Electronic journal of oncology
Volume 101 • N◦ 2 • février 2014 John Libbey Eurotext
©
Predictors of tumor response in carcinoma of uterine cervix: data from tertiary cancer centre in India Sushmita Pathy1 , Richard Muwonge2 , Subhash Chander1 , Vatsla Dadhwal3 , Jagadeshan Pandjatcharam1,4 1
Article received on July 02, 2013, accepted on September 30, 2013 Reprints: S. Pathy
All India Institute of Medical Sciences, Department of Radiation Oncology, Ansari Nagar, 110029 New Delhi, India 2 International Agency for Research on Cancer, Screening Group, Early Detection and prevention Section, 150, cours Albert Thomas, 69372 Lyon cedex 08, France 3 All India Institute of Medical Sciences, Department of Obstetrics and Gynecology, New Delhi, India 4 Kidwai Memorial Institute of Oncology, Deparment of Radiotherapy, Dr. M.H. Marigowda Road, Bangalore 560029, Karnataka, India
To cite this article: Pathy S, Muwonge R, Chander S, Dadhwal V, Pandjatcharam J. Predictors of tumor response in carcinoma of uterine cervix: data from tertiary cancer centre in India. Bull Cancer 2014 ; 101 : E13-E18. doi : 10.1684/bdc.2014.1895.
Abstract. Aim: The current study was a retrospective analysis to identify the predictors of tumor response among cervical cancer patients treated with chemoradiation and completed the intended treatment. Methods: Hundred and twenty eight patients who completed the intended therapy as per institute protocol evaluated. Patients were treated with external beam radiotherapy to a dose of 50 Gy in 27 fractions followed by 21 Gy in three sessions by intracavitory radiotherapy to point A. Postoperative patients received 16 Gy in two sessions. Predictors of tumor response after treatment were assessed using cox regression. Results: Median time to complete tumor response was 12.2 months. Clinical stage emerged as the independent prognostic
doi : 10.1684/bdc.2014.1895
Introduction Carcinoma of uterine cervix is the third most common cancer in women, and seventh overall, with estimated 530,000 new cases in 2008. More than 85% of the global burden occurs in developing countries with India having a quarter of the world cervical cancer burden [1]. In India the age standardized incidence rates is 17.6 per 100,000 people, with highest incidence in Chennai (28.0) followed by Barshi (27.4) and Delhi (19.5) [2]. The high incidence rates in India could be attributed to low socioeconomic conditions, lack of education, and poor access to health care delivery. Surgery is the mainstay of treatment for early stage I cervical cancers. However, majority of cases have loco-regionally advanced cervical cancer at presentation and can only be managed with chemoradiotherapy in order to have a better outcome in terms of overall survival, local control and prevention from progressing to distant metastasis. Adjuvantchemoradiotherapy is also shown to improve in locally controlling the Bull Cancer vol. 101 • N◦ 2 • février 2014
factor. Patients who completed chemotherapy treatment in fewer than five cycles had a non-significant increased chance of tumor response compared to those completing in five or more cycles. Conclusion: Poor tumor response after treatment is largely explained by advanced disease stage at diagnosis. A strategy for early detection and access to screening facilities will improve outcome. Fewer chemotherapy cycles may be beneficial in patients with locally advanced carcinoma cervix. Key words: carcinoma cervix, chemotherapy, predictors, radiotherapy
disease in the presence of high risk factors particularly close margins, and positive lymph nodes. A meta-analysis including 19 trials supported the use of chemoradiation over radiation alone [3]. Concomitant chemoradiotherapy has therefore been the preferred treatment for locally advanced carcinoma cervix [4-6]. The mechanisms for the improved outcome are due to radio sensitization of tumor cells, cytotoxic action of platinum agents. Nevertheless, a more recent randomized control trial conducted by National cancer Institute of Canada compared chemoradiation to radiation alone in patients with squamous cell carcinoma (locally advanced and early stage with lymph node metastasis) failed to demonstrate any significant difference in survival in the two arms [7]. Despite technical advances the outcome of locally advanced carcinoma cervix is amenable to improvement. Many unanswered questions about treatment of advanced cervical cancer remain. The meta-analysis of earlier concurrent chemotherapy trials suggested that benefit for stage
E13
S. Pathy, et al.
III patients were minimal [8]. This encouraged us to analyse our own data to identify the predictors of tumor response among cervical cancer patients treated with chemoradiation and completed the intended treatment.
Methods Patient characteristics A total of 263 cervical cancer patients were registered in Gynaeoncology clinic of Institute Rotary cancer Hospital, All India Institute of Medical Sciences (AIIMS), from April 2006 to December 2009, out of which 160 patients were evaluable (figure 1). Patients who were referred to our centre after receiving initial treatment elsewhere and those who received palliative radiotherapy were excluded. The present study is an analysis of 128 patients with International Federation of Gynaecology and Obstetrics (FIGO) stage IA-IIIB who were investigated and treated according to the institute protocol, which is detailed below and completed the intended therapy (figure 1). All patients had confirmed histopathological diagnosis and underwent routine haematological, biochemical investigations before initiation of treatment. Chest radiograph, Computerised tomography of abdomen and pelvis, cystoscopy and sigmoidoscopy was done to rule out local spread and distant metastasis. Clinical
staging was made as per FIGO staging system. Information was collected on additional patients’ characteristics and prognostic factors including age at treatment (categorized to 50 years), clinical tumour size (4 cm), lymph nodes (N0 and N+) FIGO clinical stage (I, II, III), histology type (squamous cell, adenocarcinoma and others) haemoglobin level (10 gm/dl) chemotherapy completion (5 cycles) treatment completion time (on time and beyond) and Karnofsky Performance status.
Treatment Details External beam radiotherapy (EBRT) was given using four-field box technique after simulation by telecobalt unit (Theratron 780C) or by 6MV/16 MV photon (Elekta). External Radiotherapy to a dose of 50 Gy in 27 fractions with midline shielding after 40 Gy was given in five and half weeks. Intracavitory radiotherapy (ICRT) was given after one week of completion of EBRT by high-dose-rate (HDR) Microselectron. A total of 21 Gy in three fractions one week apart was given to point A. Each application was made under general anesthesia or spinal anesthesia, with a detail gynaecological evaluation under anesthesia followed by insertion of FletcherSuit applicator. Foleys catheter was inserted in the bladder with 7 cc of radio-opaque contrast. Adequate vaginal packing was made and stability of the applicator was ensured. 2D radiography based planning
263 patients registered
Excluded (n = 103): Received initial treatment elsewhere Received palliative radiotherapy
Evaluable patients (n = 160)
Excluded (n = 32): Not fulfilling inclusion criteria
Patients included in final analysis fulfilling criteria below (n = 128): FIGO clinical stages IA-IIIB; Investigated and treated according to institute protocol; Completed the intended therapy
Figure 1. Shows patients flowchart and methodology of the study.
E14
Bull Cancer vol. 101 • N◦ 2 • février 2014
Predictors of tumor response in carcinoma cervix
was prepared. Maximum, minimum and mean bladder and rectum doses and dose to point A were calculated based on various point doses. Chemotherapy (cisplatinum) was administered concurrently every week at a dose of 40 mg/m2 during the course of external radiotherapy for FIGO stage IIB and above. Chemotherapy was not administered during brachytherapy. Early stage was treated with surgery and adjuvant radiotherapy in the presence of high-risk histopathological features or bulky clinical disease. Postoperative patients after completion of EBRT received ovoids to a dose of 16 Gy in two fractions one week apart.
Evaluation of treatment response Patients were monitored during the course of EBRT and every week with haematological investigations and clinical examination for acute treatment related morbidities. Acute toxic effects include neutropenia, thrombocytopenia electrolyte imbalance and enteritis. Patients who completed treatment were assessed four weeks after completion of treatment. Clinical examination was carried out to assess the presence or absence of residual disease. Subsequently, all patients were called for follow-up every six weeks for the first year, every three months for the second year, every six months for the third year and thereafter every year. Pap smear was done six months after completion of treatment and thereafter every year. Radiological imaging in the form of chest radiograph, abdominal ultrasound and CT scan and biopsy of a recurrent tumour was done in the presence of clinical symptoms necessitating the investigation.
Statistical Analysis Patient characteristics and prognostic factors were presented using medians together with the interquartile range (IQR) for continuous and proportions for categorical variables. The study outcome was tumor response defined by no evidence of disease for patients who had complete response and partial response of at least 50% at four weeks (WHO criteria) and had residual disease during the follow-up. To assess the effect of patient characteristics on tumor response, hazard ratios (HRs) and their 95% confidence intervals (CIs) were obtained using Cox proportional hazard regression analysis. Follow-up time for each patient was calculated from the date of treatment to the date of complete tumor response for patients who had no evidence of disease and date of last follow-up for those who had residual disease. Multivariate regression analysis was done, by including all the patient characteristics, except histology type and Karnofsky performance status, in the same model. Statistical analysis was carried out using STATA software version 12. Bull Cancer vol. 101 • N◦ 2 • février 2014
Ethical clearance and informed consent As this was a retrospective data correction, informed consent and ethical clearance was not required and not taken.
Results The median age of the patients analysed was 50 years (IQ range 44-56). Table 1 shows the distribution of the characteristics and prognostic factors of tumor response of the patients included in the final analysis. Majority of the patients presented in stage IIIB (58%), were squamous cell carcinomas (94%), clinical tumor size of ≥4 cm (65%) with no clinical lymph nodes (83%), and haemoglobin levels above 10 mg/dl (71%). Furthermore, 53% of the patients completed
Volume 101 • N◦ 2 • février 2014 John Libbey Eurotext
©
Predictors of tumor response in carcinoma of uterine cervix: data from tertiary cancer centre in India Sushmita Pathy1 , Richard Muwonge2 , Subhash Chander1 , Vatsla Dadhwal3 , Jagadeshan Pandjatcharam1,4 1
Article received on July 02, 2013, accepted on September 30, 2013 Reprints: S. Pathy
All India Institute of Medical Sciences, Department of Radiation Oncology, Ansari Nagar, 110029 New Delhi, India 2 International Agency for Research on Cancer, Screening Group, Early Detection and prevention Section, 150, cours Albert Thomas, 69372 Lyon cedex 08, France 3 All India Institute of Medical Sciences, Department of Obstetrics and Gynecology, New Delhi, India 4 Kidwai Memorial Institute of Oncology, Deparment of Radiotherapy, Dr. M.H. Marigowda Road, Bangalore 560029, Karnataka, India
To cite this article: Pathy S, Muwonge R, Chander S, Dadhwal V, Pandjatcharam J. Predictors of tumor response in carcinoma of uterine cervix: data from tertiary cancer centre in India. Bull Cancer 2014 ; 101 : E13-E18. doi : 10.1684/bdc.2014.1895.
Abstract. Aim: The current study was a retrospective analysis to identify the predictors of tumor response among cervical cancer patients treated with chemoradiation and completed the intended treatment. Methods: Hundred and twenty eight patients who completed the intended therapy as per institute protocol evaluated. Patients were treated with external beam radiotherapy to a dose of 50 Gy in 27 fractions followed by 21 Gy in three sessions by intracavitory radiotherapy to point A. Postoperative patients received 16 Gy in two sessions. Predictors of tumor response after treatment were assessed using cox regression. Results: Median time to complete tumor response was 12.2 months. Clinical stage emerged as the independent prognostic
doi : 10.1684/bdc.2014.1895
Introduction Carcinoma of uterine cervix is the third most common cancer in women, and seventh overall, with estimated 530,000 new cases in 2008. More than 85% of the global burden occurs in developing countries with India having a quarter of the world cervical cancer burden [1]. In India the age standardized incidence rates is 17.6 per 100,000 people, with highest incidence in Chennai (28.0) followed by Barshi (27.4) and Delhi (19.5) [2]. The high incidence rates in India could be attributed to low socioeconomic conditions, lack of education, and poor access to health care delivery. Surgery is the mainstay of treatment for early stage I cervical cancers. However, majority of cases have loco-regionally advanced cervical cancer at presentation and can only be managed with chemoradiotherapy in order to have a better outcome in terms of overall survival, local control and prevention from progressing to distant metastasis. Adjuvantchemoradiotherapy is also shown to improve in locally controlling the Bull Cancer vol. 101 • N◦ 2 • février 2014
factor. Patients who completed chemotherapy treatment in fewer than five cycles had a non-significant increased chance of tumor response compared to those completing in five or more cycles. Conclusion: Poor tumor response after treatment is largely explained by advanced disease stage at diagnosis. A strategy for early detection and access to screening facilities will improve outcome. Fewer chemotherapy cycles may be beneficial in patients with locally advanced carcinoma cervix. Key words: carcinoma cervix, chemotherapy, predictors, radiotherapy
disease in the presence of high risk factors particularly close margins, and positive lymph nodes. A meta-analysis including 19 trials supported the use of chemoradiation over radiation alone [3]. Concomitant chemoradiotherapy has therefore been the preferred treatment for locally advanced carcinoma cervix [4-6]. The mechanisms for the improved outcome are due to radio sensitization of tumor cells, cytotoxic action of platinum agents. Nevertheless, a more recent randomized control trial conducted by National cancer Institute of Canada compared chemoradiation to radiation alone in patients with squamous cell carcinoma (locally advanced and early stage with lymph node metastasis) failed to demonstrate any significant difference in survival in the two arms [7]. Despite technical advances the outcome of locally advanced carcinoma cervix is amenable to improvement. Many unanswered questions about treatment of advanced cervical cancer remain. The meta-analysis of earlier concurrent chemotherapy trials suggested that benefit for stage
E13
S. Pathy, et al.
III patients were minimal [8]. This encouraged us to analyse our own data to identify the predictors of tumor response among cervical cancer patients treated with chemoradiation and completed the intended treatment.
Methods Patient characteristics A total of 263 cervical cancer patients were registered in Gynaeoncology clinic of Institute Rotary cancer Hospital, All India Institute of Medical Sciences (AIIMS), from April 2006 to December 2009, out of which 160 patients were evaluable (figure 1). Patients who were referred to our centre after receiving initial treatment elsewhere and those who received palliative radiotherapy were excluded. The present study is an analysis of 128 patients with International Federation of Gynaecology and Obstetrics (FIGO) stage IA-IIIB who were investigated and treated according to the institute protocol, which is detailed below and completed the intended therapy (figure 1). All patients had confirmed histopathological diagnosis and underwent routine haematological, biochemical investigations before initiation of treatment. Chest radiograph, Computerised tomography of abdomen and pelvis, cystoscopy and sigmoidoscopy was done to rule out local spread and distant metastasis. Clinical
staging was made as per FIGO staging system. Information was collected on additional patients’ characteristics and prognostic factors including age at treatment (categorized to 50 years), clinical tumour size (4 cm), lymph nodes (N0 and N+) FIGO clinical stage (I, II, III), histology type (squamous cell, adenocarcinoma and others) haemoglobin level (10 gm/dl) chemotherapy completion (5 cycles) treatment completion time (on time and beyond) and Karnofsky Performance status.
Treatment Details External beam radiotherapy (EBRT) was given using four-field box technique after simulation by telecobalt unit (Theratron 780C) or by 6MV/16 MV photon (Elekta). External Radiotherapy to a dose of 50 Gy in 27 fractions with midline shielding after 40 Gy was given in five and half weeks. Intracavitory radiotherapy (ICRT) was given after one week of completion of EBRT by high-dose-rate (HDR) Microselectron. A total of 21 Gy in three fractions one week apart was given to point A. Each application was made under general anesthesia or spinal anesthesia, with a detail gynaecological evaluation under anesthesia followed by insertion of FletcherSuit applicator. Foleys catheter was inserted in the bladder with 7 cc of radio-opaque contrast. Adequate vaginal packing was made and stability of the applicator was ensured. 2D radiography based planning
263 patients registered
Excluded (n = 103): Received initial treatment elsewhere Received palliative radiotherapy
Evaluable patients (n = 160)
Excluded (n = 32): Not fulfilling inclusion criteria
Patients included in final analysis fulfilling criteria below (n = 128): FIGO clinical stages IA-IIIB; Investigated and treated according to institute protocol; Completed the intended therapy
Figure 1. Shows patients flowchart and methodology of the study.
E14
Bull Cancer vol. 101 • N◦ 2 • février 2014
Predictors of tumor response in carcinoma cervix
was prepared. Maximum, minimum and mean bladder and rectum doses and dose to point A were calculated based on various point doses. Chemotherapy (cisplatinum) was administered concurrently every week at a dose of 40 mg/m2 during the course of external radiotherapy for FIGO stage IIB and above. Chemotherapy was not administered during brachytherapy. Early stage was treated with surgery and adjuvant radiotherapy in the presence of high-risk histopathological features or bulky clinical disease. Postoperative patients after completion of EBRT received ovoids to a dose of 16 Gy in two fractions one week apart.
Evaluation of treatment response Patients were monitored during the course of EBRT and every week with haematological investigations and clinical examination for acute treatment related morbidities. Acute toxic effects include neutropenia, thrombocytopenia electrolyte imbalance and enteritis. Patients who completed treatment were assessed four weeks after completion of treatment. Clinical examination was carried out to assess the presence or absence of residual disease. Subsequently, all patients were called for follow-up every six weeks for the first year, every three months for the second year, every six months for the third year and thereafter every year. Pap smear was done six months after completion of treatment and thereafter every year. Radiological imaging in the form of chest radiograph, abdominal ultrasound and CT scan and biopsy of a recurrent tumour was done in the presence of clinical symptoms necessitating the investigation.
Statistical Analysis Patient characteristics and prognostic factors were presented using medians together with the interquartile range (IQR) for continuous and proportions for categorical variables. The study outcome was tumor response defined by no evidence of disease for patients who had complete response and partial response of at least 50% at four weeks (WHO criteria) and had residual disease during the follow-up. To assess the effect of patient characteristics on tumor response, hazard ratios (HRs) and their 95% confidence intervals (CIs) were obtained using Cox proportional hazard regression analysis. Follow-up time for each patient was calculated from the date of treatment to the date of complete tumor response for patients who had no evidence of disease and date of last follow-up for those who had residual disease. Multivariate regression analysis was done, by including all the patient characteristics, except histology type and Karnofsky performance status, in the same model. Statistical analysis was carried out using STATA software version 12. Bull Cancer vol. 101 • N◦ 2 • février 2014
Ethical clearance and informed consent As this was a retrospective data correction, informed consent and ethical clearance was not required and not taken.
Results The median age of the patients analysed was 50 years (IQ range 44-56). Table 1 shows the distribution of the characteristics and prognostic factors of tumor response of the patients included in the final analysis. Majority of the patients presented in stage IIIB (58%), were squamous cell carcinomas (94%), clinical tumor size of ≥4 cm (65%) with no clinical lymph nodes (83%), and haemoglobin levels above 10 mg/dl (71%). Furthermore, 53% of the patients completed
