Contribution of Tata Memorial Centre, India, to cervical cancer care: Journey of two decades

Cervical cancer is one of the major health problems and the second most common cancer among women in India. The number of new cases of cervical cancer in India in 2020 was 1,23,907, contributing to 18.3 per cent of all cancers in women1. Although the incidence of cervical cancer continues to decrease, most women are still diagnosed in a locally advanced stage due to the lack of population-based screening programmes in India2. In the past two decades, Tata Memorial Centre in India has made substantial contributions towards advancing education and research related to prevention, vaccination and therapeutic intervention for cervical cancer and continues to conduct clinical and community trials that may be practice-changing worldwide3. Here we summarized the contribution of Tata Memorial Centre towards the development of key evidence for various aspects of clinical management of pre-invasive, invasive and recurrent cervical cancer. The review also highlights the role of Tata Memorial Centre in shaping a broader policy on the detection and management of cervical cancer including contribution towards resource assessment and building needs for education and research nationally and internationally.

Contribution of Tata Memorial Centre

Early detection and prevention of cervical cancer: Cervical cancer screening is one of the time-tested strategies known to reduce mortality from cervical cancer4. However, wide-scale population-level screening programmes have been challenging to implement in India and other low and middle-income countries (LMICs). This has been attributed not only to the cost involved (like in HPV DNA tests) but also the need to have trained and skilled personnel to report on cytology5. Low-cost and high-fidelity techniques such as visual inspection with acetic acid (VIA) have been tested at the Tata Memorial Centre from 1998–2011 within a cluster phase III randomized trial in 142,701 women aged 30-59 yr to assess the efficacy of screening with VIA in cervical cancers6. The procedure of VIA was performed by primary healthcare workers [unlike trained cytologists or pathologists needed for Papanicolaou (Pap) staining]. The clusters were randomized to the screening group (four rounds of cancer education and screening with VIA at 24 months interval) (n=75,360) and the control group (cancer education at entry alone) (n=76,178). The screening group showed a significant 31 per cent reduction in cervical cancer mortality. The reduction in mortality achieved by VIA (a low cost and resource screening test) suggests that this may be a good low-cost option for both high- and low-income countries and LMICs7.

In another cluster-randomized study of 131,746 women from 52 villages in India, HPV DNA test, VIA, cytology and standard of care for the existing healthcare system were compared. The results of the cluster-randomized study concluded that even a single round of HPV DNA testing was associated with a significant reduction in the number of advanced cervical cancers and deaths from cervical cancer8. This study, for the first time has demonstrated that if resources are limited, even a single round of testing can effectively reduce mortality. This is a major knowledge advancement as population-based screening programme for cervical cancer necessitate repeat testing at predefined intervals and have significant cost investment.

Another study concluded that further costs of population-based screening could be reduced using a two-step screening model combining VIA for primary screening followed by HPV DNA detection as a secondary test9. In this study, VIA positive women were tested for the presence of oncogenic HPV types by hybrid capture II (HC2) and with colposcopy. VIA had a high sensitivity to detect CIN grade 2+ lesions and HPV DNA and colposcopy were reserved as secondary tests that had a high specificity for detecting CIN2 + lesions9. In a subsequent study, the HPV HC2 test was shown to be a better triage test in VIA-positive women to detect >CIN210.

Tata Memorial Centre also participated in a study that investigated the detection of HPV from menstrual blood. Menstrual pads were collected from women of 30-50 yr age group from rural regions within the State of Maharashtra followed by HPV testing and colposcopy. The menstrual pad method showed acceptable sensitivity and specificity providing an alternative option for a non-invasive and simple method of cervical cancer screening11. The aforementioned trials lay a foundation for cost-effective strategies that can be implemented at national levels both within India and other countries. With the development and testing of low-cost, rapid molecular-assay technologies for HPV DNA detection, this may become viable as a primary screening tool in LMICs in the future12.

The Tata Memorial Centre also participated in a multicentric HPV vaccination study that was prematurely closed with some participants receiving only one or two doses of vaccine. This premature closure of the study provided an opportunity to study the immunogenicity of girls who received one or two doses of HPV vaccination. A seven-year follow up of this study demonstrated no persistent HPV infection in two dose cohort which was similar to a three-dose cohort suggesting that a two-dose recommendation could be safely extended to girls from 15 to 18 yr of age13. Furthermore, data on the cohort that received only one dose with a median follow up of 4.7 yr indicated similar rates of cumulative incident and persistent infections suggesting that even a single dose schedule could be an effective option for HPV vaccination14. However, there is a need for further follow up for immunogenicity and safety.

Development and validation of FIGO/TNM staging system: Experts from Tata Memorial Centre contributed to the revised International Federation of Gynecology and Obstetrics (FIGO) staging for cervical cancer in 201815. In a study of 632 locally advanced cervical cancer (LACC) patients, the investigators categorized outcomes as per the old and new FIGO and the AJCC (American Joint Committee on Cancer) staging system. The three-year disease-free survival (DFS) for stages IB, IIA, IIB, IIIA, IIIB, IIIC1, IIIC2 and IVA were 100, 93, 84, 53, 77, 74, 61 and 61 per cent, respectively. Patients with lymphadenopathy had significantly inferior outcomes when compared to node-negative patients (62.9 vs. 77.8%, P=0.002) validating the importance of nodal involvement in cervical cancer. However, DFS was poorer in patients with ≥3 para-aortic and pelvic nodes than <3 para-aortic and pelvic nodes (13.6% vs. 56.3%, P=0.001). Patients with primary tumour volume >30 cm³ also had worse three-year DFS than those with ≤30 cm³ volume (67.4 vs. 78.5%, P=0.002)16. The results of the validation study16 suggest that further refinement of staging is needed with the incorporation of tumour and nodal volume. Based on the observations within this large cohort, modification of the existing TNM and FIGO system was suggested to allow reclassification of patients in four different stage groupings with high fidelity16. These observations in the future may help with further evolution of the FIGO and TNM staging.

Practice changing therapeutic trials: National Cancer Institute alert in 1999 recommended that concurrent chemoradiation should be the standard of care for the management of LACC17; however, the clinical trials testing the role of concurrent chemoradiation had under-representation of patients in stage III B18. A Phase III trial (NCT 00193791) was initiated at the Tata Memorial Centre to test the benefit of concurrent chemoradiation in stage III B and accrued 850 patients from 2003 to 2011. After a median follow up of 88 months, the study reported in 2018 that the five-year DFS was significantly higher in the chemoradiation arm (52.3%) when compared to RT alone arm (43.8%). Eight per cent overall survival (OS) benefit was observed in the chemoradiation arm (54 vs. 46%). This study provided level one evidence in favour of concurrent weekly cisplatin chemotherapy when compared to RT alone in stage III B patients19. The absolute benefit observed within this phase III trial was eight per cent as compared to three per cent reported by the individual patient data meta-analysis20.

Tata Memorial Centre was also among the first ones to test and report the role of neoadjuvant chemotherapy (NACT) and surgery as compared to chemoradiation in stage IB2-IIB cervical cancer. A phase III randomized controlled trial of 635 patients with FIGO stage IB2, IIA and IIB cervical cancer was designed to test the superiority of NACT-surgery ± adjuvant chemo-radiation over chemoradiation21. However, at the final analysis, inferior outcomes for the primary end point of DFS were observed in the NACT-surgery arm. At a median follow up of 58.5 months, five-year DFS of 69.3 per cent was observed in NACT-surgery and 76.7 per cent in chemoradiation arm (P=0.03)21. The results provided an important answer about the lack of role of NACT-surgery in the management of stage IB2-IIB cervical cancer. The results were later corroborated by the EORTC trial22.

Pattern of care and implementation studies

In addition to performing randomized trials, the Tata Memorial Centre has consistently reported outcomes of patients treated outside clinical trials. In a large retrospective study, outcomes of 6234 patients with stage IB-IIIB cervical cancer treated with radical intent between 1979 and 1994 were reported23. At a median follow up of 68 months, the DFS for stage IB-IIA was 60-62 per cent at eight years with either surgery, radical radiation therapy or pre-operative radiation followed by surgery. For stage IIB and IIIB, a DFS of 56 and 40 per cent, respectively, was achieved at eight years23. Concurrent chemoradiation results in a five-year OS rate of 66 per cent and DFS of 58 per cent in LACC24. In another study, five-year outcomes were reported for patients treated in 2010. Overall, a five-year DFS of 62 per cent was observed in stage II and 45 per cent in stage III suggesting improvement over the historical series. Grade 3 or higher late toxicity was seen in only five per cent of the patients25.

From 2014, Tata Memorial Centre adopted radiation dose escalation through brachytherapy and reported three-year outcomes in 2018. In a series of 339 patients of LACC, the audit reported three-year local control, DFS and OS of 94.1, 83.3 and 82.7 per cent, respectively, in stage IB-IIB (FIGO 2009). For stage III-IVA, the corresponding rates were 85.1, 60.7 and 69.6 per cent, respectively26. These studies reported improvement in outcomes as a result of the effective delivery of clinical care for all patients. In another study of 632 patients, there was a further reduction in local (6.1%), pelvic nodal (4.9%), para-aortic nodal (6.9%) and distant relapse (9.4%), suggesting efficient adoption of full course concurrent chemoradiation, primary and nodal dose escalation in routine clinical practice. It was noteworthy that the three-year DFS in pelvic and para-aortic node positive and stage IV A disease (III C1 and III C2) patients was 74, 61 and 61 per cent, respectively16.

Selected studies from Tata Memorial Centre have reported the outcomes of patients who are poorly represented in LACC clinical trials. Outcomes of patients with lower vaginal involvement are not adequately reported in the literature and these patients are often not included in prospective clinical trials. In an audit study that reported 118 patients with lower third vaginal involvement, the three-year DFS and OS of FIGO stage IIIA patients were significantly better than IIIB patients (71 vs. 56%, P=0.02 and 76 vs. 66%, P=0.01) after a median follow up of 30 months. The overall three-year DFS and OS rates were 61.5 and 69.8 per cent, respectively. Concurrent chemotherapy and the absence of persistent disease independently predicted survival on multivariate analysis27. Another study reported outcomes of LACC patients who present with obstructive uropathy at the time of initial diagnosis28. In this group of patients, percutaneous nephrostomy and double-J stenting was performed. Only 50 per cent of patients could receive radical chemoradiation. While the median survival was 10 months for all the overall patients, select patients with good performance status, small tumour size and serum creatinine <3 mg/dl who could complete chemoradiation had superior clinical outcomes as compared to others (30 vs. 10 months). The study identified a small number of patients with advanced stage who could potentially become long-term survivors and the clinical features listed above to assist in patient selection28.

Clinical audits of outcomes have also been reported of patients with neuroendocrine histology and special clinical circumstances as coexisting HIV infection293031. As a proportion of patients with cervical cancer also present with very advanced disease, a report from the institution summarises the outcomes following extreme hypofractionation for palliation for very advanced cervical cancer. A study evaluating the outcomes of 100 patients with palliative RT reported 100, 49 and 33 per cent response to bleeding, discharge and pain respectively, with a single fraction of 10 Gy. By second fraction, more than 50 per cent pain relief was reported in up to 60 per cent of patients with a median survival of seven months32. In another study in patients with locally advanced metastatic rectal cancer use of a hypofractionated schedule of 25 Gy in 5#, we observed control of pain and bleeding in up to 80 per cent of patients33. A phase III randomized study has recently been initiated to test the superiority of rapid palliation schedule (25 Gy/5#/one week) over a protracted schedule of 30 Gy/3#/7-8 wk. A total of 230 patients will be included34.

Tata Memorial Centre is actively involved as a part of the National Cancer Grid of India initiative for treatment homogenization. Institutional representatives participated in the development of resource stratified treatment guidelines. Furthermore, a national resource assessment was undertaken to study the feasibility of broader implementation of guidelines35. A multi-institutional study is presently underway to evaluate the implementation of guidelines which will provide valuable information about the broader level implementation of these policies.

Studies investigating advances in external beam technology

In a phase II randomized trial of 200 patients that evaluated dose escalation in LACC through a combination of intensity-modulated radiotherapy (IMRT) and standard brachytherapy, patients were randomized into IMRT and brachytherapy [50 Gy/25#/5 wk followed by 5# of 7 Gy high-dose rate (HDR) weekly brachytherapy to point A] (Study arm) or 3D conformal radiation and brachytherapy (40 Gy/20#/4 wk with midline shielding at 20 Gy followed by 7 Gy HDR brachytherapy to point A x 5# weekly) (standard arm). After a median follow up of 62 months, no difference was observed in locoregional control (LRC), DFS and acute toxicities between the two arms. However, late rectal toxicity was significantly higher in the study arm as compared to the standard arm secondary to dose escalation. [grade III/IV: 15 versus 2 per cent (P<0.005)]36.

The Centre has also recently published the results of a phase III randomized trial comparing 3DCRT with image-guided IMRT (IG-IMRT) in patients receiving post-operative radiation for cervical cancer. The trial was designed to detect a reduction in grade ≥2 late bowel toxicity with the application of IG-IMRT37. At a median follow up of 46 months, the three year cumulative incidence of grade ≥2 late GI toxicity was 21.1 vs 42.4 per cent (P<0.01) in IG-IMRT arm and 3DCRT arm respectively38. In a sub-analysis from this study, the feasibility of bone marrow sparing was reported39; however, simultaneous bowel and bone marrow sparing needs further investigation.

A previous study compared the performance of two techniques of IMRT delivery [volumetric modulated arc therapy (VMAT) and fixed field IMRT], and reported improved performance of VMAT in terms of dose homogeneity and conformity with significant improvements in organs at risk sparing40. Further investigators compared two different techniques of VMAT (halcyon vs. tomotherapy). Essential equivalence was noticed between halcyon and helical tomotherapy based plans for IMRT in cervical cancer41. The role of machine learning in improving IMRT planning process fidelity and efficiency was studied. A prospective study reported no difference between clinical and machine learning generated treatment plans inferring that the automated machine-based plans were robust42. These observations may be important and facilitate the transition of high-quality care delivery with high fidelity in high volume centres.

Contribution to brachytherapy

Brachytherapy is an integral part of cervical cancer treatment, and Tata Memorial Centre has been actively involved in prospective clinical studies at national and international trials43. In a phase III randomized study of 830 patients that compared low-dose rate to HDR, no significant difference was observed in LRC, survival or toxicity between the two44. Tata Memorial Centre was also a major contributor to the phase III randomized study, initiated by the International Atomic Energy Agency (IAEA), across seven countries in 601 patients to compare two fractionation regimens for HDR brachytherapy. Of the 601 patients, 257 patients studied at Tata Memorial Centre contributed. The study compared 7 Gy × 4# with 9 Gy × 2#. Local control was significantly higher for the arm using 7 Gy × 4#. No differences in OS or adverse effects were found between the arms. This was the first trial that reported benefit of brachytherapy dose escalation45.

Experts from Tata Memorial Centre have also contributed to the international publication of International Commission on Radiation Units and Measurements (ICRU) Report 8946. Within an international collaborative framework advanced brachytherapy principles (EMBRACE and EMBRACE II) were tested at the centre. In collaboration with the medical university of Vienna, Tata Memorial Centre piloted a technical modification in brachytherapy technique to improve brachytherapy coverage in patients with a poor response to external radiation. Close to 90 per cent of patients in this study had a residual disease in the distal parametrium/pelvic wall following external radiation. The results showed local control, DFS and OS of 76/72, 56/50, 62/54 for three and five years, respectively, with acceptable toxicities even in patients with poor response47. A simulation study was conducted to evaluate the macroeconomic benefits with MR-IGBT over conventional brachytherapy that demonstrated significant economic gains with IGBT48. There is a lack of level I evidence supporting the wide-scale implementation of image-guided brachytherapy. While Tata Memorial Centre is presently conducting a phase III randomized study to test the superiority of 2D over image guided brachytherapy49.

Currently image-guided brachytherapy is desirable, there is a considerable deficit in imaging scanners throughout the country as compared to developed countries50. In an online survey conducted among radiation oncologists of India, it was noted that many of the centres commonly carried out CT and X-ray imaging with Point A prescription being the most common methodology for planning51. Low-cost imaging modalities such as CT and ultrasonography (USG) were more in practice compared to MRI. In a prospective study that evaluated CT based target and organs-at-risk (OAR) delineation with the help of MRI at diagnosis and real-time transrectal ultrasound (TRUS) information during brachytherapy planning, the results were comparable to MRI-based approach5253. In a feasibility study of using transabdominal ultrasound for brachytherapy planning, 60 patients of stage IIB to III B cervical cancer were included. Brachytherapy was performed with ultrasound guidance. Reference points were defined on sonography as a surrogate for target and extrapolated on to CT. Point A planning with OAR optimization was done. With a median follow up of 49 months, the five-year local and pelvic control, DFS and OS were 93.3, 85, 71.7, and 71.7 per cent, respectively. Hence, USG may be an acceptable alternative to MRI brachytherapy54.

In addition to the above techniques, our institution has developed innovative techniques in brachytherapy of cervical cancer. A novel technique using a single ovoid and a central tandem was reported for patients who have reduced vaginal space that limits standard intracavitary procedures. During the first application, central tandem and a single ovoid is inserted, the other ovoid being replaced by a rubber tube. For the subsequent application, contralateral ovoid is inserted55. This modified high dose rate (HDR) intracavitary technique may prove an alternative for centres where interstitial brachytherapy for cancer of the cervix is not available.

Tata Memorial Centre undertook a nationwide audit to understand brachytherapy availability for cervical cancer and estimated the burden of LACC in each State and the required infrastructure for EBRT and brachytherapy to meet the current needs of cervical cancer56. Brachytherapy deficit was noted in 14 Indian States, and it was estimated that additional 58 brachytherapy units would be needed. It was estimated that, with the current BT deficit, approximately 14,000 women in India may have delayed or no access to brachytherapy. As the infrastructural deficit may take time to expand, various strategies to improve brachytherapy access including altered abbreviated brachytherapy fractionation schedules are being investigated57.

Recurrent cancers

Local recurrences following suboptimal surgery without appropriate adjuvant treatment are not uncommon in patients with cervical cancer in India and in many other LMICs where adequate surgical expertise or referral practices for adjuvant radiotherapy may not be in place58. Even after definitive non-surgical treatments, 30-50 per cent relapse rates have been reported in the literature59. In a retrospective study of patients with recurrent vault/vaginal cancers treated with interstitial brachytherapy from 2000 to 2008, modest clinical outcomes with three-year locoregional relapse-free survival, DFS and OS of 62.4, 59.5 and 73.4 per cent, respectively, with acceptable late toxicities were seen60. A prospective phase II study integrating tomotherapy-based intensity-modulated radiotherapy (IMRT), interstitial brachytherapy and 3D planning for all patients reported five-year actuarial DFS and OS of 75 and 71 per cent with grade III/IV rectal and genitourinary toxicity in five and three per cent patients, respectively61. In a phase II study of MRI and PET-guided interstitial brachytherapy for post-surgical vaginal recurrences of cervical cancer, patients were recruited with residual or recurrent disease after hysterectomy. Patients underwent baseline T2W MRI, 18 F-FDG, 18 F-FLT and 18 F-F Miso PET and received EBRT 50 Gy/25#. MRI was performed at brachytherapy and utilized for the delineation of clinical target volume. While patients with parametrial disease at baseline received interstitial brachytherapy (16-20 Gy/4-5#), those with vaginal disease received intracavitary brachytherapy (12-14 Gy/2-4#). Image-guided radiation and brachytherapy was associated with good-to-excellent local control, DFS and OS of 84, 73 and 74.5 per cent, respectively, in patients with vaginal recurrences of cervical cancer62. Functional score stability was observed from baseline till 24 and 60 months. For the symptom domain, at two and five years post-radiotherapy, the difference in toxicities were not significant.

Only a few studies worldwide have focussed on re-irradiation of gynaecological cancers with brachytherapy ± external radiation. The Centre has published outcomes in this setting using individualized intracavitary/interstitial brachytherapy techniques. With a median follow up of 25 months, two-year local control, DFS and OS were 44, 42, and 52 per cent, respectively. This study also showed that the outcome was better with higher doses63. An update of this series showed two-year local control, DFS and OS of 54, 50 and 64 per cent, respectively. Grade 3 or higher toxicity was seen in 22.4 per cent patients64. This was superior to systemic chemotherapy, targeted therapy or immunotherapy for treating central relapses that were not amendable for surgery65.

Toxicity reduction studies

In a series of studies, dose-volume response has been reported for various normal pelvic organs with an aim to reduce long-term toxicity. The relationship between dose-volume parameters and severe bowel toxicity was analyzed in a study of 71 patients undergoing post-operative radiation for cervical cancer. Grade 2 or higher bowel toxicity was seen in 30.9 per cent grade 3 or higher bowel toxicity in 12.6 per cent patients. On univariate analysis, V15 small bowel (SB) <275 ml, V30 SB <190 ml, and V40 SB <150 ml and V15 large bowel (LB) <250 ml and V40 LB <90 ml predicted for the absence of grade 3 or higher toxicity. On multivariate analysis, only V15 SB and V15 LB were significant66. In another study of 103 patients, V30 peritoneal cavity ≥900 ml and use of concurrent chemotherapy independently predicted acute toxicity. The presence of acute grade ≥III toxicity independently predicts late toxicity67. While the QUANTEC essentially reported a correlation of radiation dose with acute toxicity, our study was the first to report a correlation of radiation dose with late toxicity and possible potential of advanced techniques in toxicity reduction which was subsequently tested and validated in a phase III trial (PARCER NCT 01279135)38. Other studies have focussed on reporting dose-volume and toxicity correlation for rectum, duodenum, bone marrow and vagina6869. Prospective work from the institution has also focussed on reporting the temporal trends of late toxicity after pelvic radiation. This reporting of temporal trends may help in providing better understanding of reporting late toxicity as prevalence rather than incidence.

Translational research studies

Tata Memorial Centre also has actively contributed to translational research in cervical cancer in patients undergoing treatment. Some of the notable contributions include the role of viral integration site (episomal) and persistent HPV viral load as a predictor of recurrence after chemoradiation70. In a prospective study, patients with persistent HPV 16/18 infection had a significantly higher overall and locoregional relapses (49 and 32%) as compared to those who had clearance of HPV by nine months (28 and 11% with P=0.024 and P=0.02, respectively). Furthermore, persistent HPV infection by 24 months showed a significant impact on LRC and recurrence-free survival (RFS)7172.

Our institution was also one of the few institutions to prospectively test the impact of cancer stem cells on outcomes of cervical cancer following chemoradiation. Pre-treatment and post biopsies were performed for 150 patients. At baseline, moderate-to-strong immunohistochemical expression of SOX-2, OCT-4, Nanog, CD44 and Podoplanin was observed in 12.8, 4.8, 24.4, 15.5 and 1.3 per cent patients, respectively. At a median follow up of 30 months, locoregional and distant relapse was observed in 12.2 and 23.1 per cent patients. The three-year DFS was 87 per cent. Moderate-to-high CSC expression and CD44 low status (P<0.04) independently predicted for locoregional relapse-free survival73. The results of these studies have also been incorporated in to the international multicentric study BIOEMBRACE-174, which is investigating markers of epithelial–mesenchymal translation, stemness and immune response with disease recurrence in patients treated with chemoradiation for LACC recruited within EMBRACE studies.

Research and education

The Tata Memorial Centre has contributed towards training and education in gynaecological cancers within the framework of educational collaborations with the IAEA conducting training of international participants for gynaecological brachytherapy. In collaboration with Elekta, the Centre conducts the brachytherapy training school Brachy Academy75, and the institution also accepts scholars from global health programmes.

Conclusion & future directions

Tata Memorial Centre is presently leading the investigation of the role of HIV protease and Protein kinase B (Akt) inhibitor Nelfinavir in improving disease-free survival in patients receiving chemoradiation for LACC in an ongoing randomized trial76. In another ongoing randomized trial optimal fractionation for the palliation of advanced cervical cancer is being studied34.

To summarize, Tata Memorial Centre has provided substantial contributions in various aspects of cervical cancer prevention, treatment and research over the past three decades. This includes contribution towards prevention through screening, practice-changing therapeutic trials, advancing radiation technologies, brachytherapy, educational and training activities and national and international collaborations.