We are facing several problems in the field of clinical oncology research, by which I mean that study results are being collected, analysed and reported in many different ways, and we tend not to speak the same language within our discipline, writes Robbert S. Puijk (Amsterdam University Medical Centers and OLVG Hospital, Amsterdam, The Netherlands). These shortcomings are the reason we took the initiative to develop new guidelines regarding preferred clinical outcome measures following image-guided tumour ablation.
Image-guided thermal and non-thermal tumour ablation techniques have become indispensable therapeutic options for a variety of cancer types. For smaller-size malignant tumours, international guidelines have already adopted thermal ablation as a first-line treatment option1-7. The continuing emergence of novel treatment options and growing demand for minimally invasive image-guided tumour ablation techniques have raised the need for evidence-based IO, and with that comes the need for clear documentation of oncologic outcome parameters.
Shortcomings in how to collect, analyse and report data were the motive for our team to develop a modified Delphi consensus project8 in order to provide a framework of key opinion leader recommendations on patient-, procedure-, and tumour-related definitions, starting and ending time definitions, survival time definitions, time-to-event end points, and patient-reported outcome measures. The team consisted of members of the Society of Interventional Oncology (SIO), two independent epidemiologists and Carine Bellera, biostatistician from the French-based Definition for the Assessment of Time-to-Event End Points in Cancer Trials (DATECAN) group.
A total of 62 IO experts from Europe, the United States and Asia, working in 48 centres, eventually joined the evaluating committee. The panellists were asked to rate multiple statements and to provide additional comments whenever desirable. The article can be found online as open access publication in the Radiology journal.
To make the purpose and final results of this project clear, I can provide a few common practical examples which we hope to have solved by now. One of the most straightforward, but misleading, situations is when overall survival, for example for locally advanced pancreatic cancer, is analysed using the date of diagnosis to the date of death in one study and the date of the interventional procedure to the date of death in the other study. In order to compare survival probabilities in terms of Kaplan-Meier estimates and avoid misinterpretations as mentioned earlier, the experts reached consensus on starting and ending time definitions by documenting the time from the date of detection of disease (diagnosis), the date of the start of neo-adjuvant or induction therapy, and the date of the first intervention.
Another confusing situation might occur when different outcome measures are misused or used interchangeably, such as local tumour progression-free survival (LTPFS), primary- or secondary technique efficacy and local control. LTPFS is intended to reflect the effectiveness of the first local treatment, while local control should be used to reflect the final success after additional overlapping (completion) procedures (+/- alternative treatment methods). Furthermore, we tend to merely analyse and report a variety of outcome measures and procedure- or tumour-related outcomes per patient only while results of this article show that this is strictly speaking incorrect. To overcome this issue the experts reached consensus on whether to address outcomes per patient, per session or per tumour. As previously published by Muneeb Ahmed and colleagues, efficacy of a certain ablation device or image-guiding technique to eradicate tumours can best be illustrated by comparing the time to local tumour progression.9 Suppose a situation where a patient has been locally treated for three liver metastases and local tumour progression is being analysed and reported on a per patient basis. In this case, progression of one of the three tumours already counts as an event while the other two tumours where treated successfully, leading to an impractical LTPFS curve while comparing data analysed and reported on a per tumour basis. Therefore, and for the fact that multiple index tumours in one patient cannot be regarded as independent as these tumours are potentially correlated, the experts agreed that one should address and report several parameters both on a per patient and per tumour basis.
Another common issue is related to the fact that, prior to our article, no clear agreements have been made concerning the documentation of complications in the field of IO, as opposed to surgical series where the Clavien-Dindo classification is established to be used.10 Although consensus was not reached for the preferred validated classification system to document, analyse, and report complications and adverse events, the experts have agreed that one should indicate the most recent version of a certain validated classification system so that complications and adverse events can be categorised consistently according to severity, time of occurrence, and likelihood of the event being related to the procedure.
We believe that the clear definitions in these comprehensive guidelines will undoubtedly provide a necessary foundation for scientific reproducibility between randomised and retrospective studies as they will ensure an objective and reliable interpretation of results, allow for accurate comparison of outcomes, and avoid misinterpretations among physicians and researchers. The interchangeable setup of this current consensus document allows us to follow up this project regularly and, hopefully in the near future, helps us to extend these guidelines to regional and systemic cancer treatments to eventually attract the participation of other (inter)national medical societies. Therefore, we encourage all of our colleagues to adopt the recommendations outlined in this proposal in order to facilitate universal communication of scientific advances in the field of clinical oncology research.
Robbert S Puijk is an interventional radiology resident at the OLVG Hospital, (Amsterdam, The Netherlands) and researcher at the Amsterdam University Medical Centers, location VUmc (Amsterdam, The Netherlands)
- Meijerink MR, Puijk RS, van Tilborg A, Henningsen KH, Fernandez LG, Neyt M, et al. Radiofrequency and Microwave Ablation Compared to Systemic Chemotherapy and to Partial Hepatectomy in the Treatment of Colorectal Liver Metastases: A Systematic Review and Meta-Analysis. Cardiovasc Intervent Radiol. 2018;41(8):1189-204.
- Nieuwenhuizen S, Puijk RS, van den Bemd B, Aldrighetti L, Arntz M, van den Boezem PB, et al. Resectability and Ablatability Criteria for the Treatment of Liver Only Colorectal Metastases: Multidisciplinary Consensus Document from the COLLISION Trial Group. Cancers (Basel). 2020;12(7).
- Forner A, Reig M, Bruix J. Hepatocellular carcinoma. Lancet. 2018;391(10127):1301-14.
- Xu XL, Liu XD, Liang M, Luo BM. Radiofrequency Ablation versus Hepatic Resection for Small Hepatocellular Carcinoma: Systematic Review of Randomized Controlled Trials with Meta-Analysis and Trial Sequential Analysis. Radiology. 2018;287(2):461-72.
- Uhlig J, Ludwig JM, Goldberg SB, Chiang A, Blasberg JD, Kim HS. Survival Rates after Thermal Ablation versus Stereotactic Radiation Therapy for Stage 1 Non-Small Cell Lung Cancer: A National Cancer Database Study. Radiology. 2018;289(3):862-70.
- Uhlig J, Kokabi N, Xing M, Kim HS. Ablation versus Resection for Stage 1A Renal Cell Carcinoma: National Variation in Clinical Management and Selected Outcomes. Radiology. 2018;288(3):889-97.
- Feng K, Yan J, Li X, Xia F, Ma K, Wang S, et al. A randomized controlled trial of radiofrequency ablation and surgical resection in the treatment of small hepatocellular carcinoma. J Hepatol. 2012;57(4):794-802.
- Jones J, Hunter D. Consensus methods for medical and health services research. BMJ. 1995;311(7001):376-80.
- Ahmed M, Solbiati L, Brace CL, Breen DJ, Callstrom MR, Charboneau JW, et al. Image-guided tumor ablation: standardization of terminology and reporting criteria–a 10-year update. Radiology. 2014;273(1):241-60.
- Clavien PA, Barkun J, de Oliveira ML, Vauthey JN, Dindo D, Schulick RD, et al. The Clavien-Dindo classification of surgical complications: five-year experience. Ann Surg. 2009;250(2):187-96.