Carboplatin and paclitaxel for advanced, residual and recurrent cervical carcinoma: a tertiary care centre experience

Introduction

Carcinoma cervix is the second most common cancer among Indian women and most of the women present in an advanced stage^[1]. A recurrence rate of 10%- 20% has been reported following primary surgery or radiotherapy in women with early stage cancer cervix, which goes up to 70% in patients with nodal metastases and/or locally advanced disease^[2,3]. A total pelvic failure rate after radiotherapy alone in stage IB, IIA, IIB, III and IVA is 10%, 17%, 23%, 42% and 74% respectively has been reported. The most common distant recurrence sites are paraaortic nodes (81%), lungs (21%) and supraclavicularnodes (7%)^[4]. The prognosis of patients with advanced, persistent or recurrent disease is poor, with a 1-year survival rate of 15 - 20%^[5]. These cases are treated with systemic chemotherapy often with a palliative intent. Carcinoma cervix is relatively chemo-resistant disease. Till date cisplatin remains the most active and most extensively studied agent in carcinoma cervix with dose ranging from 50mg/m2 to 100mg/m2 every 3 weekly. The response rate (RR) was 50% in chemo naive patients with cisplatin 50 mg/m2 in a Gynecologic Oncology Group (GOG) study^[6].

Various combination chemotherapies have been tried over the years. Moore et al compared cisplatin plus paclitaxel with cisplatin alone in patients with stage IVB, recurrent, or persistent squamous cell carcinoma of the cervix. Response rate was 36% in cisplatin plus paclitaxel group whereas cisplatin alone showed response in 19% of the cases^[5]. Few other authors found that the combination of cisplatin and paclitaxel in advanced or recurrent carcinoma of cervix yielded an objective RR of 46% ^[7,8].

Carboplatin has substituted cisplatin for palliative chemotherapy in cancer cervix. A randomized phase III trial by Kitagawa et al^[9] compared paclitaxel and carboplatin (TC) and paclitaxel and cisplatin (TP) regimen in patients with recurrent and advanced cancer cervix and demonstrated a RR of 58.8% (95% CI, 48.6 to 68.5%) in the TP group and 62.6% (95% CI, 52.3 to 72.2%) in the TC group. In GOG 179, combination of cisplatin and topotecan was compared with cisplatin alone and an improvement in overall survival was seen with combination arm^[10]. Despite the better clinical outcomes, the toxicity profile of these regimens is very important as the objective is largely palliative in these cases.

This study provides the experience with combination chemotherapy with paclitaxel and carboplatin in women with advanced, residual and recurrent cancer cervix in a tertiary care center at Bengaluru, India.

Material and Methods

All patients with advanced, residual and recurrent carcinoma cervix treated carboplatin–paclitaxel combination chemotherapy in the Department of Gynaecologic Oncology from July 2015 to October 2019 have been reviewed.

Carcinoma cervix beyond Stage IIB were considered as advanced disease and only those requiring adjuvant chemotherapy with Paclitaxel and Carboplatin after chemoradiation were included in the study.

Residual disease was defined as presence of disease 12 weeks after completion of chemo-radiation. Disease occurrence 6 months after completion of treatment was or disease occurrence within 6 months with a prior documented complete response was defined as recurrence. Clinical recurrence was confirmed by radiologic investigations and histopathology wherever feasible.

All the cases were re-categorised according to the FIGO staging system 2018 at initial diagnosis.

Patients with Neuroendocrine tumour of cervix, Eastern Cooperative Oncology Group (ECOG) performance status >2 and who have received less than 3 cycles of chemotherapy were excluded from the study.

Before starting chemotherapy, all patients underwent complete physical and gynaecologic examination, assessment of performance status, complete blood count with differential, liver and renal function tests. Glomerular filtration rate (GFR) was measured by 24 h urine creatinine clearance method. Radiologic studies including chest radiograph, ultrasound abdomen and pelvis, chest CT scan in cases of lung or mediastinal metastases, and PET CT whole body were done as indicated and depending on patient affordability. Imaging of the brain and bone was performed when clinically indicated.

Pre-treatment laboratory requirements included leukocyte count ≥3000/µL, neutrophil count ≥1500/µL, platelet count ≥100,000/µL, serum creatinine ≤1.5 mg/ dl, serum bilirubin ≤1.5 times the upper institutional limit of normal, and aspartate aminotransferase and alkaline phosphatase ≤2 times the upper institutional limit of normal.

The chemotherapy regimen consisted of intravenous carboplatin (AUC 6 for chemotherapy naïve patients and AUC 5 for patients who had received to concurrent chemo-radiation) over 1 hour and paclitaxel 175 mg/ m2 over 3-hour infusion every 21 days for 6 cycles after appropriate premedications. Paclitaxel was administered prior to carboplatin. The carboplatin dose was calculated using Calvert formula. Complete blood count was repeated weekly during treatment. Physical examination, renal and liver function tests, performance status and toxicity evaluation were performed before each cycle. If white blood cell had decreased to <3000/µL and platelet count <100,000/ µL on the first day of the next cycle, chemotherapy was withheld until hematologic recovery of white blood cell to <3000/µL and platelet count of <100,000/ µL. An episode of febrile neutropenia at any time in the treatment course required 20% dose reduction for subsequent cycles. The treatment was discontinued if progressive disease or, unacceptable toxicity and in case of patient’s preference.

Response to treatment was assessed clinically after 3 cycles of chemotherapy. Radiological assessment was done whenever possible. Imaging was not feasible in all patients due to cost constraints. Complete response (CR) was defined as the disappearance of all clinical evidence of disease. Partial response (PR) was defined as symptomatic improvement with reduction in volume of the disease. Progressive disease (PD) was defined as deterioration of symptoms and increase in volume of the disease. Stable disease (SD) was any response not applicable to CR, PR, and PD (symptomatic improvement with no change in volume of the disease). Survival was assessed from the date of post-recurrence or diagnosis with stage IV disease or initial diagnosis for residual disease to last follow up or, death due to any cause. Symptom free interval was measured from the completion of treatment to relapse or, last follow up. Toxicity related to treatment was graded according to the NCI Common Terminology Criteria for Adverse Events, Version 3.0.

Results

A total of 50 women with advanced, residual and recurrent carcinoma cervix were treated with carboplatin – paclitaxel protocol during this period. Out of 50 women, 16 women had taken less than three cycles of chemotherapy and three women are still on this treatment. This analysis was performed on 31 women who had completed minimum three cycles of chemotherapy. Patient characteristics are listed in Table 1.

There were nine women with advanced disease, five with residual disease and 17 women with recurrent disease. Among 16 women who had received less than three cycles of chemotherapy and excluded from the analysis; 10 women had recurrent disease, four had advanced and two had residual disease. Among these patients, five patients completed treatment elsewhere, six women were lost to follow up and four women discontinued chemotherapy due to financial constraints.

Among the advanced disease, stage IV A, IV B and III C2(r) were three, five and one women respectively. In stage IVA disease group, two patients received palliative radiotherapy. One patient with stage IV B with right psoas collection positive for malignancy also received palliative radiotherapy (30Gy/10#). Another patient with stage IV B (inguinal node positive) received chemo-radiation with cisplatin and brachytherapy after paclitaxel and carboplatin chemotherapy (Table 2).

Boost to para-aortic region was given to patient with enlarged para-aortic lymph nodes. Residual disease was diagnosed in patients who had received complete chemoradiation with cisplatin and brachytherapy. One patient with residual disease underwent hysterectomy after paclitaxel-carboplatin chemotherapy. Details of treatment in recurrent disease were shown in table 3.

In the recurrent group, most patients (11/17) had pelvic recurrence within the radiated field. Five patients had distal recurrence and one patient had both regional and distal recurrence.

A total of 153 courses of chemotherapy were given (median - 6 courses with range 3-7). Majority of the patients received paclitaxel with dose of 175 mg/m2 over 3-hour infusion and Carboplatin was AUC 6 for stage IV disease and AUC 5 for residual and recurrent patients. During the entire treatment course, six patients had a treatment delay by two weeks.

Clinical response to treatment was assessed after 3 cycles of chemotherapy. All patients had imaging (either ultrasound, CT scan, MRI or PET CT) before starting the chemotherapy. Response assessment could be done in 28 patients as three patients were non-compliant to follow up. In 7 (25%) patients, PET CT was done after chemotherapy. In rest of the patients, response assessment was clinical due to financial constraints. The overall RR was 85.7%. CR and PR were equal (12/28, 42.5%) as depicted in table 4.

All complete responders in recurrent groups (6) were previously irradiated. Among these 6 patients, 5 had disease only in the previously irradiated field and one had peritoneal carcinomatosis with enlarged retroperitoneal lymph nodes. Two patients (2/9) with stage IV B disease had complete response. Among partial responders (5) in recurrent group, all are previously irradiated (3 had disease within and 2 had disease outside the irradiated field. Three patients (10.7%) demonstrated stabilization of disease (All had local recurrence, 2- previously irradiated and 1-no radiation). One patient progressed while on chemotherapy (peritoneal metastaes). She was clinical stage I B2 at initial diagnosis and underwent radical hysterectomy with pelvic and para-aortic lymph node dissection. Histopathology showed para-aortic lymph node positive and hence, she received complete concurrent chemo-radiation including paraaorticirradiation.

The median survival from date of diagnosis till last follow up/death for the entire group was 18.5 months (95% CI, 14.4 - 19.67 months). The median survivals in recurrent and Stage IV disease were 11 months (95% CI, 5 – 17 months) and 18.5 months (95% CI, 10.7 – 26.3 months) respectively. The median disease free survival for patients with residual disease was 26 months (95% CI, 21.8-30.2 months). The median symptom free survival in complete responders was 13 months (95% CI, 8 - 18 months).

Combination chemotherapy with paclitaxel and carboplatin is well tolerated. The toxicities are summarized in table 5. The toxicities were graded according to the worst toxicity occurred in a patient during entire treatment period. The most frequently observed toxicity was anaemia (96.77%, grade 1-11, grade 2 - 17 and grade 3 -2). Majority of the women had haemoglobin less than 12 gm/dl before starting treatment.

Febrile neutropenia was observed in only one patient and was managed with Granulocyte coloby stimulation factor (G-CSF), antibiotics and other supportive management. Other four patients had grade 4 neutropenia during nadir, but it recovered prior to chemotherapy. Among these four patients, none of them developed febrile neutropenia. As there was hematological recovery spontaneously, no G-CSF support was given and no dose reduction was required.

Discussion

The life expectancy of patients with advanced or recurrent cervical cancer is significantly reduced and less than 20% survive for 1 year ^[5]. The goal of treatment in these patients is only palliation and to better the quality of life. Systemic chemotherapy is often required. However, administration of chemotherapy to this group of patient is quite challenging.

Cervical cancer has a high response rate to taxane and platinum agents^[5]. However, response to chemotherapy might be reduced in patients who would have received prior pelvic radiation, limiting bone marrow reserve and reduced chemotherapy dose. Previously irradiated tumour tissue will have compromised blood supply and hence reduced drug delivery. Some patients will have obstructive uropathy and/or, compromised renal function which will further restrict or preclude the use of drugs such as platinum agents. Prior exposure to cisplatin with radiation might cause resistance to platinum agents too.

Paclitaxel and carboplatin have been tried in several studies for the management of advanced, recurrent (not amenable for surgery) and residual disease of cancer cervix. Our study demonstrated an overall response in 85.7% of patients. CR and PR, both were equal (12/28, 42.5%). Similar results have been demonstrated by a Japanese study who have studied combination chemotherapy with paclitaxel 135 to 180 mg/m2 as a 3-hour intravenous infusion followed by carboplatin (AUC 3-5) over 1 hour in 62 patients younger than 70 years and demonstrated an RR of 78%^[11].

A study including 25 women with advanced or recurrent carcinoma of the cervix treated with carboplatin (AUC 5-6) and paclitaxel 155 to 175 mg/m2 and reported an overall RR of 40% (20% PR and 20% CR) which is significantly less as compared to this study^[12]. 14/25 patients in this study had progressive disease. In another small retrospective study, 15 patients with recurrent or persistent disease were treated with carboplatin and paclitaxel and demonstrated an overall clinical RR of 60% with CR of 26.7% and PR of 33.3%^[13] . In a study which identified 62 patients with stage IVB, recurrent or persistent cancer of cervix from tumour registry who were treated with either cisplatin and paclitaxel or carboplatin and paclitaxel, the objective response was observed in 53% of patients who received pactitaxel and carboplatin. RR was 51% in patients who were previously treated with chemoradiation which is lesser than RR occurred in the present study^[14]. Another study demonstrated an overall RR of 53% (CR - 16% and PR - 37%)^[15].

Recently, a phase III randomized trial evaluated the clinical benefits of paclitaxel plus carboplatin with conventional paclitaxel plus cisplatin in 253 patients with metastatic or recurrent cervical cancer. CR rate was 3.9% in the TP group and 7.1% in the TC group.

The overall RR was 58.8% (95% CI, 48.6 to 68.5%) in the TP group and 62.6% (95% CI, 52.3 to 72.2%) in the TC group[9]. The site of recurrence whether it is inside or outside the previously irradiated field has been reported to be significantly associated with response to chemotherapy^[16]. In our series response to chemotherapy was observed in 73.34% (11/15) of patients with recurrent disease. All patients with recurrent disease were previously irradiated. CR was observed in 40%, (6/15). Among these 6 complete responders, 5 had disease only in the previously irradiated field and one had peritoneal carcinomatosis and enlarged retroperitoneal lymph nodes. Among partial responders (5) in recurrent group, 3 had disease within and 2 had disease outside the irradiated field. In contrast a study^[15] demonstrated that 68% of the responding patients and all patients who had a CR had disease outside the irradiated field, where as in those who had disease inside the irradiated field, the RR was 30%. Prognostic factors in cervical cancer patients (190) treated with chemotherapy were studied and the RR for patients with disease outside the irradiated field was 25.3% whereas RR was only 5.3% if the disease was inside the irradiated field^[16]. Unlike above reports, few other studies failed to detect a difference in RR between those with and without prior concurrent treatment in patients with recurrent cervical cancer treated with cisplatin and vinorelbine and carboplatin and paclitaxel, respectively^[14,17].

The median survival was 18.5 months in this study. These figures are similar to other studies^{[9, 12, 13]}. Overall survival was significantly lesser (11-13 months) in some studies unlike the present study^{[14, 15]}. The symptom free interval for complete responders in this study was 13 months which is lower than the progression free survival (PFS) reported elsewhere (16 months)^[12]. There was no significant change in performance status during the treatment in the present study.

This study showed that chemotherapy with a combination of carboplatin and paclitaxel is a well-tolerated regimen, with the most common toxicity being hematologic. The most common toxicity was anaemia as most of the women were anaemic before starting chemotherapy. According to National Family Health Survey - 4, more than 50% of Indian women (age 15-49 years) are anaemic all the time^[18]. The most significant toxicity was grade 3 or 4 neutropenia (35.48%) which is in agreement with other studies^[15]. Another study reported grade 3 to 4 neutropenia in 76.2% of patients which is much higher than this study^[9]. Treatment for febrile neutropenia was required in only one patient. Gastrointestinal toxicities were very few in this series. This may be because of adequate support with proper premedication, antiemetics and proton pump inhibitors after chemotherapy. There was no significant neurotoxicity/ nephrotoxicity/ hepatotoxicity and allergic reactions in our patients. No treatment-related deaths were recorded in this series.

The drawbacks of this study are that it is retrospective with small sample size. It is also limited by clinical response assessment (75%) which is not an accuratemethod. Despite the fact that all patients in residual and recurrent disease group had received prior radiotherapy (63.6% - concurrent chemoradiation), this study demonstrated an overall RR of 85.7%, with equal number of complete and partial responses.

Conclusion

Combination chemotherapy with carboplatin and paclitaxel has reasonable activity and is well tolerated in patients with previously treated with concurrent chemoradiation. These chemotherapeutic agents can be delivered in full dose and on schedule with a manageable toxicity profile. Given the fact that the combination of paclitaxel and carboplatin is being administered with a palliative intent, this regimen shows a good response and survival in patients with advanced, residual, or recurrent cancer cervix.