ESTRO 2024 Congress Report
At ESTRO 2024, we had a presentation of some great scientific abstracts from phase III trials in locally advanced cervical cancers as well as large cohort studies for recurrent and metastatic gynaecological cancers. Below we discuss one abstract of a phase III trial that tested the addition of immunotherapy to chemoradiation and brachytherapy in cervical cancer and two abstracts from large cohorts of patients of oligometastatic cancers and reirradiation.
A Phase 3 Study of Pembrolizumab + Chemoradiotherapy for High-Risk Locally Advanced Cervical Cancer
Cerrotta A. et al.
Proffered paper 3560
Since the last two decades, there have been multiple phase III trials that have investigated the addition of systemic chemotherapy in either neoadjuvant or adjuvant setting to improve the clinical outcomes of locally advanced cervical cancer. The main aim of these phase III trials was to test if the addition of chemotherapy improved overall survival over what was achieved using concurrent chemoradiation and brachytherapy (e.g., OUTBACK study that tested the addition of adjuvant chemotherapy). Metanalysis of adjuvant chemotherapy trials reported no benefit of this strategy. Similarly, the use of neoadjuvant chemotherapy did not lead to an improvement in overall survival in a vast majority of clinical trials (EORTC trial and Tata Neoadjuvant Chemotherapy trials). Recently INTERLACE phase III trial reported improvement in overall survival with the use of weekly neoadjuvant chemotherapy when administered before chemoradiation and brachytherapy.
Immunotherapy has recently been approved for use in recurrent metastatic setting. However, phase III trials in locally advanced settings did not demonstrate improved outcomes. CALLA trial, which utilised concurrent and adjuvant Durvulamab, reported failure to meet the primary endpoint. The ENGOT-cx11/GOG3047/KEYNOTE-A18 study (ClinicalTrials.gov, NCT04221945) randomised eligible patients with newly diagnosed, previously untreated, high-risk LACC (International Federation of Gynaecology and Obstetrics 2014 stage IB2‒IIB with node-positive disease or stage III‒IVA) to receive 5 cycles of pembrolizumab 200 mg or placebo Q3W plus CCRT, then 15 cycles of pembrolizumab 400 mg or placebo Q6W. The CCRT regimen included 5 cycles (with an optional 6th dose) of cisplatin 40 mg/m2 QW plus EBRT followed by brachytherapy. Patients were stratified by planned EBRT type (intensity-modulated radiotherapy [IMRT] or volumetric-modulated arc therapy [VMAT] vs non-IMRT or non-VMAT), stage at screening (stage IB2‒IIB vs III‒IVA), and planned total radiotherapy dose (<70 Gy vs ≥70 Gy equivalent dose in 2-Gy fractions [EQD2]). Primary endpoints were progression-free survival (PFS) per RECIST version 1.1 by investigator or histopathologic confirmation and overall survival (OS). The study randomised 1060 patients with locally advanced cervical cancer. Patients in either arm received chemoradiation and brachytherapy. In the experimental arm, 10.4% of patients had discontinuation of cisplatin, pembrolizumab in 1.5% of patients, EBRT in 0% of patients, and brachytherapy in 0% of patients. The corresponding treatment-related AE discontinuation rates in the placebo plus CCRT group were 10.8%, 0.9%, 0%, and 0.2%, respectively. At a median follow-up of 17.9 months, Pembrolizumab plus CCRT significantly improved PFS versus placebo plus CCRT (hazard ratio, 0.70 [95% CI, 0.55‒ 0.89; P = 0.0020]). The median PFS was not reached in either group.
Cerrotta et al. presented the updated survival at ESTRO 2024, but with only 103 events (42.9% maturity), the addition of pembrolizumab to CCRT showed a favourable trend in OS (hazard ratio, 0.73 [95% CI, 0.49‒1.07]); these data did not cross the boundary of statistical significance at this interim analysis.
Though the initial PFS results and OS trends provide encouraging data, it is noteworthy that the first analysis was undertaken when many patients were yet actively on treatment, suggesting the need for further mature follow-up and analysis. It will also be worthwhile in this study to see the proportion of patients undergoing immunotherapy after disease progression in the standard arm as that could potentially impact the OS analysis. Further structured details may help in interpreting the impact of immunotherapy on OS. If the final analyses of Keynote A18 demonstrate improvement in survival, then the standard of care would be redefined. Robust implementation strategies will be needed in such a case to allow women in high-incidence regions to have access to such treatments.
References of Interest
Mileshkin LR, Moore KN, Barnes EH, et al. Adjuvant chemotherapy following chemoradiotherapy as primary treatment for locally advanced cervical cancer versus chemoradiotherapy alone (OUTBACK): an international, open-label, randomised, phase 3 trial. Lancet Oncol. 2023 May;24(5):468-482. doi: 10.1016/S1470-2045(23)00147-X. Epub 2023 Apr 17. PMID: 37080223; PMCID: PMC11075114.
Gupta S, Maheshwari A, Parab P, Mahantshetty U, Hawaldar R, Sastri Chopra S, Kerkar R, Engineer R, Tongaonkar H, Ghosh J, Gulia S, Kumar N, Shylasree TS, Gawade R, Kembhavi Y, Gaikar M, Menon S, Thakur M, Shrivastava S, Badwe R. Neoadjuvant Chemotherapy Followed by Radical Surgery Versus Concomitant Chemotherapy and Radiotherapy in Patients With Stage IB2, IIA, or IIB Squamous Cervical Cancer: A Randomized Controlled Trial. J Clin Oncol. 2018 Jun 1;36(16):1548-1555. doi: 10.1200/JCO.2017.75.9985. Epub 2018 Feb 12. PMID: 29432076.
Horeweg N, Mittal P, Gradowska PL, Boere I, Nout RA, Chopra S. A systematic review and meta-analysis of adjuvant chemotherapy after chemoradiation for locally advanced cervical cancer. Crit Rev Oncol Hematol. 2022 Apr;172:103638. doi: 10.1016/j.critrevonc.2022.103638. Epub 2022 Feb 18. PMID: 35189325.
Kenter GG, Greggi S, Vergote I, Katsaros D, Kobierski J, van Doorn H, Landoni F, van der Velden J, Reed N, Coens C, van Luijk I, Colombo N, Steen-Banasik EV, Ottevanger N, Casado A; EORTC-55994 Study Group. Randomized Phase III Study Comparing Neoadjuvant Chemotherapy Followed by Surgery Versus Chemoradiation in Stage IB2-IIB Cervical Cancer: EORTC-55994. J Clin Oncol. 2023 Nov 10;41(32):5035-5043. doi: 10.1200/JCO.22.02852. Epub 2023 Sep 1. PMID: 37656948.
Monk BJ, Toita T, Wu X, et al: Durvalumab versus placebo with chemoradiotherapy for locally advanced cervical cancer (CALLA): a randomised, double-blind, phase 3 trial. The Lancet Oncology 24:1334–1348, 2023 [cited 2024 Jan 12]
Lorusso D, Xiang Y, Hasegawa K, et al: Pembrolizumab or placebo with chemoradiotherapy followed by pembrolizumab or placebo for newly diagnosed, high-risk, locally advanced cervical cancer (ENGOT-cx11/GOG-3047/KEYNOTE-A18): a randomised, double-blind, phase 3 clinical trial [Internet]. The Lancet 403:1341–1350, 2024[cited 2024 May 13]
SBRT and artificial intelligence in oligometastatic GYN-cancers: a real-world study.
Macchia G. et al.
603 Proffered paper
The use of radical and ablative doses for oligometastatic cancers is on the rise and is supported by the long-term follow-up of completed randomised trials. However, a vast majority of these clinical trials have an underrepresentation of women with gynaecological cancers. Recent clinical trials that investigate immunotherapy in combination with standard chemotherapy for gynaecological cancers demonstrate improvement in survival. However, in clinical practice, a vast majority of oligometastases are treated with locally directed therapies like SBRT. With the emerging evidence of systemic therapies, it is important that patients are carefully selected for local therapies. In this case knowledge of patients who are likely expected to have a complete response after SBRT is highly desirable. Currently, there are no established predictive models for clinical outcomes of gynecologic oligometastatic cancer treated with SBRT. Literature suggests that achieving a complete response after SBRT may influence oncologic outcomes, but this relationship is not well-defined. The availability of a predictive model can help select appropriate patients for SBRT.
Macchia et al. reported an artificial intelligence-based model using real-world data to predict complete response to SBRT in women with oligometastases and report 2-year local control on a “per lesion” basis. The study also reported Distant Metastases Free Survival (DMFS), PFS, and OS. Overall, 501 patients from 21 radiation oncology institutions with 846 gynaecological metastases (ovarian: 449, Cervix: 125, Uterus: 272), representing one of the largest series of oligometastatic gynaecological cancers. The majority of lesions were ovarian (449, 53.1%), uterine (272, 32.1%), or cervical 125 (14.8%) origin. Multiple fraction radiotherapy was used to treat 762 metastases (90.1%). The most frequent schedule was 24 Gy in 3 fractions (13.4%). The features of lesions and treatment details are depicted in the table below.
The use of SBRT leads to an 85.9% objective response rate (complete response of 63.7% and partial response of 22.3%). The complete response rate was higher in ovarian and endometrial primary (65.2% and 64%) as compared to the lesions of cervical origin (58.4%). SBRT was associated with a 2-year local control rate of 79.2%. The 2-year local control (91.2% vs 52.5%, p=0.001), PFS (35.4% vs 10.8%, p=0.01) and OS (89.4% vs 76.2%, p=0.001) were superior in patients who developed complete response after SBRT as in prostate, colorectal, breast and lung settings (4-7) though out of field progression remained high in both cohorts.
Model development for prediction of complete response was undertaken using the Least Absolute Shrinkage and Selection Operator (LASSO) method to select the most relevant variables from the dataset. In this case for prediction of CR Macchia et al considered Age (≤63 years versus >63 years), Primary tumour site (cervix versus ovary versus endometrium), Disease burden (one lesion versus more than one lesion), Type of lesion (nodal versus parenchymal), Previous radiotherapy (yes versus no), Planning Target Volume (PTV as a continuous variable) and Biologically Effective Dose (BED as a continuous variable). After LASSO selection of significant variables, these were used to create a Classification And Regression Tree (CART) model. CART is a decision tree learning technique that results in a tree structure, where each internal node represents a "test" on an attribute (e.g., whether a coefficient is above a certain value), each branch represents the outcome of the test, and each leaf node represents a class label (in this case, complete response or not). The performance of the CART model is evaluated using receiver operator characteristic (ROC) curves and the area under the curve (AUC).
For this large series of patients with oligometastatic, machine learning analysis showed a poor ability to find covariates strong enough to predict CR in the whole series. Analysing the malignancies separately, in uterine cancer, if dose (BED10)>78.3Gy, the CR probability was 75.4%; moreover, if the volume was less than 13.7cc, the CR probability became 85.1%. In ovarian cancer, if the lesion was a lymph node, the CR probability was 71.4%; if the volume was less than 17.0cc, the CR probability rose to 78.4%. No covariate predicted the CR for cervical lesions.
In conclusion, this is the largest study (21 centres, 846 lesions, 501 patients) to identify complete response predictors to SBRT. Based on the results, a robust and consistent CART model, employing 4 clinical variables is proposed. For ovarian cancer lesions, the type of lesion (lymph node or parenchyma) was the most important variable, whereas for uterine cancer lesions, the BED10 was the most important one. Volume inverse correlation was also observed for ovarian cancer and uterine cancer. For cervical cancer, no variables are sufficiently predictive to create a model. This could also be attributed to a small cohort in this case and highlights the need for also possibly greater sample size. This may also be related to primary histological subtype (squamous) and HPV-related disease biology where a larger majority of the population may achieve a complete response.
This study highlights that complete response acts as a major driver for prognosis. Therefore, every effort must be taken to eradicate oligo-disease. The ability to predict CR through machine learning models could be a starting point to drive treatment choices in the context of personalised oncology and possibly delay systemic therapies to a later stage. Though retrospective in design and with limited data about radiation dose and use of systemic therapies, the strength of the study remains in the large collaborative effort and consistent availability of chosen variables. Further future expansion of collaborative effort is expected.
The limitations of the study are represented by the retrospective nature of the data and the modest number of clinical and dosimetric parameters available for modelling of the dose-response relationship. It is possible that other variables with potential influence on outcome prediction, like molecular subtypes, have not been imputed. However, the limited number of clinical variables should also be viewed as a benefit, as such models are less likely to overfit and easier to use. This results in a more fundamental and comprehensible model. For further detailed information and patient characteristics readers can access Macchia G et al. Gynecol Oncol. 2024 Jan 24; 184:16-23.
Further references of interest
G. Macchia, R. Lazzari, N. Colombo, et al., A large, multicenter, retrospective study on efficacy and safety of stereotactic body radiotherapy (SBRT) in Oligometastatic ovarian Cancer (MITO RT1 study): a collaboration of MITO, AIRO GYN, and MaNGO groups, Oncologist. 25 (2) (2020) e311–e320.
G. Macchia, A. Nardangeli, C. Laliscia, et al., Stereotactic body radiotherapy in oligometastatic cervical cancer (MITO-RT2/RAD study): a collaboration of MITO, AIRO GYN, and MaNGO groups, Int. J. Gynecol. Cancer 32 (6) (2022) 732–739.
G. Macchia, D. Pezzulla, M. Campitelli, et al., Efficacy and safety of stereotactic body radiotherapy (SBRT) in oligometastatic uterine cancer (MITO-RT2/RAD study): a large, real-world study in collaboration with AIRO GYN, MITO and MANGO groups, Int J Radiat Oncol Biol Phys 117 (2) (2023) 321–332.
D.R. Gomez, G.R. Blumenschein Jr., J.J. Lee, et al., Local consolidative therapy versus maintenance therapy or observation for patients with oligometastatic non-small cell lung cancer without progression after first-line systemic therapy: a multicentre, randomised, controlled, phase 2 study, Lancet Oncol. 17 (2016) 1672–1682.
P. Iyengar, Z. Wardak, D.E. Gerber, et al., Consolidative radiotherapy for limited metastatic non-small-cell lung cancer: a phase 2 randomized clinical trial, JAMA Oncol. 4 (2018), e173501.
R. Phillips, W.Y. Shi, M. Deek, et al., Outcomes of observation vs stereotactic ablative radiation for Oligometastatic prostate Cancer: the ORIOLE phase 2 randomized clinical trial, JAMA Oncol. 6 (5) (2020) 650–659. [6] D.A. Palma, R. Olson, S. Harrow, et al., Stereotactic ablative radiotherapy for the comprehensive treatment of Oligometastatic cancers: long-term results of the SABR-COMET phase II randomized trial, J. Clin. Oncol. 38 (25) (2020) 2830–2838.
P. Ost, D. Reynders, K. Decaestecker, et al., Surveillance or metastasis-directed therapy for oligometastatic prostate cancer recurrence (STOMP): five-year results of a randomized phase II trial, J. Clin. Oncol. 38 (2020) (6_suppl 10-10).
Outcomes of pelvic reirradiation with stereotactic radiotherapy for gynaecological cancer recurrence
Thomas B. et al.
2829 Proffered paper
Thomas B. et al. from Royal Marsden reported single institution outcomes of patients undergoing reirradiation for infield recurrences of gynaecological cancers from 2012-2021 at ESTRO 2024. Pelvic recurrence of gynaecological cancer is associated with significant morbidity and can be challenging to treat, with surgery particularly when arising in a previously irradiated site. Highly focal re-irradiation can provide sustained infield control, providing an opportunity to delay the initiation of systemic therapies. There is however limited published data regarding the long-term efficacy and safety of pelvic reirradiation using SBRT in gynaecological malignancies1–4, and a lack of guidance on case selection, technique, dose and fractionation5 Thomas presented an analysis of a large cohort of patients (n=73) from Royal Marsden who underwent reirradiation using SBRT for gynaecological cancers. This included 45% of patients with cervix cancer, 37% with uterine cancer, 8% with ovarian cancer and 5% each with vulva and vaginal cancer. In this cohort, 60% received treatment for soft tissue relapse, 25% for nodal relapse and 15% for a positive surgical margin. A vast majority of patients were treated with cyberknife (65%), C-Arm Linac (27%) or MR-Linac (8%). The median time to reirradiation was 28 months (4-360 months). The median target volume was 18.7 cc (2.1-98.1 cc), and the median SBRT dose was 30 Gy/5# (24-33 Gy/ 3-5#). Of these 73 patients, 8 received a second course of SBRT. The endpoints of local, regional and distant failure, PFS and OS, as well as acute and chronic toxicity data (CTCAE grading), were analysed. The median time to local failure of the treated lesion was 35.7 months (95% CI 24.7 months not reached) whereas the median PFS was 12.9 months (95% CI: 7.9-19.5 months). The median OS was 43.5 months (95% CI 33.8-69.9 months). Superior survival outcomes were observed in women with endometrial than cervical cancer (69.9 months vs. 36.6 months). Overall, 40% of patients failed locally at the treated site with or without other relapses, 45% failed locoregionally elsewhere and 15% failed at distant sites. Time from radiotherapy (r=0.74) but not GTV/CTV volume (r = 0.25) correlated with a longer time to treatment failure. SBRT reirradiation was well tolerated with acute G3 toxicity reported in 2/73 patients. Late G3 toxicity was documented in 2 /73 patients occurring >2 years after treatment. This report from Royal Marsden represents one of the large series for oligorecurrent gynaecological cancers reirradiation and SBRT was associated with excellent clinical outcomes. The study also reported that reirradiation can provide durable local control and delay time to systemic therapy by >12 months. Our outcomes are comparable to other smaller published case series.1-4
References of interest
- Ling DC, Vargo JA, Burton SA, Heron DE, Beriwal S. Salvage Curative-Intent Reirradiation Stereotactic Body Radiation Therapy for Isolated Pelvic and/or Paraortic Recurrences of Gynecologic Malignancies. Pract Radiat Oncol. 2019;9(6):418-425. doi:10.1016/j.prro.2019.05.012
- Park HJ, Chang AR, Seo Y, et al. Stereotactic Body Radiotherapy for Recurrent or Oligometastatic Uterine Cervix Cancer: A Cooperative Study of the Korean Radiation Oncology Group (KROG 14-11). Anticancer Res. 2015;35(9):5103-5110.
- Seo Y, Kim MS, Yoo HJ, et al. Salvage stereotactic body radiotherapy for locally recurrent uterine cervix cancer at the pelvic sidewall: Feasibility and complication. Asia Pac J Clin Oncol. 2016;12(2):e280-288. doi:10.1111/ajco.12185
- Macchia G, Pezzulla D, Cilla S, et al. Stereotactic Body Reirradiation in Gynaecological Cancer: Outcomes and Toxicities from a Single Institution Experience. Clin Oncol (R Coll Radiol). 2023 Oct;35(10):682-693.
- Slevin F, Aitken K, Alongi F, et al. An international Delphi consensus for pelvic stereotactic ablative radiotherapy re-irradiation. Radiother Oncol. 2021; 164:104-114. doi: 10.1016/j.radonc.2021.09.010
Dr Supriya Chopra, MD
Professor, Radiation Oncology
ACTREC, Tata Memorial CentreMumbai, India.
Dr Gabriella Macchia, MD
Chair Radiation Oncology Unit and Deputy Scientific Director
Responsible Research Hospital
Campobasso, Italy