Figures
Abstract
Background
The European Association for the Study of the Liver (EASL) criteria and the modified Response Evaluation Criteria in Solid Tumors (mRECIST) are currently adopted to evaluate radiological response in patients affected by HCC and treated with loco-regional procedures. Several studies explored the validity of these measurements in predicting survival but definitive data are still lacking.
Aim
To conduct a systematic review of studies exploring mRECIST and EASL criteria usefulness in predictive radiological response in HCC undergoing loco-regional therapies and their validity in predicting survival.
Methods
A comprehensive search of the literature was performed in electronic databases EMBASE, MEDLINE, COCHRANE LIBRARY, ASCO conferences and EASL conferences up to June 10, 2014. Our overall search strategy included terms for HCC, mRECIST, and EASL. Loco-regional procedures included transarterial embolization (TAE), transarterial chemoembolization (TACE) and cryoablation. Inter-method agreement between EASL and mRECIST was assessed using the k coefficient. For each criteria, overall survival was described in responders vs. non-responders patients, considering all target lesions response.
Results
Among 18 initially found publications, 7 reports including 1357 patients were considered eligible. All studies were published as full-text articles. Proportion of responders according to mRECIST and EASL criteria was 62.4% and 61.3%, respectively. In the pooled population, 1286 agreements were observed between the two methods (kappa statistics 0.928, 95% confidence interval 0.912–0.944). HR for overall survival (responders versus non responders) according to mRECIST and EASL was 0.39 (95% confidence interval 0.26–0.61, p<0.0001) and 0.38 (95% confidence interval 0.24–0.61, p<0.0001), respectively.
Conclusion
In this literature-based meta-analysis, mRECIST and EASL criteria showed very good concordance in HCC patients undergoing loco-regional treatments. Objective response according to both criteria confirms a strong prognostic value in terms of overall survival. This prognostic value appears to be very similar between the two criteria.
Citation: Vincenzi B, Di Maio M, Silletta M, D’Onofrio L, Spoto C, Piccirillo MC, et al. (2015) Prognostic Relevance of Objective Response According to EASL Criteria and mRECIST Criteria in Hepatocellular Carcinoma Patients Treated with Loco-Regional Therapies: A Literature-Based Meta-Analysis. PLoS ONE 10(7): e0133488. https://doi.org/10.1371/journal.pone.0133488
Editor: Seung Up Kim, Yonsei University College of Medicine, KOREA, REPUBLIC OF
Received: January 22, 2015; Accepted: June 29, 2015; Published: July 31, 2015
Copyright: © 2015 Vincenzi et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Data Availability: All relevant data are within the paper and its Supporting Information files.
Funding: The authors have no support or funding to report.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Hepatocellular carcinoma (HCC) represents today the fifth most common cancer diagnosis and the third most common cause of cancer-related deaths [1]. Several risk factors have been identified, including chronic hepatitis B and/or C viral infections, some inherited errors of metabolism (i.e. hemocromatosis, Wilson’s disease, α1-antitrypsin deficiency), primary hepatic immune disease and primary biliary cirrhosis [2]. More recently, a higher risk of liver cancer development has also been reported in patients affected by systemic metabolic syndrome, diabetes mellitus and non-alcoholic fatty liver disease [3]. Since 60%-80% of patients with newly diagnosed HCC have cirrhosis of the liver, ultrasonography and AFP testing every 6–12 months are routinely performed to promote an early detection of malignant nodule transformation in asymptomatic patients. Despite screening programs fewer than 20% of HCC are curable at the time of diagnosis and, given the presence of co-existent chronic liver disease in most cases, valuation of the underlying liver function is essential in therapeutical decision, since it can affect treatment efficacy and influence tolerability profile [4]. Current guidelines from the American Association for the Study of Liver Diseases for intermediate-stage HCC recommend loco-regional approaches for those patients with localized disease not suitable for surgical resection/transplantation [5]. By inducing alteration in local temperature (radiofrequency ablation, microwave ablation, cryoablation) or determining selective catheter-based infusion of particles in cancer supplying arterial branches (chemoembolization), these procedures lead to tumor necrosis and ensure disease control [6]. Radiological response is a well-recognised surrogate endpoint in the assessment of treatment efficacy in phase II studies, whereas survival remains crucial for phase III [7]. However conventional response evaluation criteria (WHO, World Health Organization and RECIST, Response Evaluation Criteria in Solid Tumors) have shown poor correlation with survival outcome in HCC patients, since they do not address measures of antitumor activity other than tumour shrinkage (which is based on the sum of bidimensional measurements of target lesions) [8]. To overcome this limitation, a modification of the response assessment was developed starting from 2001 in order to include the concept of tumour viability (tumoral tissue showing arterial uptake in the arterial phase of contrast-enhanced imaging techniques) and discriminate treatment efficacy from early failure [6]. The European Association for the Study of the Liver (EASL) criteria and the modified Response Evaluation Criteria in Solid Tumors (mRECIST) were adopted in the evaluation of radiological response in patients affected by HCC and treated with loco-regional procedures. EASL and mRECIST criteria differ from each others in terms of number of target lesions (all versus < = 2) and calculation method (bidimensional versus unidimensional) as reported in Table 1. Several studies [9, 10]indicate that evaluating the largest two lesions is generally the most useful procedure for measuring TACE responses under both EASL and mRECIST, even if the optimal number of lesions is not formally indicated in mRECIST criteria.
Up to now no large prospective validation is available for both mRECIST and EASL criteria and further studies are needed to confirm the validity of these measurements and their correlation with survival. Here we present a literature-based review gathering together all published retrospective studies comparing mRECIST and EASL criteria predictivity of tumor response and survival outcomes.
Methods
Selection of studies
Study selection was conducted according to the preferred reporting items for systematic reviews and meta-analyses (PRISMA) statement [11] (S1 PRISMA Checklist). A comprehensive search of the literature was performed in electronic databases EMBASE, MEDLINE and COCHRANE LIBRARY, from February 2010 up to June 2014. The references within the identified articles were then manually searched for additional studies. Our overall search strategy included terms for HCC, mRECIST, and EASL. To be eligible for inclusion, studies had to met the following criteria: (1) Loco-regional procedures included transarterial embolization (TAE), transarterial chemoembolization (TACE) and cryoablation, (2) response assessment after loco-regional treatments was evaluated according to both mRECIST and EASL criteria, (3) availability of data about overall survival and (4) reported k coefficient as measurement of mRECIST and EASL concordance, or availability (in the text or in a table) of the information needed to calculate it. Moreover, meeting abstracts presented in the most recent International Meetings (American Society of Clinical Oncology, American Association for the Study of Liver Diseases, and European Association for the Study of the Liver), personal presentation and no published data from ongoing study were explored and included if the above criteria were respected. No language limitation were observed, as all selected works were written in English.
Data extraction, clinical end points and quality assessment
By reading the full texts of the selected citations, two investigators (M.S. and L.D.) independently evaluated each identified for eligibility and quality, and then extracted the following data: name of all authors, year of publication, number of enrolled patients, type of loco-regional treatment, reported hazard ratio (HR) for overall survival (OS) according to mRECIST and EASL criteria and k coefficient of concordance in each study. Since progression free survival was not available for all studies, differently from overall survival outcome, these data were not extrapolated.
Statistical analysis
After data were abstracted, the authors proceed to their analysis using Review Manager (RevMan 5), the software used for preparing and maintaining Cochrane Reviews. Inter-methods concordance between similar categorical items of the two criteria was measured using the k coefficient. The strength of agreement based on k values was interpreted as follows: k less than 0.21, poor; k of 0.21–0.40, fair; k of 0.41–0.60, moderate; k of 0.61–0.80, good; and k greater than 0.80, excellent [12]. Hazard ratio (HR) for overall survival were used for meta-analysis, considering responders (complete or partial response) versus non responders patients, using the generic inverse variance outcome type in RevMan. To account for the heterogeneity of studies, a random-effects model was applied. For both mRECIST and EASL criteria, funnel plots were used to grossly exclude publication bias.
Results
Of 18 titles identified in the original search, 7 reports including 1357 patients were considered eligible for analysis (Fig 1). All trials, available as full-text articles, have been conducted retrospectively. The overall collection data period was March 2000 –June 2014. Diagnosis of HCC was confirmed by biopsy or radiologic imaging techniques according to the guidelines in each study.
11 studies were considered not eligible for analysis, as reported in Table 2. One was focused on the prognostic role of number of target lesions more than on that of response criteria [13]. Another study reported response and survival data at specific time point, so that it was not possible to extract the overall data [14]. In five works authors chose RECIST 1.0, RECIST 1.1 or volumetric RECIST version as comparator response criteria [15–19]. Lee IJ et al focused on correlation between radiologic response (according to WHO, RECIST, mRECIST and EASL criteria) and pathologic post resection viability [20]. Finally three studies compared the prognostic value of all known criteria (WHO, RECIST, mRECIST and EASL) to each other but it was not possible to draw all parameter we need for analysis [10, 21, 22]. However, data from one of these excluded studies [19] were obtained from a second, more recent publication [23] describing the same series of patients.
Baseline patients characteristics were homogeneous among retrospective cohorts and are listed in Table 3. General exclusions criteria were: inadequate target lesion (infiltrative pattern or largest lesion <1 cm); (2) presence of an additional primary malignancy in other organ; (3) presence of extrahepatic lesions; (4) presence of uncontrolled functional or metabolic disease. Of note, only two studies [13, 22] limited the inclusion to Child-Pugh class A patients, while the other studies included a minority of Child-Pugh B patients, and even a small percentage of Child-Pugh C patients in 1 case [9]. Most enrolled patients underwent TACE as initial therapy and then repeated TACE on demand at 4–8 weeks after the first cycle. One study [9] used drug-eluting beads (DEB) TACE. In all cases femoral artery approach was preferred. Arteries selective catheterization was followed by TACE following the internal guidelines of the different institution. Chemotherapeutics agents were doxorubicin (in three studies [9, 23, 24] or cisplatin (in two studies [22, 25]. Only one among the considered studies used cryoablation as loco-regional procedure [26]. Tumour measurements were performed according to the EASL and mRECIST criteria and assessment of response was carried out by contrast-enhanced spiral computed tomography (CT) or gadolinium-enhanced magnetic resonance imaging (MRI) after 4–8 weeks from treatment, depending on each study. In detail, as reported in Table 4, the procedure used was CT in the majority of cases, with the exception of one study using MRI [9] and two studies [13, 25] using CT or MRI (mostly CT in the study by Gillmore et al; no further details in the study by Kim BK et al). As shown in Table 4, most studies considered the overall response, while the study by Kim CJ et al [27] reported only the index lesion, that is currently validated for follow-up of patients with HCC [15] (and it is recommended by the American Association for the Study of Liver Diseases).
Response
Both criteria embraced the following four response categories: complete response (CR), partial response (PR), stable disease (SD) and progressive disease (PD). Objective response (OR) included both CR and PR. As reported in Table 1, according to RECIST, CR was defined as the absence of arterially enhanced areas; PR and PD, as a greater than 30% decrease and a greater than 20% increase, respectively, of the sum of the longest diameters of the enhancing target lesions; and SD, as neither PR nor PD. According to EASL criteria, PR and PD were defined as a greater than 50% decrease and a greater than 25% increase, respectively, of the sum of the cross products of the enhancing target lesions. The appearance of new HCC lesions denoted PD under both criteria, confirmed when their diameter exceeded 1 cm or when the lesion became at least 1 cm larger on progressive scans.
Table 5 shows the response assessed with EASL criteria after loco-regional therapy administration in the 7 considered studies. Table 6 shows the response assessed with mRECIST criteria after loco-regional therapy administration in the same studies.
The number of responders according to mRECIST and EASL criteria was 847 / 1357 (62.4%) and 832 / 1357 (61.3%), respectively.
Kappa statistics (available or calculated by data described in the paper in all the 7 studies) showed very high concordance between responses assessed by using EASL and mRECIST criteria (Table 7). In the pooled population of the 7 studies (Table 8), out of 1357 patients, 1286 agreements were observed between the two methods (94.77% of the observations, kappa statistics 0.928, 95% confidence interval 0.912–0.944).
Survival
With the limitation of the small number of studies included, funnel plots for both EASL and mRECIST estimates did not show asymmetry (S1 Fig and S2 Fig), so there was no clear evidence of publication bias.
Hazard Ratio for overall survival (responders versus non responders) according to mRECIST criteria (Fig 2) was 0.39 (95% confidence interval 0.26–0.61, p<0.0001), with a statistically significant heterogeneity among the studies (I2 82%, p<0.00001).
Similarly, Hazard Ratio for overall survival (responders versus non responders) according to EASL criteria (Fig 3) was 0.38 (95% confidence interval 0.24–0.61, p<0.0001), with a statistically significant heterogeneity among the studies (I2 84%, p<0.00001).
Discussion
Tumor response assessment is a cornerstone in cancer patient management, both in everyday clinical practice and as auxillary surrogate end point of survival for the evaluation of treatment efficacy in clinical studies.
Since new biological agents and loco-regional procedures exert their antitumoral activity by inducing tumour necrosis, with rare changes in volume shrinkage, traditional WHO and RECIST criteria do not always represent an appropriate tool for response evaluation, as they are based on dimensional criteria with no indication of lesion density changes. This concept has been very clearly described in several studies where both WHO and RECIST showed poor correlation with survival [8]. This observation, that can be assumed for most of solid cancer, becomes extremely relevant in HCC management, for which antiangiogenic drugs (i.e. sorafenib) and selective ablative procedures represent the standard of care for inoperable disease. To overcome these limitations, the EASL and mRECIST criteria have been suggested to be a better way of assessing tumor response in HCC patients. Several studies, recently reviewed [28], demonstrated their superiority over conventional criteria. The evaluation of tumour viability, represents the most relevant change in EASL/mRECIST criteria compared to traditional WHO/RECIST.
To the best of our knowledge, our meta-analysis is the first comprehensive paper aiming to address the superiority in assessing response of one criterion over the other. Since WHO and RECIST criteria are well recognized as inadequate, we only considered EASL and mRECIST methods and compare their predictivity of survival. Our results show how both EASL and mRECIST response evaluation methods can be of help in predicting long-term survival in HCC patients treated with TACE, with no proven advantage of one method over the other. Our meta-analysis also showed a statistically significant heterogeneity for both mRECIST and EASL criteria. This statistical heterogeneity could be related to relevant differences between the included studies, not only in terms of patients characteristics, but also in the technique used for the treatment procedure, and in the timing of instrumental assessment that, as shown in Table 4, was not the same among the series. However, each single study showed a better prognosis for responders compared to non-responders, with only quantitative heterogeneity in the hazard ratio. Heterogeneity of collection data together with retrospective nature of included studies, represent a large limitation for our work.
Give our results, we believe that EASL and mRECIST criteria deserve further evaluation as response assessment methods in HCC patients undergoing TACE, and that larger prospective trials should be encouraged in the future.
Conclusion
In this literature-based meta-analysis, mRECIST and EASL criteria showed a very good concordance in HCC patients undergoing loco-regional treatments. Objective response according to both criteria confirm a strong prognostic value in terms of overall survival. This prognostic value appears to be very similar between the two methods.
Supporting Information
S1 PRISMA Checklist. Checklist of preferred reporting items for systematic reviews and meta-analyses (PRISMA) statement.
https://doi.org/10.1371/journal.pone.0133488.s001
(PDF)
Author Contributions
Conceived and designed the experiments: BV MD MS LD CS MCP GD FC EM AR DS FP GT. Performed the experiments: BV MD MS LD CS MCP GD FC EM AR DS FP GT. Analyzed the data: BV MD MS LD CS MCP GD FC EM AR DS FP GT. Contributed reagents/materials/analysis tools: BV MD MS LD CS MCP GD FC EM AR DS FP GT. Wrote the paper: BV MD MS LD CS MCP GD FC EM GB AR DS FP GT.
References
- 1. Parkin DM, et al. Global cancer statistics, 2002. CA Cancer J Clin, 2005. 55(2): p. 74–108. pmid:15761078
- 2. Fattovich G., et al., Hepatocellular carcinoma in cirrhosis: incidence and risk factors. Gastroenterology, 2004. 127(5 Suppl 1): p. S35–50. pmid:15508101
- 3. Michelotti GA, Machado MV, Diehl AM. NAFLD, NASH and liver cancer. Nat Rev Gastroenterol Hepatol, 2013. 10(11): p. 656–65. pmid:24080776
- 4. Llovet JM, Burroughs A, Bruix J. Hepatocellular carcinoma. Lancet, 2003. 362(9399): p. 1907–17. pmid:14667750
- 5. Ribero D, et al., Selection for resection of hepatocellular carcinoma and surgical strategy: indications for resection, evaluation of liver function, portal vein embolization, and resection. Ann Surg Oncol, 2008. 15(4): p. 986–92. pmid:18236112
- 6. Bruix J, et al., Clinical management of hepatocellular carcinoma. Conclusions of the Barcelona-2000 EASL conference. European Association for the Study of the Liver. J Hepatol, 2001. 35(3): p. 421–30. pmid:11592607
- 7. Gonzalez-Guindalini FD, et al. Assessment of liver tumor response to therapy: role of quantitative imaging. Radiographics, 2013. 33(6): p. 1781–800. pmid:24108562
- 8. Lencioni R, Llovet JM. Modified RECIST (mRECIST) assessment for hepatocellular carcinoma. Semin Liver Dis, 2010. 30(1): p. 52–60. pmid:20175033
- 9. Prajapati HJ, et al. mRECIST and EASL responses at early time point by contrast-enhanced dynamic MRI predict survival in patients with unresectable hepatocellular carcinoma (HCC) treated by doxorubicin drug-eluting beads transarterial chemoembolization (DEB TACE). Ann Oncol, 2013. 24(4): p. 965–73. pmid:23223331
- 10. Shim JH, et al. Which response criteria best help predict survival of patients with hepatocellular carcinoma following chemoembolization? A validation study of old and new models. Radiology, 2012. 262(2): p. 708–18. pmid:22187634
- 11. Moher D, et al. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med, 2009. 6(7): p. e1000097. pmid:19621072
- 12.
DA . Practical Statistics for Medical Research. 1991, London, England: Chapman & Hall.
- 13. Kim BK, et al., Prospective comparison of prognostic values of modified Response Evaluation Criteria in Solid Tumours with European Association for the Study of the Liver criteria in hepatocellular carcinoma following chemoembolisation. Eur J Cancer, 2013. 49(4): p. 826–34. pmid:22995582
- 14. Boatta E, et al. Endovascular treatment of hepatocellular carcinoma with drug eluting microparticles (DC-Beads): CT evaluation of response to the treatment. Indian J Radiol Imaging, 2013. 23(2): p. 126–33. pmid:24082476
- 15. Shuster A, et al. Response Evaluation Criteria in Solid Tumors (RECIST) criteria are superior to European Association for Study of the Liver (EASL) criteria at 1 month follow-up for predicting long-term survival in patients treated with transarterial chemoembolization before liver transplantation for hepatocellular cancer. J Vasc Interv Radiol, 2013. 24(6): p. 805–12. pmid:23562641
- 16. Price TR, et al. Evaluation of response after stereotactic body radiotherapy for hepatocellular carcinoma. Cancer, 2012. 118(12): p. 3191–8. pmid:22025126
- 17. Riaz A, et al. Imaging response in the primary index lesion and clinical outcomes following transarterial locoregional therapy for hepatocellular carcinoma. JAMA, 2010. 303(11): p. 1062–9. pmid:20233824
- 18. Lin M, et al. Quantitative and volumetric European Association for the Study of the Liver and Response Evaluation Criteria in Solid Tumors measurements: feasibility of a semiautomated software method to assess tumor response after transcatheter arterial chemoembolization. J Vasc Interv Radiol, 2012. 23(12): p. 1629–37. pmid:23177109
- 19. Sato Y, et al. Tumor response evaluation criteria for HCC (hepatocellular carcinoma) treated using TACE (transcatheter arterial chemoembolization): RECIST (response evaluation criteria in solid tumors) version 1.1 and mRECIST (modified RECIST): JIVROSG-0602. Ups J Med Sci, 2013. 118(1): p. 16–22. pmid:23167460
- 20. Lee IJ. Correlating radiologic response criteria with pathologic tumor viability in HCC patients undergoing localized radiation followed by surgical resection. ASCO Meeting Abstracts, 2013. 31(4_suppl): p. 217.
- 21. Meza-Junco J, et al. Locoregional radiological treatment for hepatocellular carcinoma; Which, when and how? Cancer Treat Rev, 2012. 38(1): p. 54–62. pmid:21726960
- 22. Duke E, et al. Agreement between competing imaging measures of response of hepatocellular carcinoma to yttrium-90 radioembolization. J Vasc Interv Radiol, 2010. 21(4): p. 515–21. pmid:20172741
- 23. Choi J, et al. Clinical significance of the best response during repeated transarterial chemoembolization in the treatment of hepatocellular carcinoma. J Hepatol, 2014. 60(6): p. 1212–8. pmid:24486088
- 24. Jung ES, et al. Comparison of the methods for tumor response assessment in patients with hepatocellular carcinoma undergoing transarterial chemoembolization. J Hepatol, 2013. 58(6): p. 1181–7. pmid:23395691
- 25. Gillmore R. et al. EASL and mRECIST responses are independent prognostic factors for survival in hepatocellular cancer patients treated with transarterial embolization. J Hepatol, 2011. 55(6): p. 1309–16. pmid:21703196
- 26. Li H, et al. EASL and mRECIST responses are independent predictors of survival in hepatocellular carcinoma patients treated with cryoablation. Eur J Gastroenterol Hepatol, 2013. 25(5): p. 620–7. pmid:23325276
- 27. Kim CJ, et al. Radiologic response to transcatheter hepatic arterial chemoembolization and clinical outcomes in patients with hepatocellular carcinoma. Liver Int, 2014. 34(2): p. 305–12. pmid:23890360
- 28. Maida M, et al. Assessment of treatment response in hepatocellular carcinoma: a review of the literature. Future Oncol, 2013. 9(6): p. 845–54. pmid:23718305