proper control. In addition, our results demonstrate that cisplatin-based chemoradiation can be used on a routine basis as the patients treated were consecutive and not selected. Thus, median number of cisplatin courses administered, dose intensity, dose of radiation and treatment time were comparable to, or even better than, that reported in these trials [2–6]. The widespread use of chemoradiation still faces some resistance from radiotherapists in many regions in the world, mainly those from developing countries. In a recent survey conducted by A. Duenas-Gonzalez (presented at the International Network for Cancer Treatment and Research meeting, Brussels, Belgium, April 2001, unpublished data), revealed that in most Latin American countries, radiation alone is still the most commonly used modality. We have retrospectively compared (unpublished data) our results of chemoradiation with radiation alone which show a highly significant difference in terms of local control and survival for the concomitant approach. In conclusion, this non-randomized comparison of neoadjuvant chemotherapy followed by surgery or chemoradiation with definitive cisplatin-based chemoradiation shows equivalent results in terms of response and survival. A randomized phase III trial is required to establish the value of neoadjuvant chemotherapy/surgery with or without adjuvant chemoradiation versus standard chemoradiation.

References 1. Mohar A, Frias-Mendivil M. Epidemiology of cervical cancer. Cancer Invest 2000; 18: 584–590. 2. Peters WA III, Liu PY, Barrett R et al. Concurrent chemotherapy and pelvic radiation therapy compared with pelvic radiation therapy alone as adjuvant therapy after radical surgery in high-risk early-stage cancer of the cervix. J Clin Oncol 2000; 18: 1606–1613. 3. Morris M, Eifel PJ, Lu J et al. Pelvic radiation with concurrent chemotherapy compared with pelvic and para-aortic radiation for high-risk cervical cancer. New Engl J Med 1999; 340: 1137–1143.

1218 4. Keys HM, Bundy BN, Stehman FB et al. A comparison of weekly cisplatin during radiation therapy versus irradiation alone each followed by adjuvant hysterectomy in bulky stage IB cervical carcinoma: a randomized trial of the Gynecology Oncology Group. New Engl J Med 1999; 340: 1154–1161. 5. Rose PG, Bundy BN, Watkins EB et al. Concurrent cisplatin-based chemoradiation improves progression-free survival in advanced cervical cancer: results of a randomized Gynecologic Oncology Group study. New Engl J Med 1999; 340: 1144–1153. 6. Whitney CW, Sause W, Bundy BN et al. A randomized comparison of fluorouracil plus cisplatin versus hydroxyurea as an adjunct to radiation therapy in stages IIB–IVA carcinoma of the cervix with negative para-aortic lymph nodes. A Gynecologic Oncology Group and Southwest Oncology Group study. J Clin Oncol 1999; 17: 1339– 1348. 7. Wong LC, Ngan HYS, Cheung ANY et al. Chemoradiation and adjuvant chemotherapy in cervical cancer. J Clin Oncol 1999; 17: 2055–2060. 8. Roberts KB, Urdaneta N, Vera R et al. Interim results of a randomized trial of mitomycin C as an adjunct to radical radiotherapy in the treatment of locally advanced squamous-cell carcinoma of the cervix. Int J Cancer 2000; 90: 206–223. 9. Curtin JP, Hoskins WJ, Venkatraman ES et al. Adjuvant chemotherapy versus chemotherapy plus pelvic irradiation for high-risk cervical cancer patients after radical hysterectomy and pelvic lymphadenectomy (RH-PLND): a randomized phase III trial. Gynecol Oncol 1996; 61: 3–10. 10. Tattersall MHN, Ramirez C, Coppleson M. A randomized trial of adjuvant chemotherapy after radical hysterectomy in stage IB–IIA cervical cancer patients with pelvic lymph node metastases. Gynecol Oncol 1992; 46: 176–181. 11. Lahousen M, Haas J, Pickel H et al. Chemotherapy versus radiotherapy versus observation for high-risk cervical carcinoma after radical hysterectomy: a randomized, prospective multicenter trial. Gynecol Oncol 1999; 75: 196–201. 12. Chauvergne J, Rohart J, Heron JF. Randomized phase III trial of neoadjuvant chemotherapy and radiotherapy versus radiotherapy in stage IIB, III carcinoma of the cervix: a cooperative study of the French Oncology Centers. Proc Am Soc Clin Oncol 1988; 7: 136. 13. Tattersall MH, Ramirez C, Coppleson M. A randomized trial comparing platinum-based chemotherapy followed by radiotherapy versus radiotherapy alone in patients with locally advanced cervical cancer. Int J Gynecol Oncol 1992; 2: 244–251. 14. Kumar L, Kaushal R, Nandy M et al. Chemotherapy followed by radiotherapy versus radiotherapy alone in locally advanced cervical cancer: a randomized study. Gynecol Oncol 1994; 54: 307–315. 15. Souhami L, Gil RA, Allan SE et al. A randomized trial of chemotherapy followed by pelvic radiotherapy in stage IIIB carcinoma of the cervix. J Clin Oncol 1991; 9: 970–977. 16. Sundorf K, Trope CG, Hogberg T et al. Radiotherapy and neoadjuvant chemotherapy for cervical carcinoma. A randomized multicenter study of sequential cisplatin and 5-fluorouracil and radiotherapy in advanced cervical carcinoma stage 3B and 4A. Cancer 1996; 77: 2371–2378. 17. Tattersall MNH, Lorvidhaya V, Vootiprux V et al. A randomized trial of epirubicin and cisplatin chemotherapy followed by pelvic radiation in locally advanced cervical cancer. J Clin Oncol 1995; 13: 444–451. 18. Herod J, Burton A, Buxton J et al. A randomised, prospective, phase III clinical trial of primary bleomycin, ifosfamide and cisplatin (BIP)

19.

20.

21.

22.

23.

24.

25. 26.

27.

28.

29.

30.

31.

32.

33.

34. 35.

chemotherapy followed by radiotherapy versus radiotherapy alone in inoperable cancer of the cervix. Ann Oncol 2000; 11: 1175–1181. Sardi J, Giaroli A, Sananes C et al. Long-term follow-up of the first randomized trial using neoadjuvant chemotherapy in stage IB squamous carcinoma of the cervix: the final results. Gynecol Oncol 1997; 67: 61–69. Sardi J, Sananes C, Giaroli A et al. Neoadjuvant chemotherapy in cervical carcinoma stage IIB: a randomized controlled trial. Int J Gynecol Cancer 1998; 8: 441–450. Sardi J, Giaroli A, Sananes C et al. Randomized trial with neoadjuvant chemotherapy in stage IIIB squamous carcinoma of the cervix uteri: an unexpected therapeutic management. Int J Gynecol Cancer 1996; 6: 85–93. 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 2002; 20: 179–188. Duenas-Gonzalez A, Lopez-Graniel C, Gonzalez A et al. A phase II study of gemcitabine and cisplatin combination as induction chemotherapy for untreated locally advanced cervical carcinoma. Ann Oncol 2001; 12: 541–547. Lopez-Graniel C, Reyes M, Chanona G et al. Type III radical hysterectomy after induction chemotherapy for patients with locally advanced cervical carcinoma. Int J Gynecol Cancer 2001; 11: 210– 217. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958; 53: 457–481. Mantel M. Evaluation of survival data and two rank order statistics arising in its consideration. Cancer Chemother Rep 1966; 50: 163– 170. Sugiyama T, Nishida T, Muraoka Y et al. Radical surgery after neoadjuvant intra-arterial chemotherapy in stage IIIB squamous cell carcinoma of the cervix. Int Surg 1999; 84: 67–73. Minagawa Y, Kigawa J, Irie T et al. Radical surgery following neoadjuvant chemotherapy for patients with stage IIIB cervical cancer. Ann Surg Oncol 1998; 5: 539–543. Sugiyama T, Nishida T, Hasuo Y et al. Neoadjuvant intraarterial chemotherapy followed by radical hysterectomy and/or radiotherapy for locally advanced cervical cancer. Gynecol Oncol 1998; 69: 130– 136. Kigawa J, Minagawa Y, Ihshihara H et al. The role of neoadjuvant intraarterial infusion chemotherapy with cisplatin and bleomycin for locally advanced cervical cancer. Am J Clin Oncol 1996; 19: 255– 259. Sardi J, Sananes C, Giaroli A et al. Neoadjuvant chemotherapy in locally advanced carcinoma of the cervix uteri. Gynecol Oncol 1990; 38: 486–493. Burnett AF, Roman LD, Garcia AA et al. A phase II study of gemcitabine and cisplatin in patients with advanced, persistent, or recurrent squamous cell carcinoma of the cervix. Gynecol Oncol 2000; 76: 63– 66. Duenas-Gonzalez A, Hinojosa-Garcia LM, Lopez-Graniel C et al. Weekly cisplatin/low dose gemcitabine combination for advanced and recurrent cervical carcinoma. Am J Clin Oncol 2001; 24: 201– 203. Eddy GL. Neoadjuvant chemotherapy before surgery in cervical cancer. Monogr NCI 1996; 21: 93–96. Pignata S, Silvestro G, Ferrari E et al. Phase II study of cisplatin and vinorelbine as first-line chemotherapy in patients with carcinoma of the uterine cervix. J Clin Oncol 1999; 17: 756–760.

1219 36. Sugiyama T, Nishida T, Kumagai S et al. Combination therapy with irinotecan and cisplatin as neoadjuvant chemotherapy in locally advanced cervical cancer. Br J Cancer 1999; 81: 95–98. 37. Zanetta G, Lissoni A, Pellegrino A et al. Neoadjuvant chemotherapy with cisplatin, ifosfamide and paclitaxel for locally advanced squamous-cell cervical cancer. Ann Oncol 1998; 9: 977–980. 38. Kies MS, Haraf DJ, Athanasiadis I et al. Induction chemotherapy followed by concurrent chemoradiation for advanced head and neck cancer: improved disease control and survival. J Clin Oncol 1998; 16: 2715–2721.

39. Ampil FL, Mills GM, Burton GV. Induction chemotherapy followed by concomitant chemoradiation for cervical esophageal cancer. Am J Gastroenterol 1999; 94: 2325–2326. 40. Ardizzoni A, Grossi F, Scolaro T et al. Induction chemotherapy followed by concurrent standard radiotherapy and daily-dose cisplatin in locally advanced non-small-cell lung cancer. Br J Cancer 1999; 81: 310–315. 41. Dueñas-Gonzalez A, Lopez-Graniel C, Gonzalez AE et al. Induction chemotherapy with gemcitabine and oxaliplatin for locally advanced cervical carcinoma. J Clin Oncol 2002; In press.