Enhanced Radiosensitization with Interferon-α-2b and Cisplatin in the Treatment of Locally Advanced Cervical Carcinoma

GYNECOLOGIC ONCOLOGY ARTICLE NO.

67, 309 –315 (1997)

GO974879

Enhanced Radiosensitization with Interferon-a-2b and Cisplatin in the Treatment of Locally Advanced Cervical Carcinoma Richard G. Stock, M.D.,* Peter Dottino, M.D.,† T. Scott Jennings, M.D.,† Mitchell Terk, M.D.,* J. Keith DeWyngaert, Ph.D.,* Ann Marie Beddoe, M.D.,† and Carmel Cohen, M.D.† *Department of Radiation Oncology and †Department of Gynecologic Oncology, Mount Sinai School of Medicine, New York, New York 10029 Received January 21, 1997

Purpose. To evaluate the efficacy and toxicity of interferon-a-2b (IFN-a) and cisplatin given concomitantly with radiation therapy (RT) in the treatment of locally advanced cervical carcinoma. Materials and Methods. Twenty-one patients with stage bulky Ib–IIIb (Ib, 2; IIa, 2; IIb, 8; IIIb, 9) cervical carcinoma were treated with combined IFN-a (5 million IU) subcutaneously three times per week and cisplatin (25 mg/m2) iv infusion over 2 h weekly for 7 weeks, given concomitantly with RT (4500 cGy of external beam plus 2 brachytherapy procedures). Total radiation doses delivered ranged from 7500 to 9960 cGy (median, 9300 cGy). Follow-up ranged from 16 to 33 months (median, 25 months). Results. The 2-year local control rate was 100%. The only sites of disease recurrence were distant. Freedom from distant metastases, disease-free survival, and overall survival at 2 years was 76%. Late complication rates were high. Grade 4 rectosigmoid, bladder, and small bowel complication rates were 49, 18, and 23% at 2 years. Late toxicity was seen earlier than expected with rectosigmoid complications observed 5 to 11.5 months (median, 8 months) after completion of treatment. Conclusion. Combination IFN-a and cisplatin produced a marked effect of enhanced radiosensitization as evidenced by 100% local tumor control and high late normal tissue complication rates. Due to the unacceptable late toxicity, its routine clinical use cannot be recommended. Further investigation is needed to determine whether a therapeutic window exists such that the use of lower doses of IFN-a, cisplatin, or RT can increase tumor control with more acceptable normal tissue toxicity. © 1997 Academic Press

vanced cervical cancer is important and improvements in this endpoint of therapy can translate into increased survival [2, 3]. Attempts at improving local control and distant control for this subset of patients with the use of chemotherapy in conjunction with irradiation have produced mixed results. This led investigators to examine other therapeutic agents. Biologic response modifiers such as interleukins, interferons (IFN), and cytokines have recently been used in the treatment of human malignancies. Recombinant interferon-a has been shown to demonstrate activity in both in vivo models and human neoplasms [4, 5]. It has also been shown to enhance the activity of both chemotherapy and RT [6]. Specifically, IFN has been shown to augment the activity of cisplatin in human mesothelioma xenografts in vivo and in human clinical trials of advanced non-small cell lung cancer [4, 7]. IFN has been shown to act as a radiosensitizer with significant potentiation of cytoxic activity in cervical cell lines [8]. In addition, there is evidence that IFN may provide protection from the damaging effects of radiation [9, 10]. These unique properties of IFN led us to design a phase I–II trial to test the efficacy and safety of interferon-a-2b (IFN-a) in combination with cisplatin and RT in the treatment of locally advanced cervical carcinoma. This report details the tumor control and late toxicity following this regimen. MATERIALS AND METHODS Patient Eligibility

INTRODUCTION There has been a marked decrease in cancer-related deaths from cervical cancer. This has primarily been due to screening programs which have led to a decreased incidence of invasive disease. Despite this success, the cure rate for women presenting with locally advanced disease has remained poor with contemporary therapy offering less than a 50% survival rate. The standard treatment for locally advanced cervical carcinoma has been radiation therapy (RT). The poor results with irradiation alone have partially been due to the inability of this modality to locally control disease [1]. Local control in ad-

All patients required a histologic diagnosis of carcinoma of the uterine cervix. Eligibility criteria were as follows: International Federation of Gynecology and Obstetrics (FIGO) bulky stage Ib or IIa with cervical diameter greater than 6 cm or stage IIb, IIIa, IIIb, or IVa or the presence of lymphatic metastases at pretreatment staging, Karnofsky performance status of at least 70% and an anticipated survival of at least 6 months, adequate hematologic function with ANC . 2000/mm3 and platelet count .150,000/mm3, adequate renal function with serum creatinine ,2.5 mg/ml, or creatinine clearance .50 ml/min, bilirubin levels ,2.1 mg/ml, and serum transaminases/alkaline

309

0090-8258/97 $25.00 Copyright © 1997 by Academic Press All rights of reproduction in any form reserved.

310

STOCK ET AL.

phophatases less than or equal to 2.5 times normal. Patients could not have a history of treatment with chemotherapy or radiation therapy less than or equal to 2 years prior to diagnosis of cervical cancer. The placement of uretheral stents prior to therapy was allowed. All patients were tested negative for the human immunodeficiency virus. Informed consent was obtained for all patients. Patient and Tumor Characteristics There were 21 patients entered into the protocol whose ages ranged from 25 to 72 with a median of 46 years. There were 7 white, 6 black, 6 Hispanic, and 2 Asian patients. The International Federation of Gynecology and Obstetrics (FIGO) stages included Ib in 2 patients, IIa in 2 patients, IIb in 8 patients, and IIIb in 9 patients. Histologies included squamous cell carcinoma in 17 patients and adenosquamous in 4 patients. The tumors were graded as poorly differentiated in 12 patients, moderately differentiated in 6 patients, and well differentiated in 3 patients. Staging All patients underwent an examination under anesthesia. Nineteen patients underwent laparoscopic paraaortic lymph node sampling. Two of these patients were found to have metastatic paraaortic lymph nodes. Fifteen of these patients also underwent laparoscopic pelvic lymph node dissections. One of these patients had positive pelvic nodes. This patient also had paraaortic metastases. All patients had negative chest X-rays. Computed tomography of the abdomen and pelvis was performed in all patients prior to therapy. Treatment All patients started therapy within 1 week of staging. Chemotherapy, immunotherapy, and radiation therapy started concomitantly. IFN-a was given at 5 million units subcutaneously, 3 times per week for 7 weeks throughout radiation therapy. IFN-a was given during external beam therapy (5 weeks) and during the weeks of the 2 brachytherapy procedures. Cisplatin was given at 25 mg/m2 iv over 2 h weekly for a total of 7 courses. Cisplatin was given with both external irradiation and brachytherapy. One patient who received only one brachytherapy procedure received only 6 weeks of cisplatin and IFN-a. External irradiation was delivered to a dose of 4500 cGy using daily 180 cGy fractions given at 5 fractions per week. All patients were treated using 15-MV X rays. Sixteen patients were treated to a pelvic field using a four-field (anterior/ posterior and 2 lateral) technique. Two patients were treated to a pelvic–inguinal field using an anterior/posterior technique. One patient was treated to a pelvic–inguinal field using a four-field technique with concomitant electron boost to the inguinal nodes. The indication for treating the inguinal region was tumor involvement of the lower one-third of the vaginal

TABLE 1 Total Brachytherapy Doses Delivered

Prescribed dose (cGy) Milligram hours (mg Ra eq 3 h) Point A dose (cGy) Tolerance points (brachytherapy 1 external beam) Rectal maximum (cGy) Bladder maximum (cGy) Sigmoid (cGy)

Range

Median

3000–5460 2800–8129 3087–7048

4800 5192 4771

5710–8776 5701–8305 5690–8768

7186 7542 6836

canal. Two patients were treated to a pelvic–paraaortic field using a four-field technique. Both of these patients had documented paraaortic nodal metastases. Prophylactic irradiation of the paraaortic nodes in patients without pathologic evidence of paraaortic adenopathy was not given. Brachytherapy was delivered using either the Henschke tandem and ovoid applicator or the Syed–Neblet interstitial applicator. The Henschke applicator was used in 18 patients and the Syed–Neblet applicator was used in 3 patients. The decision to use an interstitial implant was made when the patient’s anatomy and/or tumor bulk created a situation in which a tandem and ovoid application would not have provided adequate dose coverage of the tumor volume. One patient was treated with the Henschke applicator for one application and the Syed–Neblet applicator for the second application. Two applications were used for 20 patients. The first implant application was given 2 weeks after the completion of external beam irradiation and the second application was given 2 weeks after the first. One patient received only one application due to the discovery of metastatic disease following the first implant. The brachytherapy dose was prescribed to the isodose line which was felt to cover the tumor volume. The prescribed dose rates ranged from 50 to 60 cGy per hour (median, 60 cGy). There was no standard implant dose. The doses given via the implants were customized to each patient, with consideration to the amount of disease present, the implant geometry, and the resulting proximity of the bladder and rectum. The total milligram hours (milligram radium equivalent 3 duration of implant in hours), prescribed doses, and doses to point A for the brachytherapy procedures are listed in Table 1. In addition, the maximum doses delivered to bladder, rectum, and sigmoid were calculated for each implant. In tandem and ovoids implants, the rectal maximum point was calculated as the point along a line drawn through the center of the ovoids, perpendicular to the horizontal axis of the lateral X-ray film, at the anterior rectal mucosal surface as visualized by barium injected into the rectum (see Fig. 1). This point was positioned to lie directly beneath the tandem on the anterior film. In an interstitial implant, the rectal point was defined as a point on the anterior rectal mucosa,

ENHANCED RADIOSENSITIZATION WITH IFN-a AND CISPLATIN

311

FIG. 1. Lateral X-ray film showing placement of tandem and ovoid applicator with tolerance points. B, maximum bladder point; R, maximum rectal point; S, sigmoid point.

as defined by barium, on a lateral film which lay closest to the needles. This point was located in the center of the implant as defined from left to right on the anterior film. The bladder maximum point was chosen as the point on the most posterior aspect of the foley balloon which lay in the center of the balloon from left to right and from superior to inferior (see Fig. 1). In addition, a new point was introduced to approximate the maximal sigmoid dose in tandem and ovoid implants. The sigmoid point was defined as the point midway along a line drawn perpendicular to the tandem constructed from a point located half the distance between tandem tip and flange connecting the tandem to the sacrum (see Fig. 1). The point was located directly beneath the tandem on an anterior film. Sigmoid points were not calculated for interstitial implants. The cumulative doses delivered to the rectal and bladder maximum points as well as the sigmoid points are listed in Table 1. Local failure was defined as recurrence within the irradiated field. Distant failure was considered disease recurrence outside of the irradiated field. Actuarial survival, failure, and complication rates were calculated using the methods of Kaplan and Meier [11]. Differences in actuarial rates were calculated using the log-rank test [12, 13]. Acute and late complications were

graded using the Radiation Therapy and Oncology Group (RTOG) scale [14].

RESULTS The follow-up ranged from 16 to 33 months (median, 25 months) for patients free from disease recurrence. The local control at 2 years was 100%. All patients had core needle biopsy samples taken of the cervix and parametria at the time of the second brachytherapy procedure. The biopsies were negative in all but three patients. These three patients all had subsequent negative biopsies taken after completion of therapy. The freedom from distant metastasis rate at 2 years was 76% (see Fig. 2). Five patients developed distant metastases and all died of their disease. Distant metastatic sites included the following: abdominal carcinomatosis, 2 patients; bone, 2 patients; liver, 2 patients; lung, 2 patients; adrenal, 1 patient; heart, 1 patient; colon, 1 patient. None of these patients showed evidence of local failure. Patients failing included the two with documented paraaortic metastases at presentation. The stage, histology, grade, and paraaortic and pelvic nodal status of the five patients who developed distant metastases are listed in

312

STOCK ET AL.

FIG. 2.

Freedom from distant metastases.

Table 2. The actuarial 2-year overall and disease-free survival rates were 76%. Acute Toxicity The regimen was well tolerated. There was no hematologic or renal toxicity noted during the administration of the therapy. All patients were classified as having grade 0 hematologic and renal toxicity. During the course of external beam irradiation, acute gastrointestinal toxicity was observed in the following: grade 0, 3 patients (14%); grade 1, 10 patients (48%); grade 2, 8 patients (38%). Acute genitourinary symptoms were observed as follows: grade 0, 9 patients (43%); grade 1, 11 patients (52%); and grade 2, 1 patient (5%). Overall treatment time ranged from 64 to 93 days (median, 72 days). Two patients required 1-week breaks during pelvic RT. This was due to moist skin desquamation in one patient and diarrhea and vomiting in the second. One patient required a 2-week break secondary to moist skin desquamation. Late Complications Rectosigmoid toxicity was the most common late complication. Nine patients developed grade 4 toxicity. These complications occurred earlier than expected. Rectosigmoid compli-

cations occurred from 5 to 11.5 months from the completion of treatment with a median time of 8 months. The actuarial freedom from grade 4 rectosigmoid toxicity was 51% at 2 years (see Fig. 2). Toxicity primarily involved the upper rectum and sigmoid. The distal extent of damage, as measured from the anal verge on colonoscopy, ranged from 7 to 20 cm, with a median of 12 cm. Three patients required a colon resection and temporary colostomy, followed by reanastamosis. Six patients were treated with a permanent colostomy with or without resection of the damaged area. Three patients developed grade 2 rectosigmoid complications which were conservatively managed. There were no grade 3 complications. Acute gastrointestinal toxicity predicted for late complications. Freedom from grade 4 late rectosigmoid toxicity was 71% in 13 patients who experienced grade 0 –1 acute morbidity versus 25% in 8 patients with grade 2 acute morbidity (P 5 0.01). In order to determine if this damage was related to radiation dose, a dose–response analysis was performed. While there was no clear statistically significant dose response, patients receiving a dose #7000 cGy to the rectal maximum point (9) had a freedom from developing grade 4 rectosigmoid toxicity of 73% at 2 years compared to a rate of 40% for those patients receiving a dose .7000 cGy (12) (P 5 0.2). A subset of these 9 patients included 5 patients who developed rectovaginal fistulas. These fistulas developed from 7 to 19 months (median, 15.5 months) after completion of treatment. The freedom from developing fistula formation at 2 years was 70%. Grade 4 small bowel damage occurred in 4 patients. These complications required small bowel resection. The actuarial freedom from developing grade 4 small bowel damage at 2 years was 77%. This toxicity was observed from 4 to 15 months (median, 12 months) after completion of treatment. All of the small bowel damage was seen in patients who also developed grade 4 rectosigmoid toxicity. Grade 4 late bladder complications were the next most common toxicity seen. Three patients developed this toxicity for an actuarial freedom from grade 4 bladder complications at 2 years of 82%. These complications occurred at 9, 13, and 17.5 months after completion of treatment. All three patients required a total cystectomy. There was no evidence of grade

TABLE 2 Disease Profile of Patients Developing Distant Metastases Patient

Stage

Histology

Grade

Paraaortic nodal status

Pelvic nodal status

1 2 3 4 5

IIb IIb IIIb IIIb IIIb

AS SCCA SCCA SCCA AS

P P M W P

Positive Negative Positive Negative Negative

Positive Negative NP Negative Negative

Note. AS, adenosquamous carcinoma; SCCA, squamous cell carcinoma; P, poorly differentiated; M, moderately differentiated; W, well differentiated; NP, pelvic node dissection not performed.

ENHANCED RADIOSENSITIZATION WITH IFN-a AND CISPLATIN

313

TABLE 3 Bladder and Rectum Maximum Doses for Patients Developing Grade 4 Toxicity

Patient

Grade 4 bladder toxicity

Grade 4 rectal toxicity

Total tumor dose

Rectal maximum

Bladder maximum

Total sigmoid

1 2 3 4 5 6 7 8 9

N N Y N N N Y N Y

Y Y Y Y Y Y Y Y Y

9300 8940 8920 9325 9580 9300 9780 9100 9060

7520 6111 6279 7186 8434 7376 7421 7398 8776

8305 8045 7640 6632 7571 7623 7351 7164 7748

6836 6602 6691 7345 NM 6848 NM NM NM

Note. Doses given in cGy. N, no toxicity; Y, toxicity seen. NM, sigmoid point not measured.

2–3 toxicity. In addition, all of these patients also developed rectosigmoid complications. The maximum doses delivered to the rectal, sigmoid, and bladder points for the 9 patients who developed grade 4 toxicity can be found in Table 3. DISCUSSION Poor results following standard therapy for locally advanced cervical carcinoma have led investigators to search for new treatment strategies. The first approaches involved the addition of chemotherapy to RT. Agents such as cisplatin and hydroxyurea have been used in the neoadjuvant and adjuvant setting [15, 16]. Unfortunately, results following combination chemotherapy and RT have been mixed. These findings led us to explore the use of biologic response modifiers. IFNs are a family of cytokines which possess antiproliferative, antiviral, antineoplastic, and immunoregulatory properties [17]. The cytotoxic effects of RT on cell lines have been increased with the addition of IFN. There are currently two postulated mechanisms to explain this enhanced radiosensitization. Chang investigated the effect of IFN on radiation cytotoxicity of human hypernephroma cell lines. He found that IFN and radiation had an additive effect on cell growth inhibition. In addition, radiation killing was increased by IFN pretreatment. When the distribution of cells in the cell cycle was measured, there was a significant increase in accumulation of cells at the G2–M phase of the cell cycle. This blockage at G2–M, the most radiosensitive phase of the cell cycle, was postulated to be the cause of the potentiation of radiation injury by IFN [18]. The other postulated mechanism was derived from findings that IFN caused increased radiation cell kill by reducing the shoulder of the cell survival curve. A reduction in the shoulder of the cell survival curve is believed to be due to a decreased ability of cells to repair sublethal damage caused by radiation. This led to the theory that IFN inhibited cells from accumulating sublethal damage. In order to further explore this concept, split dose experiments were performed to test whether

IFN could increase cell kill by interfering with sublethal damage repair between doses of radiation. These studies failed to demonstrate an ability of IFN to impair sublethal damage repair [19, 20]. Due to these conflicting data, the exact mechanism of IFN enhanced radiotoxicity is currently not known. Another potential benefit of combining IFN with RT was the ability of IFN to act as a radioprotector. This was suggested by data which showed that IFN could enhance postradiation therapy immunoregulatory function. Observations revealed that radiation therapy caused a decrease in the activity of natural killer cells and lymphokine-activated cells. IFN was found to cause an accelerated rebound of this depressed lymphocyte activity [10]. This suggests that IFN has a protective effect on normal cells against radiation damage. Others investigators have demonstrated that purified mouse IFN could increase the life span of nude mice after lethal doses of radiation [9]. In a study which used IFN-a-2a and cis-retinoic acid prior to RT, IFN produced a protective effect on irradiated bowel in a mouse model [21]. IFN has also been shown to affect the antineoplastic activities of chemotherapeutic agents. Increased cytoxic effects of cisplatin on HeLa cervical cell lines have been shown with the addition of IFN [6]. This effect has also been noted in the clinical setting in the treatment of non-small cell lung cancer and penile carcinoma [7, 22]. Evidence that IFN could increase radiation cytotoxicity specifically in cervical cancer was the basis behind the design of this current trial. Angioli demonstrated increased radiation cell kill with IFN in in vitro experiments of cervical cell lines [9]. Kavanagh found that the combination of 13-cis-retinoic acid and IFN was active in increasing radiation response in a phase I clinical trial of patients with locally advanced cervical carcinoma [23]. The goal of this current trial was to determine if increased cytotoxic effects could be obtained with the addition of cisplatin and IFN-a to RT while maintaining acceptable toxicity. Total radiation dose has been directly correlated to tumor

314

STOCK ET AL.

FIG. 3. Freedom from grade 4 rectosigmoid complications.

control in the treatment of cervical cancer [24, 25]. For this reason, this trial did not attempt to reduce the total dose administered, despite the addition of radiation sensitizers. The dosage of IFN chosen for this trial has previously been found to be well tolerated in combination with RT [26]. The addition of IFN and cisplatin to RT had a marked effect on both local control and treatment-related toxicity. The increase in cytotoxicity with the addition of IFN-a and cisplatin to RT was manifested by a local control rate of 100%. It is difficult to make comparisons between the antitumor activity seen in this trial and results reported with more standard therapy since this protocol was a phase I–II trial and not a phase III trial, but these findings compare favorably to pelvic failure rates of 21–34 and 36 –39% reported for stage IIb and IIIb disease, respectively, treated with RT alone [27–29]. The improved local control, disease-free, and overall survival of this regimen was partly offset by the development of severe late complications. The 2-year actuarial rates of developing grade 4 rectosigmoid, bladder, and small bowel complications were 49, 18, and 23%, respectively. Of the 16 patients alive and free of disease, 8 remain free of grade 3– 4 toxicity. Not only did this regimen increase the incidence of late complications, but it also appeared to shift the time course for their development. Late complications occurred earlier than expected. Rectosigmoid toxicity occurred from 5 months to 1 year after completion of treatment. This complication rate appeared to plateau after 1 year (see Fig. 3). In a study of late complications following RT for early-stage cervical cancer, Eifel found that the greatest risk of developing major rectal toxicity was during the first 2 years of follow-up with a marked reduction in risk after 2 years [30]. The enhanced radiosensitization observed was most likely due to the combination of the IFN, cisplatin, and radiation and not due to any one or two agents alone or in combination. Cisplatin has been used in combination with RT in multiple trials and has not been found to increase the incidence of

radiation damage to normal tissues [31–33]. In a previously published trial from our own institution, the addition of cisplatin to radiation did not increase toxicity rates [15]. It is also unlikely that IFN and radiation therapy alone were responsible for the effects observed. In a trial using RT, IFN-a-2a, and isotretinoin to treat stage IIb and III cervical cancer, no increase in toxicity was noted with a 3–5% grade 3 complication rate [34]. The dose of radiation therapy used in the current trial was also unlikely to be the sole cause for the late complications seen. Eifel, in a study of stage Ib to IIb bulky cervical carcinomas, recommended total brachytherapy doses $6000 mghr [24]. In our study, which not only included bulky stage Ib to IIb (12 patients) but also included 9 patients with stage IIIb disease, the milligram hours of brachytherapy used were less and ranged from 2800 to 8129, with a median value of 5192. In addition, these complications were unlikely due to excessive radiation dose alone delivered to the bladder, rectum, or sigmoid. The median rectal, sigmoid, and bladder maximum doses were 7186, 6836, and 7542 cGy, respectively. Perez found that major rectal and bladder complications occurred with rates less than 5% when doses in the range of 7000 – 8000 cGy were given. He found a definite increase in late toxicity with doses over 8000 cGy [25]. Similarly, Pourquier found that rectal complications increased sharply when rectal maximal doses exceeded 8000 cGy [35]. In a study by Roeske, the dose calculated to cause a 50% rectal complication rate was 9230 cGy [36]. This maximum rectal dose is much greater than the dose received by any patient in our study and yet the rectosigmoid complication rates were similar. Our data suggest that there may be a radiation dose–response relationship between total dose delivered to normal tissues and the development of late complications. Patients who received doses of #7000 cGy to the maximum rectal point had a freedom from late rectosigmoid complication rate of 73% versus only 40% for patients receiving .7000 cGy to the rectal maximum point. The trial of IFN, isotretinoin, and radiation therapy reported by Antonadou which used a lower total tumor dose (79 Gy) and did not find an excessive late complication rate supports this observation [34]. The data from this trial demonstrate that IFN-a in combination with cisplatin enhances radiation cytoxic effects on both cervical carcinoma and normal tissues. Whether IFN-a and cisplatin increase radiation damage to malignant and normal tissues equally is not known. If IFN-a and cisplatin sensitize both malignant and normal tissues equally to the effects of RT, then the combination would be of limited clinical benefit when used concomitantly with RT. If this combination is more toxic to cervical cancer cells than to normal tissues, a therapeutic window may exist such that the use of lower doses of IFN-a, cisplatin, or RT can increase tumor control with more acceptable normal tissue toxicity. Based on the findings of this trial, we must caution against the routine clinical use of combination IFN-a, cisplatin, and RT. Further investigation in cell and animal models needs to be undertaken to determine if this

ENHANCED RADIOSENSITIZATION WITH IFN-a AND CISPLATIN

therapeutic window exists. In addition, further evaluation of IFN-a and RT interactions is warranted, with particular attention to modulation of both RT and IFN-a doses. REFERENCES 1. Perez C, Camel H, Kuske R, et al: Radiation therapy alone in the treatment of carcinoma of the uterine cervix: a 20-year experience. Gynecol Oncol 23:127–140, 1986 2. Jampolis S, Andra J, Fletcher G: Analysis of sites and cause of failure of irradiation in an invasive squamous cell carcinoma of the intact uterine cervix. Radiology 115:681– 685, 1975 3. Stock RG, Chen ASJ, Karasak C: Natural history and patterns of failure in node positive cervical cancer. Cancer J Sci Am 2:256 –262, 1996 4. Carmichael J, Fergusson RJ, Wolf R, et al: Augmentation of cytotoxicity of chemotherapy by human alpha interferons in human non-small cell lung cancer xenografts. Cancer Res 46:4916 – 4920, 1986 5. Quesade JR, Reuben F, Manning JT, et al: Alpha interferon for induction of remission in hairy cell leukemia. N Eng J Med 310:15–18, 1984 6. Namba M, Yamamoto S, Tanaka H, et al: In vitro and in vivo studies on potentiation of cytotoxic effects of anticancer drugs or cobalt 60 gamma ray by interferon on human neoplastic cells. Cancer 54:2262–2267, 1984 7. Bowman A, Fergusson R, Allan S, et al: Potentiation of cisplatin by alpha-interferon in advanced non-small cell lung cancer (NSCLC): a phase II study. Ann Oncol 1:351–353, 1990 8. Angioli R, Bernd-Uwe S, Perras J, et al: In vitro potentiation of radiation cytotoxicity by recombinant interferons in cervical cancer cell lines. Cancer 71:3717–3725, 1993 9. Ortaldo J, McCoy J: Protective effects of interferon in mice previously exposed to lethal irradiation. Radiat Res 81:262–266, 1980 10. Pillai M, Balaram T, Padmanabhan T, et al: Interleukin 2 and alpha interferon induced in vitro modulation of spontaneous cell mediated cytotoxicity in patients with cancer of the uterine cervix undergoing radiotherapy. Acta Oncol 28:39 – 44, 1989 11. Kaplan EL, Meier P: Non parametric estimation from incomplete observations. J Am Stat Assoc 53:457– 481, 1958 12. Mantel N: Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chemother Rep 50:163–170, 1966 13. Cox DR, Oakes D: Analysis of Survival Data. London/New York, Chapman and Hall, 1984 14. Winchester DP, Cox JD: Standards for breast conservation treatment. Ca Center J Clin 2:134 –162, 1992 15. Lipsztein R, Kredentser D, Dottino P, et al: Combined chemotherapy and radiation therapy for advanced carcinoma of the cervix. Am J Clin Oncol 10:527–530, 1987 16. Hreshchyshyn M, Aron B, Boronow R, et al: Hydroxyurea or placebo combined with radiation to treat stages IIIB and IV cervical cancer confined to the pelvis. Int J Radiat Oncol Biol Phys 5:317–322, 1979 17. Taylor-Papadimitrou J, Balkwill R F S: Implications of new developments in interferon research. Biochem Biophys Acta 695:49 – 67, 1982 18. Chang A, Keng P: Potentiation of radiation cytotoxicity by recombinant interferons, a phenomenon associated with increased blockage at the G2–M phase of the cell cycle. Cancer Res 47:4338 – 4341, 1987 19. Dritschilo A, Mossman K, Gray M, et al: Potentiation of radiation injury by interferon. Am J Clin Oncol (CCT) 5:79 – 82, 1982

315

20. Kardamakis D, Gillies N, Souhami R, et al: Recombinant human interferon alpha-2b enhances the radiosensitivity of small cell lung cancer in vitro. Anticancer Res 9:1041–1044, 1989 21. Mason KA, Tofilon PJ: Unexpected radiation protection with 13-cisretinoic acid plus interferon alpha 2a. Cancer Chemother Pharmacol 33:435– 437, 1993 22. Mitropoulos D, Dimopolous M, Kiroudi-Vouldgari A, et al: Neoadjuvant cisplatin and interferon alpha2B in the treatment and organ preservation of penile carcinoma. J Urol 152:1124 –1126, 1994 23. Kavanagh J, Lippman S, Paredes-Espinoza M, et al: The combination of 13-cis-retinoic acid and interferon alpha 2a with radiation therapy in squamous cell carcinoma of the cervix. Int J Gynecol Cancer 6:439 – 444, 1996 24. Eifel PJ, Thoms WW, Smith TL, et al: The relationship between brachytherapy dose and outcome in patients with bulky endocervical tumors treated with radiation alone. Int J Radiat Oncol Biol Phys 28:113–118, 1993 25. Perez CA, Breaux S, Madoc-Jones H, et al: Correlation between radiation dose and tumor recurrence and complications in carcinoma of the uterine cervix: stages I and IIa. Int J Radiat Oncol Biol Phys 5:373–382, 1979 26. Torrisi J, Berg C, Harter K, et al: Phase I combined modality clinical trial of alpha-2-interferon and radiotherapy. Int J Radiat Oncol Biol Phys 12:1453–1456, 1986 27. Eifel P, Morris M, Oswald M, et al: Adenocarcinoma of the uterine cervix. Cancer 65:2507–2514, 1990 28. Marcial VA: Carcinoma of the cervix: present status and future. Cancer 39:169 –174, 1974 29. Perez CA, Kuske RR, Camel HM, et al: Analysis of pelvic tumor control and impact on survival in carcinoma of the uterine cervix treated with radiation therapy alone. Int J Radiat Oncol Biol Phys 14:613– 621, 1988 30. Eifel PJ, Levenback C, Wharton JT, et al: Time course course and incidence of late complications in patients treated with radiation therapy for FIGO stage Ib carcinoma of the uterine cervix. Int J Radiat Oncol Biol Phys 32:1289 –1300, 1995 31. Malfetano J, Keys H, Kredenster D, et al: Weekly cisplatin and radical radiation therapy for advanced, recurrent, and poor prognosis cervical carcinoma. Cancer 71:3703–3706, 1993 32. Park TK, Lee SK, Kim SN, et al: Combined chemotherapy and radiation for bulky stages I–II cervical cancer: comparison of concurrent and sequential regimen. Gynecol Oncol 50:196 –201, 1993 33. Pearcey R, Stuart G, MacLean G, et al: Phase II study to evaluate the toxicity and efficacy of concurrent cisplatin and radiation therapy in the treatment of patients with locally advanced squamous cell carcinoma of the cervix. Gynecol Oncol 58:34 – 41, 1995 34. Antonadou D, Cardamakis E, Iliopoulos, et al: Comparative study between exclusive irradiation or combined IFN alpha 2a and isotretinoin in stage IIb and III cervical carcinoma. Int J Radiat Oncol Biol Phys 36 (suppl):219, 1996 35. Pouquier H, Dubois JB, Delard R: Cancer of the uterine cervix: dosimetric guidlines for prevention of late rectal and rectosigmoid complications as a result of radiotherapeutic treatment. Int J Radiat Oncol Biol Phys 8:1887– 1895, 1982 36. Roeske JC, Mundt A, Halpern H, et al: Risk of late bladder and rectal complications following definitive radiation therapy for carcinoma of the cervix: a dosimetric analysis. Int J Radiat Oncol Biol Phys 32:(supp) 68, 1995