Results of concurrent chemotherapy and hyperthermia in patients with recurrent cervical cancer after previous chemoradiation

Abstract

Background: Concomitant hyperthermia has been shown to improve response rate after cisplatin in recurrent cervical cancer in previously irradiated patients. It is unclear whether similar response rates can be obtained in patients with a recurrence after previous platinum-containing chemoradiation.

Objective: This study aimed to evaluate the outcome of cisplatin-based chemotherapy with concurrent hyperthermia in patients with recurrent cervical cancer after radiotherapy and cisplatin.

Methods: Patients with recurrent cervical cancer after cisplatin-based chemoradiation or neoadjuvant chemotherapy followed by surgery and radiotherapy who were treated with concurrent platinum-based chemotherapy and hyperthermia were eligible for this retrospective analysis. All patients received six or eight weekly platinum-based chemotherapy cycles in combination with six or eight weekly hyperthermia sessions. The time-to-event variables were estimated using Kaplan-Meier analysis. P-values less than 0.05 were considered significant.

Results: All 38 evaluable patients were selected from the hyperthermia database in the Academic Medical Centre (Amsterdam) and the Erasmus Medical Centre (Rotterdam). Mean age at relapse was 45.7 years (range 27–74). Median time to recurrence after first-line treatment was 15 months. A total of 27 patients had a local and/or regional recurrence; 11 had disease beyond the pelvis. All planned courses of cisplatin chemotherapy and hyperthermia were administered in 17/38 patients. Median follow-up was 6.5 months. One patient died during treatment; response rate was 4/37 (14%), with one complete response. Overall survival was 23% at 12 months and 4% at 24 months. The incidence of grade 3–4 haematological complications did not exceed 10%.

Conclusion: In this retrospective study, concurrent cisplatin and hyperthermia after first-line cisplatin-containing chemoradiation showed poor response and survival. We do not recommend this treatment for recurrence of locally advanced cervical cancer.

• Chemoradiation
• chemotherapy and hyperthermia
• recurrent cervical cancer

Introduction

Cervical cancer is the third most common type of cancer worldwide in women, with an annual worldwide mortality of 275,000 [1,2]. Standard treatment for patients with cervical cancer FIGO stage IIB or higher is cisplatin-based chemoradiation or concurrent radiotherapy and hyperthermia [3,4]. Patients with recurrent disease have a poor prognosis with 1-year survival rates between 15 and 20% [5]. When disease recurs in previously irradiated areas, the prognosis is particularly poor [6]. Standard treatment of recurrent disease in previously irradiated areas consists of platinum-based chemotherapy and hyperthermia in patients who were not previously treated with chemotherapy [6–9]. Several studies investigated the benefit of palliative chemotherapy for recurrent disease. Median overall survival (OS) and progression-free survival (PFS) were improved by adding topotecan or paclitaxel to cisplatin compared to cisplatin monotherapy in two phase III studies. PFS increased from 2.9 to 4.6 months, and OS from 6.5 to 9.4 months with the addition of topotecan [10]. PFS increased from 2.8 to 4.9 months with paclitaxel [11], but the difference in OS was not significant [10]. Another phase III study could not show any significant differences between four platinum-based chemotherapy combination regimens: paclitaxel/cisplatin, topotecan/cisplatin, vinorelbine/cisplatin or gemcitabine/cisplatin [6]. Response rates were 22–29%, OS was 10–13 months and PFS was 4–5.8 months. Recently, the data of bevacizumab added to paclitaxel/cisplatin or topotecan/cisplatin were presented, with improved OS from 13.3 to 17 months for the bevacizumab-containing arms [12]. Two studies, one prospective and one retrospective, in patients with recurrent cervical cancer after radiotherapy published in 1997 showed that adding deep hyperthermia to cisplatin resulted in a high response rate of 55% with an acceptable toxicity [3,13]. The combined treatment has a higher cytotoxic synergism, without increasing side effects on bone marrow, thus providing an enhanced cell kill mechanism in the tumour [14]. Hyperthermia improves the cellular uptake of cisplatin and thereby increases platinum–DNA adduct formation, thereby enhancing the effect of cisplatin with a factor of 4 to 8, depending on temperature level and duration of treatment [3]. Cisplatin and hyperthermia became a standard treatment option for recurrent cervical cancer after previous radiotherapy in the Netherlands after two phase I/II studies [13,14]. We wondered about the efficacy of cisplatin and hyperthermia after cisplatin became part of first-line treatment with the worldwide adoption of chemoradiation as standard therapy for locally advanced cervical cancer in 2000. In the previous studies on hyperthermia with cisplatin, only eight out of 47 patients had been pre-treated with chemoradiation, whereas nowadays it must be over 95% of patients eligible for combination treatment. Therefore, we retrospectively studied the outcome of cisplatin-based chemotherapy with concurrent hyperthermia in patients with a first recurrence of cervical cancer after previous platinum-based therapy. In this paper we report on the palliative effect on local control, overall survival, response, feasibility and toxicity of these patients.

Materials and methods

Patient selection

All patients with recurrent cervical cancer who were treated with concurrent cisplatin and hyperthermia in the Academic Medical Centre, Amsterdam (AMC) and the Erasmus Medical Centre, Rotterdam (EMC) between July 1988 and December 2011 were selected from the database. Included were patients treated with platinum-based therapy as first-line treatment, either as chemoradiation or as neoadjuvant chemotherapy followed by surgery and radiotherapy. Routine pre-treatment work-up consisted of gynaecological examination under general anaesthesia and computed tomography (CT) or magnetic resonance imaging (MRI) scan of chest and abdomen. For staging, the guideline of the International Federation of Gynaecology and Obstetrics [FIGO, 1988] was used.

Chemotherapy

All patients were treated with concurrent cisplatin and hyperthermia in a weekly schedule according to hospital policy (AMC 50 mg/m² in eight schedules and EMC 70 mg/m² in six schedules). Cisplatin was dissolved in 3% NaCl and administered over 90 min, preferably during hyperthermia. All patients were hydrated with 1 L of isotonic saline over 1–2 h before hyperthermia and cisplatin (which was administered in 250 mL hypertonic 3% saline), followed by 2500 mL of isotonic saline with potassium and magnesium supplementation over 16 h. Further supportive care included anti-emetics according to institutional protocol. Patients who in addition to cisplatin-hyperthermia received lapatinib (within the realm of a clinical trial) were also included. Toxicity was scored during treatment. When grade 2–4 toxicity occurred, treatment was delayed or discontinued.

Deep loco-regional hyperthermia

Patients were scheduled for six weekly hyperthermia sessions during the period of cisplatin-based chemotherapy in the EMC and eight weekly sessions in the AMC. In the AMC the 4 Waveguide applicator (custom built) was used and in the EMC the BSD-2000 3D system (BSD Medical Systems, Salt Lake City, UT). For thermometry, Bowman probes were placed intraluminally in the bladder, vagina and rectum with closed-tip catheters. Thermal mapping along the catheters was performed every 5 min with a step size of 1 cm with a maximum map length of 14 cm. Pulse rate and blood pressure were automatically measured before, and every 5 min during treatment, and oral temperature was measured at 0, 15, 30, 60 and 90 min. Heating was started at a power output of 400 W at 77 MHz. Patients were carefully instructed to report any discomfort due to too high temperatures in normal tissue during treatment. Treatment settings for power, phase and frequency were adjusted accordingly if complaints occurred. If no complaints occurred, 100 W was added to the power output every 5 min. The treatment’s objective was to achieve intraluminal temperatures of 40–43°C as homogeneously as possible. In all patients 90 min were scheduled for each hyperthermia treatment: 30 min of heating up and 60 min of actual treatment time [3].

Response evaluation

Response was evaluated after four and eight courses in AMC, and after three and six courses in EMC. This was done by subsequent gynaecological examinations, CT scans and general medical examinations, in accordance with the guidelines of the Dutch Association of Comprehensive Cancer Centres. When patients developed complaints during treatment an earlier examination was done by gynaecological examination or CT scan. Responses noted in this study are the best responses achieved after or during treatment. Response was noted as progressive disease (PD), stable disease (SD), partial response (PR), and complete response (CR), according to WHO criteria. Any surgical treatment options were re-evaluated after treatment. Follow-up data were gathered by hand searching patient charts.

Statistical considerations

End points of this study were toxicity, tumour response, follow-up, local control and overall survival (OS). Cox regression analysis was used to estimate parameters of age and tumour stage on outcome of survival. The time-to-event variables were estimated using Kaplan-Meier analysis. P-values less than 0.05 were considered significant. Analyses were performed with PASW Statistics 20.0.

Results

Patient characteristics

Between July 1988 and December 2011, 38 patients with recurrent cervical cancer were treated with concurrent cisplatin-based chemotherapy and hyperthermia and evaluable for analysis: 22 patients from the AMC and 16 patients from the EMC. Primary treatment was chemoradiation with cisplatin in 27 patients. Three patients had neoadjuvant platinum-based chemotherapy followed by surgery and subsequent radiotherapy, and eight patients had lymph node debulking surgery followed by chemoradiation. The mean age of patients at first diagnosis was 44.2 years (range 26–73 years). Mean age of patients during relapse was 45.7 years (range 27–74 years). After first-line treatment, 29 out of 38 patients (76.3 %) had a complete response; all others had a partial response. The median time until progression or recurrence of disease after first-line treatment was 15 months (range–62). A total of 21 patients had loco-regional recurrence; the 17 other patients had lymph node metastases alone or with distant metastases. Patient characteristics are summarised in Table 1.

Table 1. Patients characteristics of 38 women receiving concurrent cisplatin and hyperthermia for recurrent cervical cancer after chemoradiotherapy.

	N (Range)
Age at primary diagnose in years	44.2 (26–73)
Age at relapse in years	45.7 (27–74)
Time to recurrence in months	15 (3–62)
Histological type
Planocellulair carcinoma	34
Adenocarcinoma	4
FIGO 1988 stage primary tumour
IAII	1
IBI	3
IBII	1
IIA	4
IIB	15
IIIA	2
IIIB	9
IVA	2
IVB	1
Previous treatment
Chemoradiation	27
Neoadjuvant chemotherapy and radiotherapy	3
Surgery before neoadjuvant chemotherapy   and radiotherapy or chemoradiation	8
Site of recurrence
Loco-regional	21
Iliac nodes	6
Para-aortic nodes	7
Regional and distant	4

Chemotherapy and hyperthermia treatment characteristics and adverse events

All patients received platinum-based chemotherapy for recurrence, according to local hospital policy. Treatment characteristics are summarised in Table 2. The aim was to administer six or eight hyperthermia sessions during chemotherapy depending on local hospital policy. Only four of 16 patients received the planned six sessions of chemotherapy in the EMC whereas 14 of 22 patients received the planned eight sessions of chemotherapy in the AMC, all other patients could not receive the planned chemotherapy courses, because of toxicity or PD. Treatment was discontinued when patients clearly progressed during treatment. Due to renal toxicity or severe side effects due to cisplatin, a switch to carboplatin was made in six patients. Nine patients had delay of chemotherapy cycles because of fatigue (n = 3) and haematological toxicity (n = 6). In six patients lapatinib was added to cisplatin because of participation in a clinical trial. All patients were evaluable for toxicity. Grade 3–4 toxicity was observed in 8/38 patients, mostly haematological toxicity (n = 5), ototoxicity (n = 1) and impaired renal function loss (n = 2). In total, 25 patients stopped treatment prematurely because of no further benefit (n = 3), toxicity (n = 6), patient refusal (n = 2) and PD (n = 14). No hyperthermia-specific toxicity was observed. One unknown cause of death was recorded during treatment with cisplatin and hyperthermia.

Table 2. Treatment characteristics of 38 women receiving concurrent cisplatin and hyperthermia for recurrent cervical cancer after chemoradiotherapy.

	%	N
Chemotherapy dose intensity*	100	17
	87.5	6
	75	4
	67	3
	50	5
	33	2
	16	1
Hyperthermia cycles completed	100	12
	87.5	5
	66	8
	50	10
	<33	3

*100%, six or eight courses.

Response rate and survival

The remaining 37 patients were evaluable for tumour response. The response rates with concurrent chemotherapy and hyperthermia are summarised in Table 3. PD occurred in 14 patients during treatment, and five patients had PD within 3 months after cessation of treatment. SD was seen in 13 patients, five patients had a PR and one patient a CR. Difference in response on second-line treatment in patients with PR after first-line treatment was PD in seven out of nine patients. Response in patients with CR after first-line treatment was PD in 12 out of 29 patients. No difference was found in response between patients with different site of recurrence.

Table 3. Clinical response of concurrent cisplatin and hyperthermia in 38 women with recurrent cervical cancer after chemoradiotherapy.

	After first treatment (chemoradiation)	After concurrent chemotherapy and hyperthermia
CR	29 (76%)	1 (3%)
PR	9 (24%)	4 (11%)
SD	–	13 (35%)
PD	–	19 (51%)
Total	38 (100%)	37 (100%)

CR, complete response; PR, partial response; PD, progressive disease; SD, stable disease.

OS was 23% at 12 months and 4% at 24 months (Figure 1). Median time of follow-up was 6.5 months. The median time until progression of disease after start of concurrent cisplatin and hyperthermia was 3.6 months (range 0–13.9). Median OS after treatment was 6.6 months (range 0.3–23.7). All patients finally died of disease. Four patients who responded to therapy underwent surgery after the end of treatment with concurrent cisplatin and hyperthermia, but later developed recurrent disease and died of cervical cancer. Cox regression analysis showed that neither FIGO stage at diagnosis nor age had any significant influence on OS with p-values of 0.549 and 0.197 respectively.

Figure 1. Overall survival of 38 women with recurrent cervical cancer.

Discussion

The problem we pose is a serious problem since recurrence of cervical cancer occurs in 10–74% of cervical cancer patients depending on several risk factors. Standard treatment for loco-regional recurrence after surgery remains chemoradiation, depending on the type of primary therapy, site of recurrence, the disease-free interval, patient symptoms and performance status [15]. In previously unirradiated patients with a loco-regional recurrence, we obtained a 2-year OS and PFS of 66% and 62% [16]. However, loco-regional recurrent cervical carcinoma in a previously irradiated area is a major clinical problem. Several treatment options have been tried. A previous study published in the Lancet showed that good palliative results could be achieved with radiotherapy and hyperthermia for recurrent rectal cancer with a response rate of 74% for the duration of median 6 months and mild toxicity. Therefore, re-irradiation and hyperthermia could be a therapeutic option [17,18].

In our hospitals we treated recurrent cervical cancer patients with platinum-based chemotherapy with concurrent deep hyperthermia. The present retrospective study was designed to determine the palliative effect of this treatment regimen after the introduction of cisplatin in first-line treatment. In contrast to previous studies, outcome was poor with OS at 12 months of only 23%. A study of the Amsterdam group in 1997 already investigated the use of concurrent cisplatin and hyperthermia in 23 previously irradiated patients with recurrent cervical cancer. An overall response rate of 52% was found, showing a benefit of adding hyperthermia to cisplatin [14]. Patients were treated with weekly cycles of loco-regional hyperthermia and cisplatin, for a maximum of 12 cycles. One-year survival was 42%. Another phase I/II study from Rotterdam in 1999 showed a response rate of 50% in 19 patients. An update from this study published in 2007 confirmed the favourably high response rate of 55% [3]. However, in both studies platinum-based chemoradiation was not included in first-line treatment. In a later report published by the Amsterdam group, carboplatin and hyperthermia were used in patients with recurrent and metastatic cervical cancer, which resulted in lower response rates of 30%. This was attributed to the fact that cisplatin might be more effective in the treatment of cervical cancer than carboplatin [15]. Several other chemotherapy combination regimens have been tested with the aim of improving response and OS, however, higher response rates came at the cost of increased toxicity, whereas OS did not improve until recently. Combination chemotherapy with topotecan and cisplatin showed both a statistically significant improved response rate and OS benefit in patients, of whom 58% received previous platinum-based chemoradiation. A new ongoing clinical trial presented the results of adding bevacizumab to paclitaxel/cisplatin or topotecan/cisplatin chemotherapy; OS was improved from 13.3 to 17 months [12].

Our study does face some biases. First, not all patients with recurrent disease have been included, only patients with symptoms, and only patients with adequate renal function were selected for treatment with concurrent cisplatin-based chemotherapy and hyperthermia. Unfortunately we have no information on the patients that were not selected. Nevertheless, it is not likely that the response in the selected patients would be worse than in the unselected patients, who were not referred for the combination treatment. There is also heterogeneity, because first-line treatment consisted of platinum-based chemoradiation or neoadjuvant platinum-based chemotherapy followed by radiotherapy. Thirdly the response to first-line treatment differed; although all patients had a response after first-line treatment, not all patients had a complete response. It may be possible that a short interval between first-line treatment and second-line treatment with concurrent chemotherapy and hyperthermia biased the results negatively.

There is no fair comparison possible between our results and that from other studies using different chemotherapy regimens because concurrent chemotherapy and hyperthermia is not a standard of care outside of the Netherlands and there are no papers that describe this kind of treatment after first-line chemoradiation with cisplatin. To our knowledge this is the first paper to analyse the outcome of concurrent platinum-based chemotherapy and hyperthermia after previous platinum-based treatment. The poor results shown here confirm the urgent need for effective strategies for recurrent cervical cancer. These may include: re-irradiation with concurrent hyperthermia, vaccination, the use of antibodies, and chemotherapy with new targeted agents.

Declaration of interest

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.