- Email: [email protected]
Radiotherapy in the management of cervical cancer in elderly patients
Radiotherapy and Oncology 56 (2000) 9±15
www.elsevier.com/locate/radonline
Radiotherapy in the management of cervical cancer in elderly patients Jacob Christian Lindegaard a, b,*, Ingrid Regitze Thranov b, Svend Aage Engelholm b a
b
Department of Oncology, Building 5, Aarhus University Hospital, DK-8000 Aarhus, Denmark Department of Oncology, The Finsen Center, Copenhagen University Hospital, DK-2100 Copenhagen, Denmark Received 31 August 1999; received in revised form 21 December 1999; accepted 16 February 2000
Abstract Purpose: To report treatment results and complications experienced by elderly patients treated with curatively intended radiotherapy for cancer of the uterine cervix. Patients and methods: One hundred and fourteen elderly patients (median 75.5 years, range 70.0±85.9) consecutively referred for curative radiotherapy in the period 1987±1996 were prospectively followed with regard to tumour control and complications. The importance of age, stage (FIGO), tumour size, histology, tumour ®xation, haemoglobin, concurrent disease, performance status (WHO) and type of radiotherapy were assessed using univariate and multivariate analyses. Results: Treatment was completed as planned in 68%, delayed in 29% and stopped prematurely in 3%. The frequency of grade 3 late complications was 11% and the actuarial probability at 5 years was 20%. Overall 5-year survival according to FIGO was 61% (I), 34% (II) and 25% (III). Cox multivariate analysis identi®ed tumour size as independent prognostic factor for tumour control, disease-free survival and overall survival. FIGO stage was predictive for late grade 2 complications. We were unable to identify signi®cant factors with respect to grade 3 complications. Age was not a signi®cant parameter for any of the investigated endpoints. Conclusion: Elderly patients in good performance status with advanced cancer of the uterine may tolerate radical radiotherapy with acceptable morbidity and reasonable survival. Radiotherapy may also be a good alternative in early stage disease for surgically un®t elderly patients. q 2000 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Cervical cancer; Elderly; Prognostic factors; Radiotherapy
1. Introduction Radiotherapy is the only curative treatment option in most cases of locally advanced cancer of the cervix [28]. However, curative radiotherapy has been regarded as problematic in elderly patients due to increased risk of morbidity [24,30]. Furthermore, several studies now indicate that selected elderly patients may tolerate radical radiotherapy and have comparable survival to that of younger patients [23,25]. Similar results have recently been reported in cervical cancer [14,15]. For early stage cervical cancer, radical surgery is most often the treatment of choice. Several retrospective studies have shown that radical hysterectomy is a safe surgical procedure with minimal postoperative complications in selected elderly patients and that age alone should not be a contraindication for radical hysterectomy [1,4,5,26]. However, as many as 50±75% of elderly patients may have pre-existing medical problems with approximately 10% suffering from severe concurrent disease [1,15]. Intra* Corresponding author.
cavitary brachytherapy may be a viable treatment alternative for this group of medically inoperable patients [9,19,28]. We have for several years offered curatively intended radiotherapy for selected elderly patients with advanced cancer of the uterine cervix, but in good performance status. Radiotherapy has also been used for surgically un®t elderly patients with early stage disease. The aim of the present study was to evaluate this treatment strategy with respect to both morbidity and treatment response.
2. Patients and methods One hundred and fourteen consecutive patients aged 70 years or older with invasive cancer of the cervix referred to the Department of Oncology, Copenhagen University Hospital in the period 1987±1996 for curative radiotherapy, were included in the study. Pre-treatment evaluation included medical history, physical examination, blood tests and chest X-ray. Biopsy, gynaecological examination and cystoscopy were performed during general anaesthesia.
0167-8140/00/$ - see front matter q 2000 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0167-814 0(00)00168-7
10
J.C. Lindegaard et al. / Radiotherapy and Oncology 56 (2000) 9±15
The patients were staged according the guidelines of the Federation of Gynaecology and Obstetrics (FIGO). In general, elderly patients with a tumour size less than 2 cm or surgically un®t patients with centrally located tumours up to 5 cm were offered intracavitary brachytherapy (BT) delivering 60±66 Gy to point-A in two fractions 1 week apart (Nucletron Selectron). BT was delivered at medium dose rate (1.80 Gy/h) using a tandem and ring applicator. The applicators were remotely after-loaded with 137Cs pellets with source strength of 1.48 GBq. A combination of external beam radiotherapy (EBRT) and BT was used for advanced disease. EBRT was delivered with 8- or 18-MV photon beams. From 1987 to 1993 the treatment strategy was 40 Gy in 20 fractions, ®ve fractions per week, delivered with anterior±posterior ®elds with a central split, followed by 60 Gy to point-A delivered in two BT sessions 1 and 2 weeks later, respectively. From 1993, CT-based dose planning became available (Varian/ dosetek). EBRT was changed to 46 Gy in 23 fractions using a four-®eld box technique. Point-A was then treated to 35 Gy in two BT fractions spaced by 1 week. Thus, overall treatment time for EBRT and BT was 6±7 weeks. Tumour control and complications to radiotherapy were evaluated prospectively by interview and gynaecological examination at 3-month intervals for the ®rst 1.5 years, at 6-month intervals in years 2±3, and once every 12 months in years 4 and 5 after treatment. Treatment-related morbidity in vagina, pelvis, intestine and bladder was graded as mild, moderate or severe by an arbitrary scale. Fistulas, perforation and complications requiring surgery were recorded separately. The grading of the recorded morbidity was translated into the Franco±Italian system using the criteria given by Chassagne et al. [2]. However, morbidity requiring surgical intervention was always considered a grade 3 complication irrespective of whether damage was lasting. The variables chosen for analysis of prognostic information were age, FIGO stage, tumour size, histology, bilateral tumour ®xation, haemoglobin, performance status (WHO), concurrent disease and type of radiation treatment. Evaluation of pre-treatment performance status was assessed retrospectively using the method described by Kamby and Sengelùv [12]. Survival time was analyzed using the Kaplan±Meier product-limit technique. Time was measured from start of treatment. The endpoints selected for survival time analyses were tumour control, disease-free survival, overall survival and grade of complication. For the actuarial analysis of tumour control, patients without recurrence were censored at last follow-up or date of death. The events for calculation of disease-free survival were recurrence or death from any cause. Patients alive without recurrence were censored at the date of last follow-up. The events for the actuarial morbidity functions were morbidity grade 1 or worse, grade 2 or worse, and grade 3 or worse, respectively. Patients with recurrence or dying before development of a
Table 1 Patient and tumour characteristics of 114 elderly patients with cancer of the uterine cervix treated with de®nitive radiotherapy Variable
Strati®cation
No.
Age (years)
Median (range) I (A/B) II (A/B) III (A/B) IVA ,2 2±5 .5 Squamous Adenocarcinoma Other No Yes Median (range) 0±1 2±3 No Yes
75.5 (70.0±85.9) 25 (1/24) 32 (8/24) 55 (6/49) 2 23 51 40 103 8 3 104 10 7.9 (4.8±9.7) 104 10 46 68
Stage (FIGO)
Tumour size (cm) Histology Bilateral ®xation Haemoglobin (mmol/l) Performance (WHO) Concurrent disease
given grade of morbidity were censored at date of recurrence or date of death. Median observation time for all patients with regard to overall survival was 2.8 (range 0.4±8.7) years. The log-rank test was used for univariate comparison (SPSS). The Cox proportional hazards model was used for multivariate analysis. Only variables with a univariate P value of less than 0.10 were included in multivariate analysis using step-wise forward selection. A probability of less than 0.05 (two-sided) was considered to indicate signi®cance. 3. Results The characteristics of the patients are given in Table 1. The median age of the patients was 75.5 (range 70.0±85.9) years. One patient was in FIGO stage IA and 24 in FIGO stage IB. Almost 50% of the patients were in stage III±IVA Table 2 Treatment characteristics according to FIGO stage of 114 elderly patients with cancer of the uterine cervix treated with de®nitive radiotherapy a FIGO
BT alone
EBRT 1 BT
I II III IV Total
20 6 1
5 25 54 1 85
a
27
EBRT alone 1 1 2
BT, brachytherapy; EBRT, external beam radiotherapy.
Total 25 32 55 2 114
J.C. Lindegaard et al. / Radiotherapy and Oncology 56 (2000) 9±15 Table 3 Treatment characteristics according to tumour size of 114 elderly patients with cancer of the uterine cervix treated with de®nitive radiotherapy Tumour size
BT alone
EBRT 1 BT
,2 cm 2±5 cm .5 cm Total
17 10
6 39 40 85
27
EBRT alone
Total 23 51 40 114
2 2
and more than one-third had a primary tumour larger than 5 cm. The vast majority of the tumours were found to be squamous cell carcinomas (95%). There were eight adenocarcinomas, two adenosquamous carcinomas and one patient with undifferentiated carcinoma. More than 90% were in good performance status. Concurrent disease was noted in 60% of the patients with 25% having more than one concurrent diagnosis. The most frequent problems were cardiovascular (37%), endocrinological (13%) or from the locomotor system (10%). Previous cancer was found in 7%. Type of radiotherapy treatment is shown according to FIGO stage (Table 2) and tumour size (Table 3). Most patients in stage I were treated with BT only. A patient with stage IA disease was treated with BT, as she was considered un®t for operative procedures involving general anaesthesia. Five patients in stage IB with large tumours were treated with combined EBRT and BT. For stage II± III most patients were treated with a combination of EBRT and BT. One patient with stage IIIB disease and poor performance status was treated only with BT. A patient with stage IIIA disease was treated with 46 Gy EBRT followed by interstitial BT giving 30 Gy to the tumour with 0.5 cm margin [11]. Complications or problems causing treatment interruption are described in Table 4. Treatment was carried through in 111 patients (97%) but was stopped prematurely in three cases (3%). One patient developed ileus during EBRT and one patient was not able to complete BT because of mental confusion. One patient did not want to complete treatment for reasons unrelated to the treatment or concurrent disease. Overall treatment time was prolonged in 33 patients (29%) with 26 patients experiencing a treatment delay for more
11
than 1 week. Perforation of uterus or vagina was the most frequent complication causing treatment delay. Acute complications to radiotherapy causing delay were diarrhoea in ®ve patients and moist desquamation in one patient. Aggravation of concurrent disease was seldom a problem, with two cases of cardiac problems and one case of mental confusion. Technical or administrative problems with delivery of BT caused delay in ten patients. Type and grade of late complications according to the modi®ed Franco±Italian system are given in Table 5. Nearly all patients experienced a grade 1 complication. The dominant types of grade 1±2 complication were moderate to severe vaginal agglutination and pelvic ®brosis. Fifteen grade 3 complications were recorded in 13 patients. One patient developed both intestinal perforation and recto-vaginal ®stulation. Two patients suffered from simultaneous vesico-vaginal and recto-vaginal ®stulas, two patients developed vesico-vaginal ®stulas and one patient developed a recto-vaginal ®stula. Six patients had intestinal stenosis ameliorated by intestinal resection or colostomy. One patient died 20 months after completion of radiotherapy from an uncontrollable peri-anal infection. This patient was tumour free and was considered dead from radiationinduced late normal tissue damage (grade 4). The actuarial probability for being free from late complications grade 1±3 as a function of observation time is shown in Fig. 1. The median latent time for grade 1 complications was 7 months. The median latent time was increased to 18 months for grade 2 complications and was not reached for grade 3 complications. The actuarial risk for developing a grade 3 complication or worse at 5 years was 20%. Overall survival according to FIGO stage I±III is shown in Fig. 2. At 5 years, overall survival was 61% for stage I, 34% for stage II and 25% for stage III. The actuarial ®gures for tumour control at 5 years were 80% for stage I, 53% for stage II and 46% for stage III (Fig. 3). Tumour control according to tumour size is shown in Fig. 4. There was no signi®cant difference in tumour control between tumours smaller than 2 cm and tumours measuring 2±5 cm (P 0:66). However, tumours larger than 5 cm were significantly more dif®cult to control (P 0:003). When looking at tumour control as function of treatment strategy, tumour
Table 4 Length and cause of treatment interruption in 114 elderly patients treated with curative radiotherapy for cancer of uterine cervix No delay (n 78) Acute irradiation morbidity Concurrent disease Compliance Vaginal perforation Uterine perforation Technica1/administrative problems Infection Total a
Two problems were present in ®ve patients.
,1 week (n 7)
.1 week (n 26)
1 4
6 2 1 4 9 6
1 7
3 31 a
1
Stopped (n 3)
Total (n 114)
1 1 1
7 4 2 4 10 10
3
4 41
12
J.C. Lindegaard et al. / Radiotherapy and Oncology 56 (2000) 9±15
Table 5 Late complications prospectively registered in 114 elderly patients with cancer of the uterine cervix treated with de®nitive radiotherapy a Site
Complication
G1
G2
Vagina
Ulceration Agglutination Fibrosis Lymphoedema Ulceration Stenosis Perforation Fibrosis
27 76 66
7 40 20 21 2 4
G3
G4
identi®ed as signi®cant prognostic variables for grade 1 and grade 2 complications. 4. Discussion
control was 67% at 5 years for the 27 patients treated with brachytherapy only and 52% for the 85 patients treated with both BT and EBRT (data not shown). Univariate analysis indicated that FIGO stage, tumour size and bilateral tumour ®xation were the most important variables with respect to tumour control, disease-free survival, overall survival and grade 1±2 complication (Table 6). We were unable to identify any prognostic variables for grade 3 complications. Univariate analysis also suggested that poor performance status and concurrent disease predicted for better local control but carried no prognostic information with regard to disease-free survival, overall survival or grade of complication. In multivariate analysis, tumour size was the only signi®cant parameter with respect to tumour control, disease-free survival and overall survival (Table 7). Bilateral tumour ®xation and FIGO stage were
Elderly patients have been categorised by age into three groups: (1) young old, 65±74 years; (2) older old, 75±84 years of age; and (3) oldest old, .85 years of age [29]. The median age in our study was 75 years with a range of 70.0±85.9. According to the de®nition above the present paper comprises and compares the treatment outcome for `young old' and `older old'. Two recent papers retrospectively analyzed the importance of age with respect to treatment outcome in cervical cancer and concluded that age is not associated with inferior tumour control or increased morbidity [14,15]. In both studies a relatively large group of patients younger than 70 years old were compared retrospectively with a smaller group older than 70. Thus, the comparisons were undoubtedly dominated by the response of the `younger old' which could have masked a `true' chronological upper limit for curative radiotherapy. However, according to our study age per se is not a signi®cant adverse prognostic factor for tumour control, survival or risk of complications for well-selected patients even older than 75. The fact that age was not a signi®cant factor for treatment outcome when analyzed separately for patients treated with BT only or with a combination of EBRT and BT further substantiates this conclusion (data not shown). Previous data have been con¯icting with regard to acute radiation morbidity in elderly patients treated with curative radiotherapy for cervical cancer. Some authors have reported a high incidence of acute sequelae, with as many as 32±41% unable to complete radiotherapy [6,24] and 13%
Fig. 1. Actuarial estimates of probability for being free from late complications prospectively registered in 114 elderly patients with cancer of the uterine cervix treated with curatively intended radiotherapy. Grading was performed according the Franco±Italian system [2].
Fig. 2. Overall survival as function of observation time in 114 elderly patients with cancer of the uterine cervix treated with de®nitive radiotherapy, strati®ed according to FIGO stage.
Pelvis Intestine Bladder Fistula Operation Death Free from complications a
6 9 1
15
1 8 6
50
101
1 113
Grading was performed according the Franco±Italian system [2].
J.C. Lindegaard et al. / Radiotherapy and Oncology 56 (2000) 9±15
Fig. 3. Tumour control as function of observation time in 114 elderly patients with cancer of the uterine cervix treated with de®nitive radiotherapy, strati®ed according to FIGO stage.
dying from treatment-related complications [6]. In contrast, others have found comparable rates of acute radiation morbidity for young and old patients [14]. Our study supports these observations since only 3% of the patients did not complete radiotherapy and no patient died from acute treatment-related complications. However, a treatment delay of 1 week or longer was quite common (23%). Nearly half of these cases were caused by uterine or vaginal perforation. Evidently insertion of the tandem in the senile uterus posed particular problems especially in those patients treated with BT only. BT was routinely postponed for 1 week, when uterine perforation was diagnosed. Since these perforations seldom caused problems, it may be possible to shorten the treatment delay either by proceeding immediately or within 1±2 days after the incident. Stage, treatment technique and radiation dose have been reported as some of the major prognostic determinants for late complications [18,20,21,27]. In the present study, bilateral tumour ®xation and stage were signi®cant factors for grade 1 and grade 2 complications, respectively. We were unable to identify any statistical signi®cant factors with
13
Fig. 4. Tumour control as function of observation time in 114 elderly patients with cancer of the uterine cervix treated with de®nitive radiotherapy, strati®ed according to tumour size.
respect to grade 3 complications. The limited number of patients probably caused this. The apparent lack of importance of performance status and concurrent disease most likely re¯ects patient selection since these patients only quali®ed for curative radiotherapy because of early stage disease. According to authoritative sources, the risk of major late sequelae of radiation therapy for stage I±-IIA should range between 3±5% and 10±15% for stage IIB±III [19]. However, comparison of late radiation morbidity between different studies is dif®cult not only because of incompatible scoring systems [3] but also because the statistical methods vary considerably. In addition, studies based on prospective design and long observation times will inherently report higher incidences than retrospective studies or studies with observation time falling short of the latency times for late irradiation complications [8,18]. Crude frequencies may vary between 0 and 20% and for actuarial estimates, ®gures from 2 to 38% have been reported [10,13,16,17,19,21,27]. As shown by Pedersen et al. [18] the actuarial incidence is higher, since many patients may die before developing a complication. Our study con®rms this observation. In our
Table 6 Univariate analysis using tumour control (TC), disease-free survival (DFS), overall survival (OS) or grade of complication (G1±G3) as endpoint a Variable
Strati®cation
TC
DFS
OS
G1
G2
G3
Age FIGO Tumour size Squamous cell Bilat. ®xation Haemoglobin Con. disease Performance Brachy. alone
75.5 I/II/III-IV ,2=2±5= . 5 y/n y/n 7.9 y/n 0±1/2±3 y/n
0.91 0.041 0.002 0.43 0.30 0.86 0.049 0.044 0.39
0.66 0.094 0.002 0.93 0.26 0.87 0.38 0.87 0.21
0.65 0.008 0.0001 0.82 0.066 0.88 0.39 0.95 0.021
0.76 0.020 0.34 0.65 0.030 0.96 0.72 0.14 0.63
0.61 0.029 0.10 0.76 0.017 0.87 0.12 0.83 0.073
0.19 0.25 0.35 0.43 0.95 0.68 0.98 0.44 0.11
a
Log-rank tests were used for comparisons between strata.
14
J.C. Lindegaard et al. / Radiotherapy and Oncology 56 (2000) 9±15
Table 7 Cox-multivariate analysis using tumour control (TC), disease-free survival (DFS), overall survival (OS) or grade of complication (G1±G2) as endpoint a
Variable RR 95% CL P value a
TC
DFS
OS
G1
G2
Tumour size 2.2 1.3±3.5 0.002
Tumour size 1.6 1.1±2.2 0.007
Tumour size 1.9 1.3±2.7 0.0003
Bilat. ®xation 2.0 1.0±4.0 0.034
FIGO 1.6 1.1±2.2 0.0093
Only variables with a univariate P value of less than 0.10 were included (see Table 5).
study the crude frequency of grade 3 complications was 11%, whereas the actuarial probability was 20% at 5 years. In the present study we deliberately considered all complications requiring major surgical intervention as a grade 3 complication. However, according to the rules of the Franco±Italian system [2], complications requiring surgical intervention with no lasting damage are considered grade 2. By the use of this de®nition, our frequency of grade 3 complications could be reduced to 9%, since at least two patients with intestinal stenosis were alleviated by resection and therefore downgraded. In the two recent retrospective papers concerning elderly patients with cervical cancer treated with de®nitive radiotherapy, overall moderate to severe morbidity crude frequencies of 9±10% were reported [14,15]. Considering the above-mentioned uncertainties, our prospective study compares favourably with these results with respect to late morbidity. In our study, tumour size was identi®ed as the most important prognostic factor with respect to both tumour control and survival. This is in accordance with numerous previous publications including younger patients [7,19,22]. The observation that poor performance status and concurrent disease predicted better local control in univariate analysis may seem surprising (Table 6). However, as discussed above, this observation merely re¯ects patient selection since most of these patients had early stage disease. The prognostic value of these two variables therefore disappeared with respect to survival in univariate analysis (Table 6) and tumour control in multivariate analysis (Table 7). In the literature, tumour control rates above 90% are expected for stage I cervical cancer treated with radical radiotherapy [7], with local control rates approaching 100% for tumours less than 1 cm treated with brachytherapy only [9]. Compared with these ®gures our tumour control rate of 80% for stage I is somewhat disappointing. It is likely that our choice of treatment may have been in¯uenced by fear of morbidity leading to a tendency for not prescribing EBRT. As demonstrated in Fig. 4, it is quite possible that the smaller tumours were treated insuf®ciently and that better tumour control and even survival would have been obtained with EBRT and BT in the patients who were treated with BT only. In view of the limited morbidity found with combination radiotherapy in the present study as well as in the studies by Mitchell et al. [14] and Mitsuhashi et al. [15], it is important to offer EBRT to a wider range of elderly
patients. On the other hand, our data show that it is possible to achieve meaningful tumour control and survival for elderly un®t patients with BT only in tumours up to 5 cm, when the patient is judged unable to sustain neither surgery nor extensive radiotherapy. The patient with stage IIIB disease treated only with BT due to poor performance status underlines this possibility. It may rightfully be argued that BT alone must be considered as a palliative for stage III disease. However, treatment intention was not clearly stated in the medical charts and the patient was therefore included. The patient died free from recurrence 18 months after the start of treatment. Whether the patient was over-staged, followed up for too short a time or actually obtained tumour control remains uncertain. 5. Conclusion Selected elderly patients with advanced cancer of the uterine cervix may tolerate radical radiotherapy consisting of both BT and EBRT with acceptable toxicity and reasonable survival. BT alone may also be a good alternative in early stage disease for the surgically un®t patient. Age per se is not a signi®cant factor for treatment outcome. For very frail patients not even expected to sustain EBRT, it may sometimes be possible to obtain meaningful tumour control even for centrally located tumours as large as 5 cm with BT only. However, combinations of EBRT and BT should always be applied whenever possible unless the tumour is very small. Acknowledgements This work was supported by Direktùr Jacob Madsens og hustru Olga Madsens Fond. References [1] Boyd ME, Groome PA. The morbidity of abdominal hysterectomy. Can. J. Surg. 1993;36:155±159. [2] Chassagne D, Sismondi P, Horiot JC, et al. A glossary for reporting complications of treatment in gynecological cancers. Radiother. Oncol. 1993;26:195±202. [3] Dische S, Warburton MF, Jones D, Lartigau E. The recording of morbidity related to radiotherapy. Radiother. Oncol. 1989;16:103± 108. [4] Fuchtner C, Manetta A, Walker JL, et al. Radical hysterectomy in the
J.C. Lindegaard et al. / Radiotherapy and Oncology 56 (2000) 9±15
[5] [6] [7] [8]
[9] [10]
[11]
[12] [13]
[14] [15] [16]
elderly patient: analysis of morbidity. Am. J. Obstet. Gynecol. 1992;166:593±597. Geisler JP, Geisler HE. Radical hysterectomy in patients 65 years of age and older. Gynecol. Oncol. 1994;53:208±211. Grant PT, Jeffrey JF, Fraser RC, et al. Pelvic radiation therapy for gynecologic malignancies in geriatric patients. Gynecol. Oncol. 1989;33:185±188. Grigsby PW. Primary radiotherapy for stage IB or IIA cervical cancer. J. Natl. Cancer Inst. Monogr. 1996:61±64. Haie-Meder C, Kramar A, Lambin P, et al. Analysis of complications in a prospective randomized trial comparing two brachytherapy low dose rates in cervical carcinomas. Int. J. Radiat. Oncol. Biol. Phys. 1994;29:953±960. Hamberger AD, Fletcher GH, Wharton JT. Results of treatment of early stage I carcinoma of the uterine cervix with intracavitary radium alone. Cancer 1978;41:980±985. Horiot JC, Pigneux J, Pourquier H, et al. Radiotherapy alone in carcinoma of the intact uterine cervix according to G.H. Fletcher guidelines: a French cooperative study of 1383 cases. Int. J. Radiat. Oncol. Biol. Phys. 1988;14:605±611. Jensen PT, Roed H, Engelholm SA, Rosendal F. Pulsed dose rate (PDR) brachytherapy as salvage treatment of locally advanced or recurrent gynecologic cancer. Int. J. Radiat. Oncol. Biol. Phys. 1998;42:1041±1047. Kamby C, Sengelùv L. Assessment of functional status in patients with invasive carcinoma of the urethelial tract. Urol. Oncol. 1996;2:43±51. Kapp KS, Stuecklschweiger GF, Kapp DS, et al. Carcinoma of the cervix: analysis of complications after primary external beam radiation and Ir-192 HDR brachytherapy. Radiother. Oncol. 1997;42:143± 153. Mitchell PA, Waggoner S, Rotmensch J, Mundt AJ. Cervical cancer in the elderly treated with radiotherapy. Gynecol. Oncol. 1998;71:291±298. Mitsuhashi N, Takahashi M, Nozaki M, et al. Squamous cell carcinoma of the uterine cervix: radiation therapy for patients aged 70 and older. Radiology 1995;194:141±145. Patel FD, Sharma SC, Negi PS, Ghoshal S, Gupta BD. Low dose rate vs. high dose rate brachytherapy in the treatment of carcinoma of the uterine cervix: a clinical trial. Int. J. Radiat. Oncol. Biol. Phys. 1994;28:335±341.
15
[17] Pedersen D, Bentzen SM, Overgaard J. Reporting radiotherapeutic complications in patients with uterine cervical cancer. The importance of latency and classi®cation system. Radiother. Oncol. 1993;28:93±186. [18] Pedersen D, Bentzen SM, Overgaard J. Early and late radiotherapeutic morbidity in 442 consecutive patients with locally advanced carcinoma of the uterine cervix. Int. J. Radiat. Oncol. Biol. Phys. 1994;29:941±952. [19] Perez CA. Uterine cervix. In: Perez CA, Brady LW, editors. Principles and practice of radiation oncology, 3rd. ed. Philadelphia, PA: Lippincott-Raven, 1997. pp. 1733±1834. [20] Perez CA, Breaux S, Bedwinek JM, et al. Radiation therapy alone in the treatment of carcinoma of the uterine cervix. II. Analysis of complications. Cancer 1984;54:235±246. [21] Perez CA, Fox S, Lockett MA, et al. Impact of dose in outcome of irradiation alone in carcinoma of the uterine cervix: analysis of two different methods. Int. J. Radiat. Oncol. Biol. Phys. 1991;21:885±898. [22] Perez CA, Grigsby PW, Chao KS, Mutch DG, Lockett MA. Tumor size, irradiation dose, and long-term outcome of carcinoma of uterine cervix. Int. J. Radiat. Oncol. Biol. Phys. 1998;41:307±317. [23] Pignon T, Scalliet P. Radiotherapy in the elderly. Eur. J. Surg. Oncol. 1998;24:407±411. [24] Sablinska B. Carcinoma of the uterine cervix in women over 70 years of age. Gynecol. Oncol. 1979;7:128±135. [25] Scalliet P. Radiotherapy in the elderly. Eur. J. Cancer 1991;27:3. [26] Shuster PA, Barter JF, Potkul RK, Barnes WA, Delgado G. Radical hysterectomy morbidity in relation to age. Obstet. Gynecol. 1991;78:77±79. [27] Sinistrero G, Sismondi P, Rumore A, Zola P. Analysis of complications of cervix carcinoma treated by radiotherapy using the FrancoItalian glossary. Radiother. Oncol. 1993;26:203±211. [28] Stehman FR, Perez CA, Kurrasch M. Uterine cervix. In: Hoskins WJ, Perez CA, Young RC, editors. Principles and practice of gynecologic oncology, 2nd. ed. Philadelphia, PA: Lippincott-Raven, 1997. pp. 785±858. [29] Yancik R, Ries LG. Cancer in the aged: an epidemiologic perspective on treatment issues. Cancer 1991;68:2502. [30] Zachariah B, Balducci L, Venkattaramanabalaji GV, et al. Radiotherapy for cancer patients aged 80 and older: a study of effectiveness and side effects. Int. J. Radiat. Oncol. Biol. Phys. 1997;39:1125±1129.
www.elsevier.com/locate/radonline
Radiotherapy in the management of cervical cancer in elderly patients Jacob Christian Lindegaard a, b,*, Ingrid Regitze Thranov b, Svend Aage Engelholm b a
b
Department of Oncology, Building 5, Aarhus University Hospital, DK-8000 Aarhus, Denmark Department of Oncology, The Finsen Center, Copenhagen University Hospital, DK-2100 Copenhagen, Denmark Received 31 August 1999; received in revised form 21 December 1999; accepted 16 February 2000
Abstract Purpose: To report treatment results and complications experienced by elderly patients treated with curatively intended radiotherapy for cancer of the uterine cervix. Patients and methods: One hundred and fourteen elderly patients (median 75.5 years, range 70.0±85.9) consecutively referred for curative radiotherapy in the period 1987±1996 were prospectively followed with regard to tumour control and complications. The importance of age, stage (FIGO), tumour size, histology, tumour ®xation, haemoglobin, concurrent disease, performance status (WHO) and type of radiotherapy were assessed using univariate and multivariate analyses. Results: Treatment was completed as planned in 68%, delayed in 29% and stopped prematurely in 3%. The frequency of grade 3 late complications was 11% and the actuarial probability at 5 years was 20%. Overall 5-year survival according to FIGO was 61% (I), 34% (II) and 25% (III). Cox multivariate analysis identi®ed tumour size as independent prognostic factor for tumour control, disease-free survival and overall survival. FIGO stage was predictive for late grade 2 complications. We were unable to identify signi®cant factors with respect to grade 3 complications. Age was not a signi®cant parameter for any of the investigated endpoints. Conclusion: Elderly patients in good performance status with advanced cancer of the uterine may tolerate radical radiotherapy with acceptable morbidity and reasonable survival. Radiotherapy may also be a good alternative in early stage disease for surgically un®t elderly patients. q 2000 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Cervical cancer; Elderly; Prognostic factors; Radiotherapy
1. Introduction Radiotherapy is the only curative treatment option in most cases of locally advanced cancer of the cervix [28]. However, curative radiotherapy has been regarded as problematic in elderly patients due to increased risk of morbidity [24,30]. Furthermore, several studies now indicate that selected elderly patients may tolerate radical radiotherapy and have comparable survival to that of younger patients [23,25]. Similar results have recently been reported in cervical cancer [14,15]. For early stage cervical cancer, radical surgery is most often the treatment of choice. Several retrospective studies have shown that radical hysterectomy is a safe surgical procedure with minimal postoperative complications in selected elderly patients and that age alone should not be a contraindication for radical hysterectomy [1,4,5,26]. However, as many as 50±75% of elderly patients may have pre-existing medical problems with approximately 10% suffering from severe concurrent disease [1,15]. Intra* Corresponding author.
cavitary brachytherapy may be a viable treatment alternative for this group of medically inoperable patients [9,19,28]. We have for several years offered curatively intended radiotherapy for selected elderly patients with advanced cancer of the uterine cervix, but in good performance status. Radiotherapy has also been used for surgically un®t elderly patients with early stage disease. The aim of the present study was to evaluate this treatment strategy with respect to both morbidity and treatment response.
2. Patients and methods One hundred and fourteen consecutive patients aged 70 years or older with invasive cancer of the cervix referred to the Department of Oncology, Copenhagen University Hospital in the period 1987±1996 for curative radiotherapy, were included in the study. Pre-treatment evaluation included medical history, physical examination, blood tests and chest X-ray. Biopsy, gynaecological examination and cystoscopy were performed during general anaesthesia.
0167-8140/00/$ - see front matter q 2000 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0167-814 0(00)00168-7
10
J.C. Lindegaard et al. / Radiotherapy and Oncology 56 (2000) 9±15
The patients were staged according the guidelines of the Federation of Gynaecology and Obstetrics (FIGO). In general, elderly patients with a tumour size less than 2 cm or surgically un®t patients with centrally located tumours up to 5 cm were offered intracavitary brachytherapy (BT) delivering 60±66 Gy to point-A in two fractions 1 week apart (Nucletron Selectron). BT was delivered at medium dose rate (1.80 Gy/h) using a tandem and ring applicator. The applicators were remotely after-loaded with 137Cs pellets with source strength of 1.48 GBq. A combination of external beam radiotherapy (EBRT) and BT was used for advanced disease. EBRT was delivered with 8- or 18-MV photon beams. From 1987 to 1993 the treatment strategy was 40 Gy in 20 fractions, ®ve fractions per week, delivered with anterior±posterior ®elds with a central split, followed by 60 Gy to point-A delivered in two BT sessions 1 and 2 weeks later, respectively. From 1993, CT-based dose planning became available (Varian/ dosetek). EBRT was changed to 46 Gy in 23 fractions using a four-®eld box technique. Point-A was then treated to 35 Gy in two BT fractions spaced by 1 week. Thus, overall treatment time for EBRT and BT was 6±7 weeks. Tumour control and complications to radiotherapy were evaluated prospectively by interview and gynaecological examination at 3-month intervals for the ®rst 1.5 years, at 6-month intervals in years 2±3, and once every 12 months in years 4 and 5 after treatment. Treatment-related morbidity in vagina, pelvis, intestine and bladder was graded as mild, moderate or severe by an arbitrary scale. Fistulas, perforation and complications requiring surgery were recorded separately. The grading of the recorded morbidity was translated into the Franco±Italian system using the criteria given by Chassagne et al. [2]. However, morbidity requiring surgical intervention was always considered a grade 3 complication irrespective of whether damage was lasting. The variables chosen for analysis of prognostic information were age, FIGO stage, tumour size, histology, bilateral tumour ®xation, haemoglobin, performance status (WHO), concurrent disease and type of radiation treatment. Evaluation of pre-treatment performance status was assessed retrospectively using the method described by Kamby and Sengelùv [12]. Survival time was analyzed using the Kaplan±Meier product-limit technique. Time was measured from start of treatment. The endpoints selected for survival time analyses were tumour control, disease-free survival, overall survival and grade of complication. For the actuarial analysis of tumour control, patients without recurrence were censored at last follow-up or date of death. The events for calculation of disease-free survival were recurrence or death from any cause. Patients alive without recurrence were censored at the date of last follow-up. The events for the actuarial morbidity functions were morbidity grade 1 or worse, grade 2 or worse, and grade 3 or worse, respectively. Patients with recurrence or dying before development of a
Table 1 Patient and tumour characteristics of 114 elderly patients with cancer of the uterine cervix treated with de®nitive radiotherapy Variable
Strati®cation
No.
Age (years)
Median (range) I (A/B) II (A/B) III (A/B) IVA ,2 2±5 .5 Squamous Adenocarcinoma Other No Yes Median (range) 0±1 2±3 No Yes
75.5 (70.0±85.9) 25 (1/24) 32 (8/24) 55 (6/49) 2 23 51 40 103 8 3 104 10 7.9 (4.8±9.7) 104 10 46 68
Stage (FIGO)
Tumour size (cm) Histology Bilateral ®xation Haemoglobin (mmol/l) Performance (WHO) Concurrent disease
given grade of morbidity were censored at date of recurrence or date of death. Median observation time for all patients with regard to overall survival was 2.8 (range 0.4±8.7) years. The log-rank test was used for univariate comparison (SPSS). The Cox proportional hazards model was used for multivariate analysis. Only variables with a univariate P value of less than 0.10 were included in multivariate analysis using step-wise forward selection. A probability of less than 0.05 (two-sided) was considered to indicate signi®cance. 3. Results The characteristics of the patients are given in Table 1. The median age of the patients was 75.5 (range 70.0±85.9) years. One patient was in FIGO stage IA and 24 in FIGO stage IB. Almost 50% of the patients were in stage III±IVA Table 2 Treatment characteristics according to FIGO stage of 114 elderly patients with cancer of the uterine cervix treated with de®nitive radiotherapy a FIGO
BT alone
EBRT 1 BT
I II III IV Total
20 6 1
5 25 54 1 85
a
27
EBRT alone 1 1 2
BT, brachytherapy; EBRT, external beam radiotherapy.
Total 25 32 55 2 114
J.C. Lindegaard et al. / Radiotherapy and Oncology 56 (2000) 9±15 Table 3 Treatment characteristics according to tumour size of 114 elderly patients with cancer of the uterine cervix treated with de®nitive radiotherapy Tumour size
BT alone
EBRT 1 BT
,2 cm 2±5 cm .5 cm Total
17 10
6 39 40 85
27
EBRT alone
Total 23 51 40 114
2 2
and more than one-third had a primary tumour larger than 5 cm. The vast majority of the tumours were found to be squamous cell carcinomas (95%). There were eight adenocarcinomas, two adenosquamous carcinomas and one patient with undifferentiated carcinoma. More than 90% were in good performance status. Concurrent disease was noted in 60% of the patients with 25% having more than one concurrent diagnosis. The most frequent problems were cardiovascular (37%), endocrinological (13%) or from the locomotor system (10%). Previous cancer was found in 7%. Type of radiotherapy treatment is shown according to FIGO stage (Table 2) and tumour size (Table 3). Most patients in stage I were treated with BT only. A patient with stage IA disease was treated with BT, as she was considered un®t for operative procedures involving general anaesthesia. Five patients in stage IB with large tumours were treated with combined EBRT and BT. For stage II± III most patients were treated with a combination of EBRT and BT. One patient with stage IIIB disease and poor performance status was treated only with BT. A patient with stage IIIA disease was treated with 46 Gy EBRT followed by interstitial BT giving 30 Gy to the tumour with 0.5 cm margin [11]. Complications or problems causing treatment interruption are described in Table 4. Treatment was carried through in 111 patients (97%) but was stopped prematurely in three cases (3%). One patient developed ileus during EBRT and one patient was not able to complete BT because of mental confusion. One patient did not want to complete treatment for reasons unrelated to the treatment or concurrent disease. Overall treatment time was prolonged in 33 patients (29%) with 26 patients experiencing a treatment delay for more
11
than 1 week. Perforation of uterus or vagina was the most frequent complication causing treatment delay. Acute complications to radiotherapy causing delay were diarrhoea in ®ve patients and moist desquamation in one patient. Aggravation of concurrent disease was seldom a problem, with two cases of cardiac problems and one case of mental confusion. Technical or administrative problems with delivery of BT caused delay in ten patients. Type and grade of late complications according to the modi®ed Franco±Italian system are given in Table 5. Nearly all patients experienced a grade 1 complication. The dominant types of grade 1±2 complication were moderate to severe vaginal agglutination and pelvic ®brosis. Fifteen grade 3 complications were recorded in 13 patients. One patient developed both intestinal perforation and recto-vaginal ®stulation. Two patients suffered from simultaneous vesico-vaginal and recto-vaginal ®stulas, two patients developed vesico-vaginal ®stulas and one patient developed a recto-vaginal ®stula. Six patients had intestinal stenosis ameliorated by intestinal resection or colostomy. One patient died 20 months after completion of radiotherapy from an uncontrollable peri-anal infection. This patient was tumour free and was considered dead from radiationinduced late normal tissue damage (grade 4). The actuarial probability for being free from late complications grade 1±3 as a function of observation time is shown in Fig. 1. The median latent time for grade 1 complications was 7 months. The median latent time was increased to 18 months for grade 2 complications and was not reached for grade 3 complications. The actuarial risk for developing a grade 3 complication or worse at 5 years was 20%. Overall survival according to FIGO stage I±III is shown in Fig. 2. At 5 years, overall survival was 61% for stage I, 34% for stage II and 25% for stage III. The actuarial ®gures for tumour control at 5 years were 80% for stage I, 53% for stage II and 46% for stage III (Fig. 3). Tumour control according to tumour size is shown in Fig. 4. There was no signi®cant difference in tumour control between tumours smaller than 2 cm and tumours measuring 2±5 cm (P 0:66). However, tumours larger than 5 cm were significantly more dif®cult to control (P 0:003). When looking at tumour control as function of treatment strategy, tumour
Table 4 Length and cause of treatment interruption in 114 elderly patients treated with curative radiotherapy for cancer of uterine cervix No delay (n 78) Acute irradiation morbidity Concurrent disease Compliance Vaginal perforation Uterine perforation Technica1/administrative problems Infection Total a
Two problems were present in ®ve patients.
,1 week (n 7)
.1 week (n 26)
1 4
6 2 1 4 9 6
1 7
3 31 a
1
Stopped (n 3)
Total (n 114)
1 1 1
7 4 2 4 10 10
3
4 41
12
J.C. Lindegaard et al. / Radiotherapy and Oncology 56 (2000) 9±15
Table 5 Late complications prospectively registered in 114 elderly patients with cancer of the uterine cervix treated with de®nitive radiotherapy a Site
Complication
G1
G2
Vagina
Ulceration Agglutination Fibrosis Lymphoedema Ulceration Stenosis Perforation Fibrosis
27 76 66
7 40 20 21 2 4
G3
G4
identi®ed as signi®cant prognostic variables for grade 1 and grade 2 complications. 4. Discussion
control was 67% at 5 years for the 27 patients treated with brachytherapy only and 52% for the 85 patients treated with both BT and EBRT (data not shown). Univariate analysis indicated that FIGO stage, tumour size and bilateral tumour ®xation were the most important variables with respect to tumour control, disease-free survival, overall survival and grade 1±2 complication (Table 6). We were unable to identify any prognostic variables for grade 3 complications. Univariate analysis also suggested that poor performance status and concurrent disease predicted for better local control but carried no prognostic information with regard to disease-free survival, overall survival or grade of complication. In multivariate analysis, tumour size was the only signi®cant parameter with respect to tumour control, disease-free survival and overall survival (Table 7). Bilateral tumour ®xation and FIGO stage were
Elderly patients have been categorised by age into three groups: (1) young old, 65±74 years; (2) older old, 75±84 years of age; and (3) oldest old, .85 years of age [29]. The median age in our study was 75 years with a range of 70.0±85.9. According to the de®nition above the present paper comprises and compares the treatment outcome for `young old' and `older old'. Two recent papers retrospectively analyzed the importance of age with respect to treatment outcome in cervical cancer and concluded that age is not associated with inferior tumour control or increased morbidity [14,15]. In both studies a relatively large group of patients younger than 70 years old were compared retrospectively with a smaller group older than 70. Thus, the comparisons were undoubtedly dominated by the response of the `younger old' which could have masked a `true' chronological upper limit for curative radiotherapy. However, according to our study age per se is not a signi®cant adverse prognostic factor for tumour control, survival or risk of complications for well-selected patients even older than 75. The fact that age was not a signi®cant factor for treatment outcome when analyzed separately for patients treated with BT only or with a combination of EBRT and BT further substantiates this conclusion (data not shown). Previous data have been con¯icting with regard to acute radiation morbidity in elderly patients treated with curative radiotherapy for cervical cancer. Some authors have reported a high incidence of acute sequelae, with as many as 32±41% unable to complete radiotherapy [6,24] and 13%
Fig. 1. Actuarial estimates of probability for being free from late complications prospectively registered in 114 elderly patients with cancer of the uterine cervix treated with curatively intended radiotherapy. Grading was performed according the Franco±Italian system [2].
Fig. 2. Overall survival as function of observation time in 114 elderly patients with cancer of the uterine cervix treated with de®nitive radiotherapy, strati®ed according to FIGO stage.
Pelvis Intestine Bladder Fistula Operation Death Free from complications a
6 9 1
15
1 8 6
50
101
1 113
Grading was performed according the Franco±Italian system [2].
J.C. Lindegaard et al. / Radiotherapy and Oncology 56 (2000) 9±15
Fig. 3. Tumour control as function of observation time in 114 elderly patients with cancer of the uterine cervix treated with de®nitive radiotherapy, strati®ed according to FIGO stage.
dying from treatment-related complications [6]. In contrast, others have found comparable rates of acute radiation morbidity for young and old patients [14]. Our study supports these observations since only 3% of the patients did not complete radiotherapy and no patient died from acute treatment-related complications. However, a treatment delay of 1 week or longer was quite common (23%). Nearly half of these cases were caused by uterine or vaginal perforation. Evidently insertion of the tandem in the senile uterus posed particular problems especially in those patients treated with BT only. BT was routinely postponed for 1 week, when uterine perforation was diagnosed. Since these perforations seldom caused problems, it may be possible to shorten the treatment delay either by proceeding immediately or within 1±2 days after the incident. Stage, treatment technique and radiation dose have been reported as some of the major prognostic determinants for late complications [18,20,21,27]. In the present study, bilateral tumour ®xation and stage were signi®cant factors for grade 1 and grade 2 complications, respectively. We were unable to identify any statistical signi®cant factors with
13
Fig. 4. Tumour control as function of observation time in 114 elderly patients with cancer of the uterine cervix treated with de®nitive radiotherapy, strati®ed according to tumour size.
respect to grade 3 complications. The limited number of patients probably caused this. The apparent lack of importance of performance status and concurrent disease most likely re¯ects patient selection since these patients only quali®ed for curative radiotherapy because of early stage disease. According to authoritative sources, the risk of major late sequelae of radiation therapy for stage I±-IIA should range between 3±5% and 10±15% for stage IIB±III [19]. However, comparison of late radiation morbidity between different studies is dif®cult not only because of incompatible scoring systems [3] but also because the statistical methods vary considerably. In addition, studies based on prospective design and long observation times will inherently report higher incidences than retrospective studies or studies with observation time falling short of the latency times for late irradiation complications [8,18]. Crude frequencies may vary between 0 and 20% and for actuarial estimates, ®gures from 2 to 38% have been reported [10,13,16,17,19,21,27]. As shown by Pedersen et al. [18] the actuarial incidence is higher, since many patients may die before developing a complication. Our study con®rms this observation. In our
Table 6 Univariate analysis using tumour control (TC), disease-free survival (DFS), overall survival (OS) or grade of complication (G1±G3) as endpoint a Variable
Strati®cation
TC
DFS
OS
G1
G2
G3
Age FIGO Tumour size Squamous cell Bilat. ®xation Haemoglobin Con. disease Performance Brachy. alone
75.5 I/II/III-IV ,2=2±5= . 5 y/n y/n 7.9 y/n 0±1/2±3 y/n
0.91 0.041 0.002 0.43 0.30 0.86 0.049 0.044 0.39
0.66 0.094 0.002 0.93 0.26 0.87 0.38 0.87 0.21
0.65 0.008 0.0001 0.82 0.066 0.88 0.39 0.95 0.021
0.76 0.020 0.34 0.65 0.030 0.96 0.72 0.14 0.63
0.61 0.029 0.10 0.76 0.017 0.87 0.12 0.83 0.073
0.19 0.25 0.35 0.43 0.95 0.68 0.98 0.44 0.11
a
Log-rank tests were used for comparisons between strata.
14
J.C. Lindegaard et al. / Radiotherapy and Oncology 56 (2000) 9±15
Table 7 Cox-multivariate analysis using tumour control (TC), disease-free survival (DFS), overall survival (OS) or grade of complication (G1±G2) as endpoint a
Variable RR 95% CL P value a
TC
DFS
OS
G1
G2
Tumour size 2.2 1.3±3.5 0.002
Tumour size 1.6 1.1±2.2 0.007
Tumour size 1.9 1.3±2.7 0.0003
Bilat. ®xation 2.0 1.0±4.0 0.034
FIGO 1.6 1.1±2.2 0.0093
Only variables with a univariate P value of less than 0.10 were included (see Table 5).
study the crude frequency of grade 3 complications was 11%, whereas the actuarial probability was 20% at 5 years. In the present study we deliberately considered all complications requiring major surgical intervention as a grade 3 complication. However, according to the rules of the Franco±Italian system [2], complications requiring surgical intervention with no lasting damage are considered grade 2. By the use of this de®nition, our frequency of grade 3 complications could be reduced to 9%, since at least two patients with intestinal stenosis were alleviated by resection and therefore downgraded. In the two recent retrospective papers concerning elderly patients with cervical cancer treated with de®nitive radiotherapy, overall moderate to severe morbidity crude frequencies of 9±10% were reported [14,15]. Considering the above-mentioned uncertainties, our prospective study compares favourably with these results with respect to late morbidity. In our study, tumour size was identi®ed as the most important prognostic factor with respect to both tumour control and survival. This is in accordance with numerous previous publications including younger patients [7,19,22]. The observation that poor performance status and concurrent disease predicted better local control in univariate analysis may seem surprising (Table 6). However, as discussed above, this observation merely re¯ects patient selection since most of these patients had early stage disease. The prognostic value of these two variables therefore disappeared with respect to survival in univariate analysis (Table 6) and tumour control in multivariate analysis (Table 7). In the literature, tumour control rates above 90% are expected for stage I cervical cancer treated with radical radiotherapy [7], with local control rates approaching 100% for tumours less than 1 cm treated with brachytherapy only [9]. Compared with these ®gures our tumour control rate of 80% for stage I is somewhat disappointing. It is likely that our choice of treatment may have been in¯uenced by fear of morbidity leading to a tendency for not prescribing EBRT. As demonstrated in Fig. 4, it is quite possible that the smaller tumours were treated insuf®ciently and that better tumour control and even survival would have been obtained with EBRT and BT in the patients who were treated with BT only. In view of the limited morbidity found with combination radiotherapy in the present study as well as in the studies by Mitchell et al. [14] and Mitsuhashi et al. [15], it is important to offer EBRT to a wider range of elderly
patients. On the other hand, our data show that it is possible to achieve meaningful tumour control and survival for elderly un®t patients with BT only in tumours up to 5 cm, when the patient is judged unable to sustain neither surgery nor extensive radiotherapy. The patient with stage IIIB disease treated only with BT due to poor performance status underlines this possibility. It may rightfully be argued that BT alone must be considered as a palliative for stage III disease. However, treatment intention was not clearly stated in the medical charts and the patient was therefore included. The patient died free from recurrence 18 months after the start of treatment. Whether the patient was over-staged, followed up for too short a time or actually obtained tumour control remains uncertain. 5. Conclusion Selected elderly patients with advanced cancer of the uterine cervix may tolerate radical radiotherapy consisting of both BT and EBRT with acceptable toxicity and reasonable survival. BT alone may also be a good alternative in early stage disease for the surgically un®t patient. Age per se is not a signi®cant factor for treatment outcome. For very frail patients not even expected to sustain EBRT, it may sometimes be possible to obtain meaningful tumour control even for centrally located tumours as large as 5 cm with BT only. However, combinations of EBRT and BT should always be applied whenever possible unless the tumour is very small. Acknowledgements This work was supported by Direktùr Jacob Madsens og hustru Olga Madsens Fond. References [1] Boyd ME, Groome PA. The morbidity of abdominal hysterectomy. Can. J. Surg. 1993;36:155±159. [2] Chassagne D, Sismondi P, Horiot JC, et al. A glossary for reporting complications of treatment in gynecological cancers. Radiother. Oncol. 1993;26:195±202. [3] Dische S, Warburton MF, Jones D, Lartigau E. The recording of morbidity related to radiotherapy. Radiother. Oncol. 1989;16:103± 108. [4] Fuchtner C, Manetta A, Walker JL, et al. Radical hysterectomy in the
J.C. Lindegaard et al. / Radiotherapy and Oncology 56 (2000) 9±15
[5] [6] [7] [8]
[9] [10]
[11]
[12] [13]
[14] [15] [16]
elderly patient: analysis of morbidity. Am. J. Obstet. Gynecol. 1992;166:593±597. Geisler JP, Geisler HE. Radical hysterectomy in patients 65 years of age and older. Gynecol. Oncol. 1994;53:208±211. Grant PT, Jeffrey JF, Fraser RC, et al. Pelvic radiation therapy for gynecologic malignancies in geriatric patients. Gynecol. Oncol. 1989;33:185±188. Grigsby PW. Primary radiotherapy for stage IB or IIA cervical cancer. J. Natl. Cancer Inst. Monogr. 1996:61±64. Haie-Meder C, Kramar A, Lambin P, et al. Analysis of complications in a prospective randomized trial comparing two brachytherapy low dose rates in cervical carcinomas. Int. J. Radiat. Oncol. Biol. Phys. 1994;29:953±960. Hamberger AD, Fletcher GH, Wharton JT. Results of treatment of early stage I carcinoma of the uterine cervix with intracavitary radium alone. Cancer 1978;41:980±985. Horiot JC, Pigneux J, Pourquier H, et al. Radiotherapy alone in carcinoma of the intact uterine cervix according to G.H. Fletcher guidelines: a French cooperative study of 1383 cases. Int. J. Radiat. Oncol. Biol. Phys. 1988;14:605±611. Jensen PT, Roed H, Engelholm SA, Rosendal F. Pulsed dose rate (PDR) brachytherapy as salvage treatment of locally advanced or recurrent gynecologic cancer. Int. J. Radiat. Oncol. Biol. Phys. 1998;42:1041±1047. Kamby C, Sengelùv L. Assessment of functional status in patients with invasive carcinoma of the urethelial tract. Urol. Oncol. 1996;2:43±51. Kapp KS, Stuecklschweiger GF, Kapp DS, et al. Carcinoma of the cervix: analysis of complications after primary external beam radiation and Ir-192 HDR brachytherapy. Radiother. Oncol. 1997;42:143± 153. Mitchell PA, Waggoner S, Rotmensch J, Mundt AJ. Cervical cancer in the elderly treated with radiotherapy. Gynecol. Oncol. 1998;71:291±298. Mitsuhashi N, Takahashi M, Nozaki M, et al. Squamous cell carcinoma of the uterine cervix: radiation therapy for patients aged 70 and older. Radiology 1995;194:141±145. Patel FD, Sharma SC, Negi PS, Ghoshal S, Gupta BD. Low dose rate vs. high dose rate brachytherapy in the treatment of carcinoma of the uterine cervix: a clinical trial. Int. J. Radiat. Oncol. Biol. Phys. 1994;28:335±341.
15
[17] Pedersen D, Bentzen SM, Overgaard J. Reporting radiotherapeutic complications in patients with uterine cervical cancer. The importance of latency and classi®cation system. Radiother. Oncol. 1993;28:93±186. [18] Pedersen D, Bentzen SM, Overgaard J. Early and late radiotherapeutic morbidity in 442 consecutive patients with locally advanced carcinoma of the uterine cervix. Int. J. Radiat. Oncol. Biol. Phys. 1994;29:941±952. [19] Perez CA. Uterine cervix. In: Perez CA, Brady LW, editors. Principles and practice of radiation oncology, 3rd. ed. Philadelphia, PA: Lippincott-Raven, 1997. pp. 1733±1834. [20] Perez CA, Breaux S, Bedwinek JM, et al. Radiation therapy alone in the treatment of carcinoma of the uterine cervix. II. Analysis of complications. Cancer 1984;54:235±246. [21] Perez CA, Fox S, Lockett MA, et al. Impact of dose in outcome of irradiation alone in carcinoma of the uterine cervix: analysis of two different methods. Int. J. Radiat. Oncol. Biol. Phys. 1991;21:885±898. [22] Perez CA, Grigsby PW, Chao KS, Mutch DG, Lockett MA. Tumor size, irradiation dose, and long-term outcome of carcinoma of uterine cervix. Int. J. Radiat. Oncol. Biol. Phys. 1998;41:307±317. [23] Pignon T, Scalliet P. Radiotherapy in the elderly. Eur. J. Surg. Oncol. 1998;24:407±411. [24] Sablinska B. Carcinoma of the uterine cervix in women over 70 years of age. Gynecol. Oncol. 1979;7:128±135. [25] Scalliet P. Radiotherapy in the elderly. Eur. J. Cancer 1991;27:3. [26] Shuster PA, Barter JF, Potkul RK, Barnes WA, Delgado G. Radical hysterectomy morbidity in relation to age. Obstet. Gynecol. 1991;78:77±79. [27] Sinistrero G, Sismondi P, Rumore A, Zola P. Analysis of complications of cervix carcinoma treated by radiotherapy using the FrancoItalian glossary. Radiother. Oncol. 1993;26:203±211. [28] Stehman FR, Perez CA, Kurrasch M. Uterine cervix. In: Hoskins WJ, Perez CA, Young RC, editors. Principles and practice of gynecologic oncology, 2nd. ed. Philadelphia, PA: Lippincott-Raven, 1997. pp. 785±858. [29] Yancik R, Ries LG. Cancer in the aged: an epidemiologic perspective on treatment issues. Cancer 1991;68:2502. [30] Zachariah B, Balducci L, Venkattaramanabalaji GV, et al. Radiotherapy for cancer patients aged 80 and older: a study of effectiveness and side effects. Int. J. Radiat. Oncol. Biol. Phys. 1997;39:1125±1129.