Introduction
Locally advanced cervical cancer remains a significant cause of mortality and morbidity in low- and middle-income countries (LMICs) like India. Reference Bray, Ferlay, Soerjomataram, Siegel, Torre and Jemal1,Reference LaVigne, Triedman, Randall, Trimble and Viswanathan2 As per International Federation of Gynecology and Obstetrics (FIGO) 2018 staging, Reference Bhatla, Berek and Fredes3 stage IVA cervical cancer is defined as involvement of the mucosa of adjacent structures such as bladder and/or rectum. Prognosis of stage IV cervical cancer is poor, and 5-year overall survival (OS) is estimated to be between 19 and 32%. Reference Bhatla, Berek and Fredes3 OS available from literature is mainly estimated for patients who undergo curative treatment. Reference Chopra, Gupta and Mathew4 However, due to the advanced stage of the disease and poor general condition, many patients undergo only palliative treatment. This implies that alleviation of painful, distressing symptoms are important clinical objectives. Approximately 3% of the cervical cancer patients present with stage IVA disease. Hitherto published literature is sparse for exclusively stage IVA cervical cancer and outcomes are usually reported along with other stages, for example, stage IB3-IVA. Given the paucity of clinical data on treatment and outcome experience of stage IVA cervical cancer, there is a lack of consensus on the management. Distressing symptoms due to local extension of the disease to the urinary bladder or rectum (urinary incontinence, haematuria, complicated urinary tract infections, faecal incontinence, bleeding per rectum and fistula), and pelvic pain are some of the challenging issues. Reference George, Das and Jeba5 The situation is often complicated by poor performance status, obstructive uropathy, bowel obstruction, lymphedema and deep venous thrombosis. Reference Mishra6 Since cervical cancer has a strong correlation with low socio-economic status, it is a challenge to implement evidence-based medicine in LMIC settings. Reference Allende-Pérez, Verástegui-Avilés, Mohar-Betancourt and Herrera-Gómez7,Reference Reichheld, Mukherjee, Rahman, David and Pricilla8 Improved quality of life by achieving optimal local control of disease without undue toxicity remains the overarching goal of treatment. Reference Allende-Pérez, Verástegui-Avilés, Mohar-Betancourt and Herrera-Gómez7 Although public health measures and population-based screening programmes can help in early detection, Reference Reichheld, Mukherjee, Rahman, David and Pricilla8 stage IVA disease remains a significant cause of morbidity and mortality in LMIC settings. This study presents the patterns of care and outcomes of stage IVA cervical cancer from a large tertiary care hospital in India. To the best of our knowledge, this is one of the largest single institutional experience of management of stage IVA cervical cancer.
Materials and Methods
The medical database of 2,476 previously untreated patients diagnosed with carcinoma cervix between January 2005 and December 2018 was retrospectively analysed. Among the 2,476 patients, 96 were diagnosed with stage IVA disease established by either cystoscopy or proctoscopy, and biopsy. The demographics and tumour characteristics were analysed in these 96 patients, while treatment modalities and outcomes were analysed in 92 patients. Four patients who did not undergo treatment at the study centre were excluded. Those who had not followed up for outpatient visits after completing treatment were termed lost to follow-up. Progression-free survival (PFS) was defined as the duration from the date of histological confirmation to the date of clinical or radiological progression. OS was defined as the duration from the date of histological evidence to the last date of outpatient visit or home visit by the palliative care team.
Statistical analyses
Analysis was performed on SPSS Version 20 (IBM Corp, Armonk, NY, USA). Descriptive statistics such as mean, median and interquartile range (IQR) were used to describe demographic variables. Kaplan–Meier curves were generated for survival outcomes. Univariate analysis was done using Log-rank test to assess the impact of variables such as age at diagnosis, differentiation, nodal status, hydroureteronephrosis (HUN) and radiotherapy doses on survival.
Results
Out of 2,476 patients, 96 (3·8%) were diagnosed with stage IVA disease (Figure 1). The mean age of patients in the study was 51·34 years (range 30–84 years). The performance status was Eastern Cooperative Oncology Group (ECOG) 1 in 85 (88·5%) and ECOG 2 in 11 (11·4%) patients. The patient characteristics are detailed in Table 1.
Figure 1. Patient selection and management algorithm.
Table 1. Patient demographics
* n, number of patients.
Baseline characteristics
The predominant presenting symptoms were bleeding per vagina in 49 patients (51%), white discharge per vagina in 21 patients (21·8%) and lower abdomen pain in 20 patients (20·8%). The other symptoms were haematuria, urinary incontinence and difficulty in passing stools. Bladder involvement was present in 85 patients (88·5%), rectal involvement was seen in 7 patients (7·3 %), while both bladder and rectal involvement was seen in 4 patients (4·2%). The commonest histology was squamous cell carcinoma in 84 patients (87·5%), 5 patients (5·2%) had adenocarcinoma, 1 patient (1·05%) had adenosquamous histology, while histology of 6 patients (6·25%) was not available. The grade was well differentiated in 3 (3·12%), moderately differentiated in 60 (62·5%), poorly differentiated in 20 (20·83%), neuroendocrine differentiation in 1 (1·04%) and unknown in 12 patients (12·5%), respectively. The staging modalities utilised were ultrasonography (n = 74, 77%), computed tomography (CT) (n = 36, 37·5%), magnetic resonance imaging (MRI) (n = 7·7·3%) and positron emission tomography-CT (n = 5, 5·2%). Thirty-four patients (35·42%) patients had node positive disease. Eighty patients (83·3%) had HUN of whom 42 (43·75%) had unilateral and 38 (39·58%) had bilateral HUN. The median creatinine was 0·995% mg (IQR: 3·6–5% mg) and the median haemoglobin was 9·25 g/dL (IQR: 8·67–9·82 g/dL). Out of the 96 patients, 23 (23·9%) had a baseline creatinine greater than 1·5% mg.
Treatment modalities
Of the 92 patients, 59 (64·13%) received radiation therapy (RT) alone, 22 (23·9%) chemoradiation (CRT), 3 (3·26%) received neoadjuvant chemotherapy (NACT) followed by RT, 1 (1·08%) received NACT followed by CRT, 4 patients (4·35%) chemotherapy alone, while 3 (3·26%) were offered best supportive care (Table 2).
Table 2. Treatment characteristics
3DCRT, 3-dimensional conformal radiation therapy; IMRT, intensity-modulated radiation therapy; EBRT, external beam radiation therapy; HDR, high dose rate; LDR, low dose rate.
The techniques used in the 85 patients who received RT were conventional in 72 (84·7%) [45 patients (52·9%) cobalt-60 and 27 patients (31·7%) linear accelerator], 3-dimensional conformal radiation therapy (3DCRT) in 10 patients (11·7%), and intensity-modulated radiation therapy (IMRT) in 3 patients (3·53%). The radical RT dose used ranged from 45 to 66 Gy in 25–33 fractions, and the most common dose schedule used was 50 Gy in 25 fractions in 25 patients (29·4%). Among these patients, 3 had received haemostatic radiotherapy, to a dose of 9 Gy in 3 fractions, followed by further RT to achieve an equivalent dose of 50 Gy in 25 fractions. Among the 82 patients who were planned for definitive RT, following completion of 45–50 Gy external beam radiation therapy (EBRT), boost was delivered in 54 patients (65·8%). The various modalities used were EBRT in 37 patients (45·12%), high dose rate (HDR) interstitial brachytherapy in 4 (4·88%), HDR intracavitary brachytherapy in 8 patients (9·76%), low dose rate intracavitary brachytherapy in 5 (6·1%), and in 12 patients (14·6%) details regarding modality of the boost treatment was not available. Among the 23 patients (28·04%) who received CRT with weekly Cisplatin 40 mg/m2, the median number of concurrent chemotherapy cycles was 3 (range 1–6). Acute toxicities noted during EBRT were grade 2 dermatitis in 16 patients (18·8%), grade 3 dermatitis in 10 (11·7%), grade 2 enteritis in 7 (8·2%) and grade 3 neutropenia in 2 patients (2·3%). Palliative RT was administered in three patients (3·52%) with 30 Gy in ten fractions dose schedule. In four patients (4·87%), systemic therapy was given in view of disease progression, ten patients (12·2%) defaulted further treatment and two (2·4%) died during EBRT.
Diversion procedures
In this study, 23 (25%) patients had fistulae out of whom 20 had vesicovaginal fistula (VVF) and 3 had rectovaginal fistula (RVF). Among the patients with fistula, 14 had the fistula pre-treatment, 4 patients developed during treatment and the rest developed post-treatment. Three patients with VVF underwent bilateral percutaneous nephrostomy (PCN), one had unilateral PCN and one underwent catheterisation. Two patients with RVF underwent diversion. These patients were treated with external beam radiotherapy with the most common schedule being 50 Gy in 25 fractions followed by boost dose of 16 Gy in 8 fractions. None of these patients received brachytherapy.
During the treatment period, 36 patients (32·5%) required a diversion procedure, of which 16 (16·6%) underwent bilateral PCN, 7 patients (7·3%) underwent unilateral PCN, 1 patient (1·04%) had bilateral PCN and rectal diversion, 2 (2·08%) had other procedures like dialysis and Double J (DJ) stenting, while 3 patients (3·12%) were catheterised with Foley’s catheter and 7 patients (7·3%) underwent rectal diversion only. PCN blockage and repositioning was one of the major problems in these patients. Four patients had reinsertion of PCN after removal due to increase in the creatinine values and seven patients underwent repositioning of the PCN. Of the 24 patients who underwent PCN insertion, internalisation could be done in 4 patients after completion of treatment and in the remaining, PCNs could not be removed (n = 19) or had to be reinserted (n = 3).
Supportive measures
The pain scores were assessed periodically, and analgesics were titrated according to World Health Organization (WHO) ladder. Details regarding pharmacological interventions for pain were available in 88 patients (91·6%). Among these patients, 85 (96·5%) required WHO step 1 analgesics like Paracetamol, 73 (82·9%) required WHO step 2 analgesics like Tramadol and 28 (31·8%) required WHO step 3 analgesics like Morphine and Buprenorphine. Adjuvant medications were prescribed in 54 patients (61·3%) which included Amitriptyline in 13 (14·7%), Mirtazapine in 26 (29·5%), Haloperidol in 5 (5·68%), Amitriptyline with Mirtazapine in 4 patients (4·54%) and Mirtazapine with Haloperidol in 6 patients (6·8%).
These patients required frequent hospitalisations due to their poor general condition and complications. Sixty-nine patients required in-patient care at least once during the course of treatment. There were 130 admissions in total and the average number of admissions was 2 (range 1–6). The most common cause for admission was PCN insertion and management of PCN-related complications followed by evaluation of fever, blood transfusion for correction of anaemia, correction of electrolyte imbalances and pain management. The other reasons were diversion procedures for rectal infiltration, urinary tract infections and enteritis.
Follow-up and outcomes
Among the 92 patients analysed for survival, the median follow-up was 12 months (2–131 months). Eighty-five patients had come for at least one outpatient visit following completion of treatment and for the remaining 7 (7·69%) who could not come for follow-up, the palliative care team made a home visit.
During the follow-up, 17 patients (18·7%) were found to be disease-free and 47 (51·1%) had disease relapse (Table 3), commonest site being local. For the remaining 28 patients (30·77%), the disease status could not be ascertained and hence they have been censored for the purpose of survival analysis beyond the date of last follow-up.
Table 3. Site of failure
The median PFS was 12 months (95% CI: 9·6–14·4 months) and median OS was 25 months (95% CI: 16·6–33·4 months). The 2-year and 3-year PFS was 30 and 20%, respectively, and the OS was 50 and 32%, respectively. The Kaplan–Meier curves are shown in Figures 2 and 3, respectively.
Figure 2. Progression-free survival (X axis—time in months, Y axis—proportion of patients surviving).
Figure 3. Overall survival(X axis—time in months, Y axis—proportion of patients surviving).
On univariate analysis, we found a statistically significant difference in PFS and OS in patients with better performance status (ECOG 1) and higher dose of RT (>50 Gy). The patients who required PCN insertion were found to have poorer outcomes, median PFS of 10 months and the median OS of 17 months as compared to those who did not have PCN placement, where it was 12 and 28 months respectively, but the difference was not statistically significant (PFS p = 0·13 and OS p = 0·07). Other factors analysed like age, haemoglobin, histology, differentiation, nodal status, site of nodal metastases and HUN did not have a significant impact on survival (Table 4).
Table 4. Univariate analysis of prognostic factors for PFS and OS
SCC, squamous cell carcinoma; PAN, paraaortic node.
Bold values are to highlight Statistical Significance p-Value <0.05.
Discussion
Cervical cancer is a major public health problem in developing countries and poses a significant burden on the social, psychological and economic aspects of an individual. Despite improvements in preventive strategies and treatment modalities, outcomes of locally advanced cervical cancer are still poor. Although radical CRT is the recommended treatment for cervical cancer stages FIGO 2018 IB3-IVA, Reference Shrivastava, Mahantshetty and Engineer9 patients with stage IVA need to be managed on an individualised basis. Reference Mabuchi, Isohashi and Okazawa10 The literature on treatment outcomes in patients with stage IVA exclusively is scarce.
Rose et al. Reference Rose, Ali, Whitney, Lanciano and Stehman11 in their retrospective study of stage IVA cervical cancer patients from four trials (Gynecologic Oncology Group protocols 56, 85, 120 and 165) treated with radiotherapy with or without concurrent cisplatin-based chemotherapy reported 44 patients (91·6%) with bladder involvement and 4 patients (8·3%) with rectal involvement similar to the present study where bladder invasion was found in 85 patients (88%) and rectal involvement was reported in 7 patients (7%).
Though ultrasound abdomen and pelvis was the most common staging investigation in this study, non-contrast MRI (Figure 4a and 4b) is an important diagnostic tool to assess the accurate local staging, nodal status and distant metastases. MRI along with diffusion-weighted imaging and calculation of apparent diffusion coefficient values have a pivotal role in prognostication, treatment planning and detection of early recurrences. It is essential in advanced cervical cancers where most patients present with obstructive uropathy, Reference Nuranna, Antonius, Laily, Kusuma and Nuryanto12 thus precluding the use of CT contrast.
Figure 4. (a) T2-weighted sagittal image of MRI of the pelvis showing bladder infiltration prior to RT (pointed out by blue arrow) (b) T2-weighted sagittal image of MRI of the pelvis showing response post-RT (pointed out by blue arrow).
Espenel et al. Reference Espenel, Garcia and Langrand-Escure13 have described their experience in the treatment of 24 patients with stage IVA disease. Patients received chemotherapy, radiotherapy including EBRT and brachytherapy, and surgery depending on the general condition and treatment response. The 5-year OS was 5·1% for stage IVA cervical cancer patients, and they reported that poor performance status was identified as a poor prognostic factor for disease-specific survival. Similarly, in this study, patients with performance status ECOG 1 had significantly better 3-year PFS (24%, p = 0·001) and 3-year OS (35%, p = 0·05) compared to those with ECOG 2 (0%).
Hata et al. Reference Hata, Koike and Miyagi14 have reported the outcomes of radiotherapy in 28 patients diagnosed with stage IVA cervical cancer. All these patients received EBRT to a median dose of 50·4 Gy in 28 fractions. Similarly, in this study, radiotherapy was the predominant treatment modality (92·4%, n = 85), and we found a statistically significant association with better outcomes among the patients who received a dose more than 50Gy (3-year PFS 24%, p = 0·05 and 3-year OS 36%, p = 0·048). However, patients with better performance status and better response are more likely to have received higher doses of radiotherapy.
Murakami et al. Reference Murakami, Kasamatsu and Morota15 in their analysis of 34 patients reported brachytherapy utilisation in 26 patients (76·5%) following completion of the planned EBRT, while the remaining 8 patients (23·5%) received EBRT boost because of poor response. However, in our study, only 17 patients (20·73%) underwent brachytherapy after EBRT and 37 patients (45·12%) received EBRT boost in view of poor performance status, poor response or persistent bladder or rectal mucosa involvement at the end of EBRT.
In this study, 83·3% of patients had HUN in varying degrees of severity. This meant that only 22 patients (23%) could receive concurrent cisplatin with RT. All the planned cycles of chemotherapy could not be completed either due to worsening renal parameters or deteriorating general condition. Similar findings were reported by Hata et al., Reference Hata, Koike and Miyagi14 where 21% received concurrent chemotherapy.
In patients with obstructive uropathy, PCN placement is one of the interventions which normalises the renal parameters early. However, in patients with carcinoma cervix stage IVA, PCN placement has to be considered only after prognostication and discussion with the patient and caregivers. Reference Mishra, Desai, Patel, Mankad and Dave16 The patients planned for PCN should have access to the health care system for adequate care and regular dressings.
The most distressing symptoms experienced by these patients were urinary and faecal incontinence as a result of fistulae. Reference Moore, Gold, McMeekin and Zorn17,Reference Biewenga, Mutsaerts, Stalpers, Buist, Schilthuis and van der Velden18 Involvement of palliative care team early in the management and decision-making would aid in optimum symptom control and oncological outcomes. Sun et al. Reference Sun, Koubaa, Limkin and Dumas19 reported 32·4% VVF in a cohort of 71 patients with bladder involvement either during or after RT which was slightly higher compared to the present study which reports 23 patients (25%) with fistulae.
The median PFS in the present cohort of patients was 12 months (95% CI: 9·6–14·4 months) and the median OS was 25 months (95% CI: 16·6–33·4 months) which is similar to published literature (10·1 and 21·2 months, respectively). Reference Rose, Ali, Whitney, Lanciano and Stehman11 The 3-year PFS was 20% and 3-year OS was 32% which is comparable to previously published study by Murakami et al. which reported 3-year PFS 28% and 3-year OS 40%. Reference Murakami, Kasamatsu and Morota15 Comparison between various studies in literature Reference Rose, Ali, Whitney, Lanciano and Stehman11,Reference Hata, Koike and Miyagi14,Reference Murakami, Kasamatsu and Morota15,Reference Khulpateea, Paulson, Carlson, Miller and Lea20 is presented in Table 5.
Table 5. Published studies on exclusively stage IVA cervical cancers
Pain is a major symptom in locally advanced cervical cancer, Reference Palat, Biji and Rajagopal21 and various pain syndromes associated with advanced cancers have been described in literature. The WHO step ladder Reference Jadad and Browman22 of pain management guides oncologists to alleviate pain and significantly aids in optimal palliation. To achieve this, palliative care physicians were involved in the primary care, symptom management and counselling of the study patients as part of the institutional disease management protocol. Care of terminally ill, nursing issues should also be explained to the caregivers. There should also be substantial support from the social workers and counsellors within the hospital, from the local administration and other support groups to smoothen and facilitate the process of treatment, follow-up and survivorship.
Limitations: Lack of rigorous follow-up data in view of the study’s retrospective nature is the limitation of the study.
Conclusion
Management of patients with stage IVA cervical cancer is complex, owing to the locally advanced nature of the tumour and a multitude of coexisting factors. RT is the mainstay of treatment with concurrent chemotherapy in carefully selected patients. The treatment needs to be individualised to achieve a fine balance between local control, toxicity and quality of life. Involvement of palliative care team early in the course of treatment adds a holistic approach to the continuum of oncological care.
Supplementary material
For supplementary material accompanying this paper visit https://doi.org/10.1017/S1460396921000443
Acknowledgements
None.
Conflicts of Interest
The authors of this study declare that there are no conflicts of interest or financial disclosures.
