Asia-Oceania J . Obstet. Gynaecot. Vol. 17, No. 1: 5-72 T9Q1

Local Thermo-Radiotherapy in Carcinoma Cervix : Improved Local Control versus Increased Incidence of Distant Metastasis

Sanjiv Sharma,l) Sandeep Singha1,l) A.P. S. Sandhu,l) Sushmita Ghosha1,l) B. D. Gupta,l) and N. S. Yadav2) Department of Radiation Therapy, Post Graduate Institute of Medical Education and Research, Chandigarh, India 2) Department of Physiotherapy, Post Graduate Institute of Medical Education and Research, Chandigarh, India 1)

Abstract I n 1986, 50 patients with stages I1 and 111 carcinoma of the cervix were entered into this prospective randomized study. Twenty-five cases (Group I) were treated only by radical radiation whereas remaining 25 cases (Group 11) received local hyperthermia in addition to radical radiation. Hyperthermia was delivered by intracavitary brachyhyperthermia approach using an endotract applicator. Both the groups were followed up for a minimum period of 18 months. Group I1 patients achieved better local control (14 out of 20 evaluable cases) than the Group I patients (11 out of 22 evaluable cases). A disturbing observation was the increased incidence of distant metastasis in Group I1 (4 out of 23 cases) as compared to Group I (1 out of 23 cases), though most of them remained disease free locally. The increasing use of hyperthermia in the management of various cancers needs to be reviewed in this context.

Key words: carcinoma cervix, intracavitary brachyhyperthermia, local control, distant metastasis

Introduction A large amount of clinical data has accumulated in recent years which shows that hyperthermia is an effective modality in cancer management.l) There is a clear evidence to prove that systemic or local hyperthermia, alone or in combination with radiotherapy2) and/or chemotherapy3) results in substantial

therapeutic benefit. Consistently promising results have been obtained in a vast variety of histologies and tumours of various sites and sizes.') There is a considerable controversy as to whether the application of hyperthermia is associated with the spread of the tumours from the primary site. The results of various studies are conflicting in nature. No alteration

-Received: Jan. 8,1990 Reprint request to: Dr. Sanjiv Sharma, Department of Radiation Therapy, Post Graduate Institute of Medical Education and Research, Chandigarh 160 012, India

5

S. SHARMA ET AL.

in the pattern or the incidence of metastasis was observed by many authorss-10) whereas some others reported a less incidence of metasta~is,l~-1~) At the same time, few authors have found an increased incidence of metastasis following the application of hyperthermia.14--20)Some authors have observed that the incidence of distant metastasis decreased with local hyperthermia whereas it increases with systemic hyperthermia.21-22) However, most of the information is based upon experimental studies only and no clinical trials have been directed towards this aspect of hyperthermia, the knowledge of which can greatly influence and modify the future clinical applications of hyperthermia. The present report analyses the results of a prospectively randomized study of local hyperthermia utilizing the intracavitary brachyhyperthermia approach combined with radiation in the treatment of the carcinoma of the cervix. The impact of thermo-radiotherapy on the local control as well as the incidence of distant metastasis shall be presented and discussed.

Material and Methods Fifty histologically proven cases of the carcinoma of the cervix were selected for this study, according to the criteria listed in Table 1. The patients were treated between March 1986 and November 1986. The details of hyperthermia and radiotherapy have been described elsewhere23) and shall only be briefly reviewed here. During local per speculum examination, Table 1. Selection criteria 1. Patients suitable for radical radiation at the time of diagnosis. 2. Histology-squamous cell carcinoma. 3. Size of tumour more than 2 cm. 4. Stages If and 111 cases only according to FIG0 Staging System. (Modified 1974) 5. Age60.

6

the exact size of the primary tumour was measured in 2 dimensions using a specially designed calliper. This measurement was utilized in the selection criteria. This mode of measurement did not take into account the parametrial infiltrations and provided the dimensions of the primary tumour only on the cervix. This was considered necessary, in view of the purely local nature of hyperthermia, as will be clearly evident subsequently. The patients were randomized blindly into 2 groups of 25 each by the sealed envelope technique. Group I patients served as the control group and received radical radiotherapy alone, by means of a 6OCo photon beam at 60 or 80 cm source to skin distance. A dose of 45 Gy in 20 fractions over 4 weeks by 2 parallel opposed anterior posterior portals was delivered, This was followed by a single intracavitary l37Cs application, delivering a dose of 35 Gy to point A. However, if the response to the initial external radiation was poor and the patient was judged not suitable for intracavitary application, a supplementary dose of 20 Gy in 10 fractions over 2 weeks was delivered using the same portals. The usual portal size was 1 4 16~cm. Group I1 patients constituted the study group and were treated by a combination of hyperthermia and the same radical radiation protocol as was used for the Group I patients. These patients received hyperthermia every alternate day i.e., 3 times a week for 4 weeks along with the initial external radiation course. A modified intracavitary brachyhyperthermia approach was adopted, Similar approach has been described earlier for the carcinoma of the esophagus.24)Using a 27.12 MHz radiofrequency (RF) machine, the tumour temperature was raised to 4243°C and maintained at that level for 30 minutes (Fig. 1). Radiation followed hyperthermia within a half an hour. Thus the Group I1 patients received 5 fractions of external radiation and 3 sessions of hyperthermia per week. Supplementary radiation whenever required was not accompanied by hyperthermia. A specially designed intraluminal RF heating system (Fig. 2) was used in order to ob-

METASTASIS AFTER LOCAL HYF'ERTHERMIA

36

Fig. 2. The radiofrequency heating system.

.* a a

..t

. I T

.* TIP OF VAGINAL ELECTRODE

Fig. 3. The mechanism of concentration of heat near the active electrode. rain a selective tumour heating. This system consisted of a small endotract (intravag~al) applicator, a large extracorporeal electrode and a R F generator operating at 27.12 MHz. The system was based upon well-known observation thar in the case of the capacitive RF

heating, electromagnetic energy is concentrated near the electrode with the smaller surface (Fig. 3) which in our case was a thin metallic electrode covered with sufficiently thick and suitably designed silicon rubber covering to provide insulation. No cooling

7

S. SHARMA

ET AL.

a-

5e-

-

Fig. 4. Diagramatic representation of the treatment set up in a sagittal section.

system was employed. The endotract applicator was placed in direct contact with the tumour, placing it in the anterior or the posterior fornix, depending upon the convenience of placement. The extracorporeal electrode was then placed either on the abdomen anteriorly or the buttocks posteriorly, depending upon the placement of the vaginal applicator such that the tumour tissue was sandwiched within the electromagnetic field (Fig. 4). It was assumed that since the maximum heat would be concentrated near the intravaginal electrode because of its smaller surface, the tumour tissue lying in close contact would take up the heat. This approach, though not an ideal one, was considered the best in our circumstances, since the direct implantation of the electrodes into the tumour was not considered feasible in day to day clinical practice. Similarly, in the usual clinical setting, it was considered difficult and impractical to get thermometry by means of direct insertion of the thermocouples into the tumour during each treatment. Hence the thermometry technique had to be modified. A fine medical grade thermocouple capable of measuring temperature up to an accuracy of 0.1"C was carefully fixed to the inner surface of the endotract applicator such that it remained in close contact both with the tumour surface as well as the applicator. A similar thermometry 8

Table 2. Composition of 2 groups Criterion

Group Group I I1 (Yrs) (Yrs)

( A ) Age median 48.07 ( B ) Stage of disease 1 IIA IIB 2 IIIB 22 ( C ) Primary clinical disease 11 Proliferative Infiltrative 11 Ulcerative 3 ( D ) Size of primary tumour 2-4 cm 6 >4 cm 19 ( E ) Histological grade (Broder's) Well differentiated 4 Moderately differentiated 19 Poorly differentiated 4 ( F ) Menstrual status Premenopausal 12 Post-menopausal 13 1 ( G ) Diabetes mellitus 1 ( H ) Hypertension ( I ) Anaemia ( < l o gmldl) 2 ( J ) Myocardial ischaemia (K) Pulmonary tuberculosis 3 ( L ) Urinary tract infection 3

50.12 2 2 21 10 13

2 7 18

5 18 4 13 12 2

3 1 1 2

technique has been described by other authors recently.84) The whole system was regulated by means

METASTASXS AFTER LOCAL HYPERTHERMKA

of an automatic onloff control. It was during the 5 seconds off phase of the machine after every 55 seconds that the thermocouple measured the temperature which was displayed on an electronic digital display system. Informed consent was taken in all the cases after the nature of the procedure was fully explained. No sedation was used for hyperthermia treatment. Table 2 shows the composition of 2 groups according to the various prognostic criteria. Both the groups turned out to be reasonably matched. All the patients were regularly followed up. Minimum follow-up is 18 months and maximum 23 months. The end point of evaluation was freedom from disease locally. Freedom from disease was defined as complete disappearance of all visible, palpable and measurable disease Locally. Since the hyperthermia delivered was purely of local nature, only local disappearance of tumour was evaluated to assess its efficacy. Any pelvic or distant recurrence was ignored in the assessment of the response as long as the patient remained disease free locally. The statistical significance was determined by 2tailed Student’s t-test.

Results All the 50 patients completed their treatment and there were no dropouts. The temperature of 43°C could be attained and maintained in 23 out of 25 patients. Two patients in each group were lost to follow-up immediately after completion of the treatment. I n Group I, one patient died of uraemia at 6 months with evidence of gross local disease. Another developed liver metastasis at 14 months but is still alive with no evidence of local disease. Another patient developed gross parametrial recurrence at 12 months, though locally she was free of disease even at 18 months follow-up period. Thus out of 22 evaluable cases at the end of 18 months, 11 (50%)were free of local disease and 1 (4.34%) had developed liver metastasis (Tables 3 and 4). The rest had evidence of uncontrolled local disease. I n Group 11, 3 patients died, one of liver

Table 3. Comparison of results of treatment in Groups I and 11 at 18 months follow-up period* Group

I

Total cases studied Cases evaluated No evidence of disease Evidence of residual disease

25 22 11 (50.00) 11 (50.00)

Group

I1 25 20 14 (70.00) 6 (30.00)

Figures in parenthesis indicate percentage. P=.O5 ** : one patient developed concomitant local disease. Rest of them were disease free locally.

metastasis at 12 months, one of bony metastasis to left femur at 15 months and one of renal failure due to bilateral hydronephrosis at 10 months. The first 2 patients were disease free locally. Thus only 20 patients were available for analysis at 18 months. Out of these, one developed bone metastasis to skull at 15 months but was locally disease free at 18 months, Another developed gross local disease along with liver metastasis at 12 months but was still alive at 18 months. Thus 14 (70%) out of 20 evaluable cases were free of disease locally and 4 (17.39%) out of total 23 cases had developed distant metastasis. The difference for local disease status was significant statistically. The difference for distant metastasis was not significant and at best can be termed as a trend, though a disturbing one. More studies with larger patients numbers and prolonged follow-up might

9

S. SHARMA ET AL.

be more revealing. No correiation was found between the development of distant metastasis and various prognostic criteria. The hyperthermia was well toierated and did not lead to any major complication. It did not in any way enhance the incidence and grade of radiation reactions.

Discussion Although successful treatment of patients with primary neoplasms is often possible, death frequently occurs from disseminated tumour. Thus the ability of the primary neoplasms to metastasize remains a major obstacle in search for a cure of human cancers. Hence, while evolving newer modalities for cancer treatment in future, the major emphasis should he placed on prevention or successful management of micrometastasis. Although hyperthermia has rapidly emerged as the fourth modality of cancer management, in addition to surgery, radiotherapy and chemotherapy, its effect on induction of metastasis is as yet not clear. It has been known for many years that tumour cells can be detected in blood stream of patients wih cervical cancer but this is of unknown prognostic significance in untreated or treated All of such patients do not necessarily develop metastasis clinically, Some kind of host resistance or sensitivity must exist to determine whether the circulating tumour cells can grow and form a metastatic focus.25) T-cell or mo~ocfonalantibody homing to the tumours and the subsequent macrophage activation may help in preventing the formation of metastasis.2e) How hyperthermia affects these phenomenon will ultimately determine its outcome upon the incidence of distant metastasis. Lucke14) in 1949 was the first person to report induction of metastasis of frog carcinoma by increase of environmental temperature. He observed that the pond temperature of spring and autumn of about 7°C was associated with very few metastasis and the frogs died of local kidney tumours but the summer temperature of up to 28°C resulted in marked increase in both the incidence and 10

number of metastasis. The exact mechanism of the effect of hyperthermia on the development of metastasis is not clear. Many theories have been proposed but this question is not settled as yet. One view is that the enzymes important in the invasion of the metastasis into the tissues and through the basement membrane of blood vessels and/or lymphatics may be more effective at higher temperat~re.~7) Another view assumes that the metabolic activity of heated tumour cells changes which may affect the genome and the expression of cell surface epitopes.28)Some feel that since the survival of circulatory tumour cells and their deposition and binding to the capillary basement membrane of the end organs such as lung requires lamina and fibronectin, it is these substances which may be affected by heat resulting in an increased incidence of metasta~is.2~) The immune system is widely implicated in the association between the hyperthermia and the development of the metastasis. Many in vitro studies have shown clearly that hyperthermia increases the incidence of metastasis by inactivating natural killer (NK) cell^.^^-^^) Another in vivo study has reported an impairment in immune reactivity after the systemic heat therapy, resulting in an increase in the number of spontaneous metastasis.22) But 2 in vivo studies have shown that whole body hyperthermia boosts NK activity thereby reducing the incidence of development of metasta~is.l~?~*) The reasons for this discrepancy between the in vivo and the in vitro results are unclear at the moment and need further elucidation. Meanwhile, an antigenic theory has been proposed to explain the decrease in the incidence of metastasis following h~perthermia.~~) According to it following the tumour heating, absorbed necrotic cancer cell byproducts provide the antigenic stimulus needed to enhance the immune system which then leads to the destruction of the distant metastatic foci. 'Very few animal studies and few clinical data currently exist concerning the effects of hyperthermia on the immune response, especially on the NK cell activity. Thus, whether hyperthermia

MmASTASIS AFTER LOCAL H ~ E R T H ~ M I A

suppresses or enhances immune defences against distant metastasis still remains a question unsettled. The relationship between the nature of hyperthermia and the development of metastasis is controversial. WhiIe some studies show that local hyperthermia increases metasta~is,~6-~7,2~) the results of some others are to the contrary.llp12) Similarly, whereas some author~1~?~**1~) feel that the systemic hyperthermia stimulates the development of metastasis, there are others who have noticed a decreased incidence with it.I3) In all these studies, development of metastasis was independent of whether the primary tumour could be controlled or not. Similar controversy exists regarding the effect of tumour temperature on the development of metastasis. Majority of studies show that the incidence of metastasis increases at subtherapeutic hyperthermia1s*21*a2)while others have found the same to be true for therapeutic hyperthermia.17) One study shows an increase in the incidence of the metastasis at both the levels of hyperthermia.20) In our study, the patients treated with local thermoradiotherapy showed a higher incidence of development of distant metastasis as compared to the patients treated with radiation alone. The difference, although not significant statistically, is of disturbing nature as it tends to negate any benefit obtained from improved local control. But we are unable to comment upon the reasons behind this observation. Since hyperthermia interacts synergistically both with radiotherapy and the chemotherapy, it is possible that addition of these modalities by their cytotoxic effect might prevent any increase in distant metastasis induced by hyperthermia. The timing of these additional modalities in relation to hyperthermia may be critical in this regard and further studies are needed to evaluate these factors.

References 1. Field SB. Hyperthermia in the treatment of cancer. Phys Med Bioll987; 32: 789-811 2. Hetzel FW, Dunn JA. Hyperthermia and radiation in cancer therapy: A review. Radiut

Phys C h m 1984; 24: 337-345 3. Haas GP, Klugo RC, Hetzel FW, Barton LE, Cerny JC. I-Iyperthermia and chemotherapy. J Ur01 1984; 132: 828-833 4. Overgaard J. Rationale and problems in the design of clinical studies. In: Overgaard J, ed. Hyperthermic Oncology, Vol. 2. London: Taylor and Francis, 1984: 325-338 5. Cater DB, Silver IA, Watkinson DA. Combined therapy with 220 KV roentgen and 10 cm microwave heating in rat hepatoma. Acta Radio2 1964; 2: 321-326 6. Wahl A, Brustad T, Mossige J, Effect of hyperthermia alone and in combination with 6oCoradiation on the growth of B16 melanoma in mice. Acta Radio1 Oncol1979; 18: 343-356 7. Hahn EW, Aifieri AA, Ho KJ. The significance of local tumour hyperthermia~~diation on the producton of disseminated disease. Int J Radiut Oncol Biol Phys 1979; 5: 819823 8. Hill SA, Denekamp J. Does local tumour heating in mice influence metastatic spread? BYJRadioll982; 55: 444-451 9. Urano M, Rice L, Epstein R, Suit HD, Chu AM. Effect of whole body hyperthermia on cell survival, metastasis frequency and host immunity in moderately and weakly immunogenic murine tumours. Cancer Res 1983 ; 43 : 1039-1043 10. Baker DG, Sager H, Constable WC, Goodchild NT. Influence of ultrasound induced hyperthermia and X-irradiation on the incidence of metastases from a solid tumour. InaasMet1984;4:111-124 11. Schechter M, Stowe SM, Moroson H. Effects of hype~hermiaon primary and metastatic tumour growth and host immune response in rats. Cancer Res 1978; 38: 498-502 12. Marmor JB, Pounos D, Hahn GM. Clinical studies with ultrasound induced hyperthermia. Natl Cancer Inst Monogr 1982; 6: 333337 13. Shen R, Hornback NB, Shidnia H, Shupe RE, Brahmi 2.Whole body hyperthermia decreases lung metastasis in lung tumour bearing mice, possibly via a mechanism involving natural killer cells. J Clin Immun 1987; 7: 246-253 14. Lucke B, Schlumberger H. Induction of metastasis of frog carcinoma by increase of environmental temperature. J Exp Med 1949; 89: 269-279 15. Dickson JA, Ellis HA. Stimulation of tumour cell dissemination by raised temperature

11

S. SHARMA ET AL.

(42°C) in rats with transplanted Yoshida tumours. Nature 1974; 248: 354-358 16. Dickson JA, Ellis HA. The influence of tumour volume and the degree of heating on the response of the solid Yoshida sarcoma to hyperthermia (4042°C). Cancer Res 1976; 36: 1188-1 195 17. Walker A, McCallum HM, Wheldon TE, Nias AHW, Abdelaal AS. Promotion of metastasis of C3H mouse mammary carcinoma by local hyperthermia. BY J Cancer 1978; 38:

561-563 18. Lord PF, Kapp DS, Morrow D. Increased skeletal metastasis of spontaneous canine osteosarcoma after fractionated systemic hyperthermia and local X-irradiation. Cancer Res

1981;41:4331-4334 19. Oda S, Koga S, Maeta M. Effects of total body hyperthermia on metastases from experimental mouse tumour. Cancer Res 1985;

45: 1532-1535 20. Nathason SD, Haas GP, Bobrowski R, et al.

21.

22.

23.

24.

12

Regional lymph node and pulmonary metastasis after local hyperthermia of melanomas in C57BL/6 Mice. I n t J Radiat Oncob Biol Phys 1987; 13 :243-249 Dickson JA, Muckle DA. Total body hypertherrnia versus primary tumour hyperthermia in the treatment of the rabbit VX-2 carcinoma. Cancer Res 1972; 32: 1916-1923 Yerushalami A. Influence on metastatic spread of whole body or local hyperthermia. EUYJ Cancer 1976; 12: 455463 Sharma S , Pate1 FD, Sandhu APS, Gupta BD, Yadav NS. A propsective randomized study of local hyperthermia as a supplement and radiosensitizer in treatment of carcinoma cervix with radiotherapy. Endowrietherapyj ~ ~ ~OPzco11989; t ~5 : 151-159 i a Kai H, Matsufuji H, Okudaira Y, Sugimachi K. Heat, drugs, and radiation given in combination is palliative for unresectable esophageal cancer. Int J Radiat Oncol Biol Phys

1988; 14: 1147-1152 25. Diddle AW, Stroles DM, Hollingsworth J, Kintaw S. Cervical Carcinoma: Cancer cells in the circulating blood. Am J Obstet Gynecol

1959; 78: 582-585 26. Frost P, Fidler IJ. Biology of metastasis. Cancer 1986; 58: 550-553 27. Liotta LA, Tryggvason K, Garbisa S, Hart I, Foltz CM, Shafie S. Metastatic potential correlates with enzymatic degradation of basement membrane collagen. Nature 1980; 284:

67-68 28. Stackpole CW. Generation of phenotypic diversity in the B16 melanoma relative to spontaneous metastasis. Cancer Res 1983 ; 43 : 3057-3064 29. Liotta LA, Mechanisms of cancer invasion and metastasis. In: De Vita VT, Hellman S, Rosenberg SA, eds. Important Advances in Oncology. New York: J. B. Lippincott, 1985; 28 30. Brahmi 2, Morse PA. The Ioss of NK activity following hyperthermia is due to impaired ability of N K cells to bind to targets. 6th International Congress of Immunology, 1986, Toronto, Canada 31. Azocar J, Yunis EJ, Essex M. Sensitivity of human natural killer cells to hyperthermia. Lancet 1982; 1 : 16-17 32. Nurmi T, Unari M, Kouvalainen K. Temperature and natural killer cell activity. Lancet

1982; 1 : 516-517 33. Onsrud M. Effects of hyperthermia on human natural killer cells. Acta Pathol Microbiob Immunol Scand (C) 1983; 91 : 1-8 34. Zanker KS, Lange J. Whole body hyperthermia and natural killer cell activity. Lancet 1982; 1 : 1079-1080 35. Strauss AA, Appel M, Saphir 0, Rabinovitz AJ. Immunologic resistance to carcinoma produced by electro-coagulation, Surg Gynecol Obstet 1965 ; 121 :989-996