Surgery, chemotherapy and whole abdominal radiotherapy in the management of advanced ovarian carcinoma

Clinical Radiology (1987)38, 269-272

Surgery, Chemotherapy and Whole Abdominal Radiotherapy in the Management of Advanced Ovarian Carcinoma G. J. S. RUSTIN, M. MINTON*, B A R B A R A SOUTHCOTT*, M. GLASER*, E. S. NEWLANDS, R. H. J. BEGENT and K. D. B A G S H A W E

Cancer Research Campaign Laboratories Department of Medical Oncology and *Department of Radiotherapy, Charing Cross Hospital, London

A pilot study on 35 women was performed to determine the feasibility of giving whole abdominal radiotherapy after surgery and chemotherapy for Stage III and IV ovarian adenocarcinoma. The planned duration of therapy was only 6 months from diagnosis. Chemotherapy consisted of four courses of cis-platinum 100 m g / m 2 alternating every 10-12 days with four courses of methotrexate 300 m g / m 2 and cyclophosphamide 500 mg/m 2 followed by folinic acid rescue. Twenty-nine of 35 (83 %) patients were shown at second look surgery after chemotherapy to have residual tumour deposits no more than 2 cm in diameter. The usual radiotherapy dose was 2400 cGy in 20 fractions over approximately 4 to 5 weeks, with kidney shielding, and was generally well tolerated. Three patients failed to complete radiotherapy, two because of progressive disease and one because of persistent nausea and vomiting. During radiotherapy the white blood count fell below 2.0 × 109/litre in seven patients (26%), and in eight patients (30%) the platelet count fell below 75 × 109/litre. However, the white blood count nadir of 1.5 x 109/litre was reached by 1300 cGy and the platelet nadir of 45 × 109/litre was reached by 1400 cGy and both then levelled or recovered despite continuing radiotherapy. The median follow up since diagnosis is 26 months. Four of 12 patients with no tumour detected at second look operation have relapsed compared with 13 of 15 patients who had detectable tumour. Toxicity of this multi-modality therapy was acceptable.

Despite increasingly aggressive surgery and chemotherapy the 5 year survival of advanced carcinoma of the ovary (Stage III and IV according to the International Federation of Gynaecology and Obstetrics (FIGO)) remains around 10% (Annual Report of Results of Treatment of Gynaecological Cancer, 1982). Radiotherapy has been shown to improve the survival of patients only if they have had complete bilateral salpingo-oophorectomy and total abdominal hysterectomy followed by whole abdominal irradiation with a pelvic boost (Dembo et al., 1979). If radiotherapy can prolong survival in patients with minimal residual disease it is possible that patients with bulky Stage III disease can also benefit from whole abdominal radiotherapy after their tumour masses have been reduced in size by surgery and chemotherapy. Such an approach has been attempted by several groups (Hainsworth et Address for correspondence: G. J. S. Rustin, Cancer Research Campaign Labs, Department of Medical Oncology, Charing Cross Hospital, London W6 8RF

al., 1983; Schray et al., 1984; Hacker et al., 1985; Rizel et al., 1985; Steiner et al., 1985). In view of the large doses of radiotherapy (3000 cGy whole abdominal radiotherapy plus pelvic boost) which frequently followed ten or more coures of chemotherapy, it is not surprising that myelosuppression and bowel toxicity were major problems. For this multi-modality approach to become accepted it requires both efficacy and acceptable toxicity. We,,therefore, designed a protocol to test the toxicity of this approach with a view to using it subsequently in a large randomised trial. We chose the dose of irradiation to the whole abdomen to be the same as that given at the Princess Margaret Hospital, Toronto (Dembo et al., 1979). We omitted the pelvic boost as it was unclear in the Toronto study whether it was of any benefit in patients who had undergone a hysterectomy, and in our experience isolated pelvic relapse is unusual in Stage III patients.. Sequential alternating chemotherapy was given following our encouraging results in other turnouts (Newlands et al., 1983) and its hypothetical advantage of minimising the development of drug resistance (Goldie and Coleman, 1979). Cis-platinum, which appears to have the highest single agent activity against ovarian adenocarcinoma (Barker and Wiltshaw, 1981; Lambert and Berry, 1985), was alternated with methotrexate and cyclophosphamide. This latter combination contains drugs which may not be cross resistant with cisplatinum, and has been shown to produce high complete response rates both alone and in combination with other drugs (Barlow et al., 1980; Barlow and Lele, 1984). We report here our results using this multi-modality approach. PATIENTS AND METHODS

From January, 1981 to December, 1984, 35 consecutive patients referred to our centre with newly diagnosed advanced ovarian cancer were treated according to the protocol detailed below. All patients had histologically proven ovarian adenocarcinoma of FIGO Stages III and IV. Patients with a history of previous malignancy were generally excluded, but one patient with previous breast cancer and another patient with previous colon cancer were treated according to the protocol. They are included in the toxicity data but are not included in the survival analysis. Patients who had received prior chemotherapy or radiotherapy were excluded. Chemotherapy consisted of cis-platinum 100 mg/m 2 as a 24 hour infusion, alternating every 10 to 12 days with cyclophosphamide 500 mg/m 2 and methotrexate 300

270

CLINICAL R A D I O L O G Y

m g / m 2 followed by folinic acid 15 mg 12 hourly for 4 doses beginning 24 h after the start of methotrexate, to a total of eight courses. A 'second look' laparotomy or laparoscopy was performed within 1 month of completing the eighth course of chemotherapy unless there was evidence of progressive disease. The aim of 'second look' surgery was to assess response to chemotherapy and, if necessary, to reduce the size of any residual mass to less than 2 cm maximum diameter. In these cases, and where there was no disease outside the abdomen and pelvis, radiotherapy was recommended. Whole abdominal radiotherapy using anterior and posterior parallel opposed stationary fields extending from 1 cm above the diaphragm to 1 cm below the obturator foramen were used in the majority of patients, although the moving strip technique was used initially in four patients. Treatment was given on a 10 MV linear accelerator with posterior lead (5 HVL) shielding to the kidneys throughout the course. The initially planned regime was 2400 cGy mid-plane dose to be given in 20 daily fractions over 26 days. After completion of radiotherapy no further treatment was given and patients were assessed monthly by clinical examination and scans as indicated. On relapse, patients were treated at the discretion of the responsible physician.

RESULTS

The median age of the patients was 59 years (range 24 to 74 years). Initial debulking surgery including bilateral salpingo-oophorectomy and hysterectomy was attempted in 29 of the 35 patients and was successfully completed in 20 patients. Only three patients had no residual macroscopic disease after primary surgery.

last patient, who had a history of alcoholism died from pneumonia and hyponatraemia having just completed eight courses of chemotherapy. There was no evidence of residual disease at post-mortem. A partial response was defined as a more than 50% reduction in the sum of the products of the two largest diameters of the measurable lesions. Excluding the three patients with no residual disease after primary surgery who remained free of disease at secondary surgery, 26 of the 32 patients (80%) responded to chemotherapy.

Toxicity of Chemotherapy

All 35 patients suffered from severe nausea and vomiting associated with the cis-platinum, and all had partial alopecia. Although myelosuppression was seen in all patients it was rarely severe but resulted in four patients developing septicaemia and five patients having dosage reduction of subsequent chemotherapy. Dosage reductions were also required in two patients because of impaired renal function, in one patient with ascites and in another patient with hearing loss. Two patients declined further chemotherapy after six courses because of toxicity. Of the patients completing eight courses of treatment all but two completed them within 4 weeks of the projected time. Other toxic effects of chemotherapy included moderate to severe parasthesiae (WHO grading 1-2) in 11 patients, one of whom had muscular weakness; reversible deterioration in renal function in three patients; mucositis severe enough to require a liquid diet in seven patients and parenteral nutrition of one patient, and symptomatic hyponatraemia in one patient already mentioned.

Radiotherapy R e s p o n s e to C h e m o t h e r a p y

Response to chemotherapy is shown in Table l and in the majority of patients was judged by 'second look' surgery. Surgery was not performed on all patients because of clear-cut evidence of progressive disease in two, because of an inability to tolerate chemotherapy in one and because of drug related death in another. This Table 1 - Response to chemotherapy of 35 patients

Stage HI

Stage IV

%

Pathologicalcomplete response (CR) 8* 0 23 Macroscopic CR cytologyunknown 4 1 14 Macroscopic CR cytologypositive 5 0 14 Partial response 7 4 31 No change 2 2 1l Progressive disease 2 0 6 *Includes three patients who had no residual macroscopicdisease after primary surgery and one patient who died and whose postmortem examinationshowed no tumour.

In nine patients cytoreductive surgery was performed at the 'second look' operation, resulting in 29 out of 35 (83%) patients having residual turnout deposits of no more than 2 cm diameter and therefore suitable for radiotherapy of the whole abdomen. Two patients had treatment discontinued because of evidence of progressive intra-abdominal disease. Of the remaining 27 patients all but four were treated by the open field technique (Table 2). One patient received only 1960 cGy in 20 fractions over 40 days because of vomiting and she was one of the two patients who also stopped chemotherapy early for the same reason. Two patients electively had slightly lower doses of 2130 and 2180 cGy respectively to the whole abdomen since a pelvic boost was given to one of them and treatment to a para-aortic strip given to the other. The dose given to four patients was higher: two patients were electively given 3000 cGy in 25 fractions and two patients had a dose increase to correct for time delays due to myelosuppression.

Table 2 - Details of radiotherapy in 27 patients

Stage III Stage IV

No. of patients

Technique Open Moving field strip

Range

23 4

19 4

1,960-3,000 2,300-3,000

4 0

Dose in cGy

No. of fractions

Time in days

Modal

Range

Modal

Range

Modal

2,400 2,400

18-29 18-25

20 20

25-67 24-40

28 33

271

ADVANCED OVARIAN CARCINOMA

Table 3 - Toxicity of whole abdominal radiotherapy in 27 patients using WHO grading

Nausea and vomiting Diarrhoea Anemia (Hb×G/100 ml)

Grade 1

H

II1

IV

% Patients

4

3 3 (8.0-9.4) 4 (2.0-2.9) 14 (50-74) 3

3 3 (6.5-7.9)

1 0 (<6.5)

41 22 41

1

0

(1.0-1.9) 6 (25-49) 4

(<1.0) 0 (<25) 0

Not recorded

Leucopenia ( W B C x 109/litre) Thrombocytopenia (plateletsx 109/litre)

(9.5-10.9) 6 (3.0-3.9) 2 (75-99) 8

The majority of patients, 22 out of 27 (81%), completed their treatment within 14 days, and 15 patients (56%) completed within 7 days, of the planned time of 26 days. Treatment in three patients was prolonged up to 19 days and in two patients by 37 and 41 days because of myelosuppression. Myelosuppression was the main toxic effect of treatment (Table 3). The white cell count usually reached its nadir by 1300 cGy (Fig. 1). Similarly, thrombo cytopenia reached a nadir at 1400 cGy and then either levelled or recovered (Fig. 2). Vomiting and diarrhoea were not severe problems and responded to oral medica-

-4.0-

3.0 | g 2.0~o 1 . 0 H-

0

I

0

I 500

I I000

[ 1500

I 2000

I 2400

Dose of radtati0n,cGy Fig. 1 - Relationship between nadir white blood count and dose of radiation in seven patients whose count fell below 2.0 x ]09/litre. Duration of radiotherapy: 0=28 days; e = 3 f , 35 or 36 days; [ i ] = 4 2 days; 11=63 days.

150

0

81 56

tion in all patients. Only in one patient did vomiting interrupt therapy. Six patients (22%) complained of extreme tiredness, in spite of anaemia being corrected if the haemoglobin fell below 10 g/100 ml. There has, to date, been no late toxicity identified despite regular estimations of blood count, liver and renal function. ¢

Survival and Treatment of Relapses This section refers to 33 patients and excludes the two patients with previous malignancies. The median period of those currently alive from diagnosis to the date of analysis (1 February, 1986) is 26 months (14 to 59). Seventeen of the 27 patients completing radiotherapy have relapsed (13 of 23 Stage III, 4 of 4 Stage IV). All but three of the 17 relapses were first noted in the abdomen. Relapses occurred in two of the seven patients in pathological complete remission with negative cytology, in two of the five patients who were macroscopically in complete remission but had no cytology performed and in three of the four patients who had no macroscopic disease but positive cytology after chemotherapy. Of the 11 patients with macroscopic disease at 'second look' only one has not relapsed. She had para-aortic lymphadenectomy and extra radiotherapy to that area. The median duration of survival and median progression-free survival of the 27 Stage III patients was 26 months and 19 months respectively (Fig. 3), and, for the whole group of 33 patients, 26 months and 17 months respectively. Eight of the 17 patients who relapsed had additional chemotherapy. Three of these patients were given cisplatinum (cis-platinum 20 mg/m 2 and chlorambucil in two, cis-platinum 100 mg/m 2 and cyclophosphamide in one) resulting clinically in two complete remissions and one partial response from over 6 to 11 months' duration. A variety of drugs were given as second-line treatment to the other patients.

E 100 I00 i19

> 8C

o. 50

46 6(3

*6

>,

I--

.---- 4c 25 0

[ 0

I 500

I I000

I 1500

I 2000

I 2400

I I] I

Dose of radiaho%cGy

Fig. 2 - Relationship between nadir platelet count and dose of radiation in eight patients whose count fell below 75 x 109/litre. Duration of radiotherapy: 0=25, 27 or 31 days; o=35 days; D=42 days.

II

I

I

o 20

n

I

I

Years

Fig. 3 - Kaplan Meier survival curve of 27 patients with Stage IIl ovarian cancer. _l-=patients alive; N=27.

272

CLINICAL RADIOLOGY

DISCUSSION

This study demonstrates that the great majority of patients can tolerate this combined modality therapy. Apart from those patients who failed to complete the planned treatment because of inadequate tumour response only two patients failed to complete because of toxicity. In one patient the chemotherapy can be implicated in causing her fatal hyponatraemia. Another patient refused further treatment because of the toxicity. In another patient nausea and vomiting resulted in her only completing 80% of the planned doses of both chemotherapy and radiotherapy. In four studies where 3000 cGy whole abdominal radiotherapy with or without boosts was planned after chemotherapy, 13, 20, 40 and 41% of the patients in the respective studies stopped radiotherapy because of myelosuppression (Hainsworth et al., 1983; Schray et al., 1984; Hacker et al., 1985; Steiner et al., 1985). Myelosuppression was seen in the lower dose of 2400 cGy given in this study but no patient failed to complete therapy for this reason. Early in the study radiotherapy was delayed when the total white cell count fell below 2 x 109/litre or the platelet count fell below 75 x 109/litre. However, when it was shown that the white count and platelet count did not fall below 1.5 and 45x109/litre respectively even if radiotherapy was continued, subsequent patients did not have radiotherapy delayed because of myelosuppression. Although Hacker et al. (1985) found that 30% of patients who lived more than 4 months after radiotherapy developed radiation induced intestinal obstruction, no similar problems were noted in this study. This is probably a reflection of the lower whole abdominal dose and omission of the pelvic boost. The planned duration of treatment from the start of chemotherapy to the end of radiotherapy was only 6 months in this study. 'CHAP-5' (cyclophosphamide, hexamethylmelamine, doxorubicin, cis-platinum) which has been advocated as one of the most effective chemotherapy regimens for advanced ovarian carcinoma has a planned duration of treatment of 16 months (Neijt et al., 1984). There is, however, little difference in the median duration of survival of 31 months with CHAP-5, and 26 months in this study. If this result is sustained in studies with more patients, the shorter duration of therapy in our design represents a major improvement in the patients' quality of life. We decided to give radiotherapy to patients in complete remission because of the risk of intra-abdominal relapse. Gershenson et al. (1985) found that 24% of patients shown by laparotomy with multiple biopsies to be in complete remission relapsed within 32 months. As complete remission was defined in our study only by laparoscopy a higher relapse rate would be expected. There were fewer relapses in patients who were in complete remission at 'second look' surgery than in those with positive cytology and all but one of those patients with macroscopic disease have relapsed. This pilot study has shown that this multi-modality therapy is tolerated by the great majority of patients and the results are in line with the best currently being

achieved elsewhere. A multi-centre trial comparing radiotherapy with continuing chemotherapy has therefore been started to show whether the multimodality approach can increase the long term complete remission rate of patients with advanced ovarian cancer, and to identify which patients are most likely to benefit from this approach. Acknowledgement. We are grateful to our gynaecologicalcolleagues who referred the patients and performed the surgery. REFERENCES Annual report of the results of treatment in Gynaecological Cancer (1982). Vol. 18 (Statements for 1973-75 inclusive) Radiumhemmet, Stockholm S-10401, Sweden. Barker, GH & Wiltshaw, E (1981). Use of high dose Cisdichlorodiammine Platinum (II) (N.S. C-119875)Following failure of previous chemotherapy for advanced carcinoma of the ovary. British Journal of Obstetrics and Gynaecology, 88, 1192-1199. Barlow, JJ & Lele, SB (1984). Cis-platin-MECY (methotrexateleucovorin 'rescue' plus eyclophosphamide) versus eis-platinumCHAD (cyclophosphamide, hexamethylmelamine doxorbicin and cisplatin) as initial chemotherapy in Stage III-IV ovarian carcinoma. Cancer Treatment Reports, 68, 1433-1438. Barlow, JJ, Piver, S & Lele, SB (1980). High dose methotrexate with 'rescue' plus cyclophosphamide as initial chemotherapy in ovarian adenoearcinoma. Cancer, 46, 1333-1338. Dembo, AJ, Bush, RS, Beale, FA, Bean, HA, Pringle, JF, Sturgeon, J et al. (1979). Ovarian carcinoma: improved survival following abdominopelvic irradiation in patients with a completed pelvic operation. American Journal of Obstetrics and Gynaecology, 134, 793-800. Gershenson, DM, Copeland, LJ, Warton, JT, Atkinson, EN, Sneige, N, Edwards, CL et al. (1985). Prognosis of surgically determined complete responders in advanced ovarian cancer. Cancer, 55, 1129-1139. Goldie, JH & Coldman, AJ (1979). A mathematical model for relating the drug sensitivity of tumours to the spontaneous mutation rate. Cancer Treatment Reports, 63, 1727-1733. Hacker, NF, Berek, JS, Burnison, CN, Heintz, PM, Juillard, GJF & Lagasse, LD (1985). Whole abdominal irradiation and salvage therapy for epithelial ovarian cancer. Obstetrics and Gynaecology, 65, 60-66. Hainsworth, JD, Arnold, M, Johnson, DH, Burnett, LS, Jones, HW & Greco, A (1983). Advanced minimal residual ovarian carcinoma: abdominopelvic irradiation following combination chemotherapy. Obstetrics and Gynaecology, 61,619-623. Lambert, HE & Berry, J (1985). High dose cis-platin compared with high dose cyclophosphamide in the management of advanced ovarian cancer (FIGO Stages III and IV): report from the North Thames Co-operative Group. British Medical Journal, 290, 889893. Neijt, JP, Bokkel Huinink, WWt, van der Burg, MEL, van Oosterom, AT, Vriesendorp, R, Kooyman, CD, et al. (1984). Randomised trial comparing two combination chemotherapy regimens (HEXACAF vs CHAP-5) in advanced ovarian carcinoma. Lancet, 2, 594600. Newlahds, ES, Begent, RHJ, Rustin, GJS, Parker, D, Bagshawe, KD (1983). Further advances in the management of malignant teratomas in the testis and other sites. Lancet, i, 948-952. Rizel, S, Biran, S, Anteby, SO, Brufman, G, Sulkes, A, Milwidsky, A et al. (1985). Combined modality treatment for stage III ovarian carcinoma. Radiotherapy and Oncology, 3, 237-244. Schray, M, Martinez, A, Howes, A, (1984). Toxicity of whole abdominal irradiation in advanced ovarian cancer following surgery, chemotherapy and re-operation. Proceedings of the American Society of Clincial Oncology, 170. Steiner, M. Rubinov, R, Borovik, Cohen, Y & Robinson, E (1985). Multimodal approach (surgery, chemotherapy and radiotherapy) in the treatment of advanced ovarian carcinoma. Cancer, 55, 27482752.