doi: /hepr Response Evaluation Criteria in Cancer of the Liver (RECICL) (2015 Revised version)

Transcription

1 bs_bs_banner Hepatology Research 2016; 46: 3 9 doi: /hepr Special Report Response Evaluation Criteria in Cancer of the Liver (RECICL) (2015 Revised version) Masatoshi Kudo, Kazuomi Ueshima, Shoji Kubo, Michiie Sakamoto, Masatoshi Tanaka, Iwao Ikai, Junji Furuse, Takamichi Murakami, Masumi Kadoya and Norihiro Kokudo, for the Liver Cancer Study Group of Japan Committee for Response Evaluation Criteria in Cancer of the Liver, Liver Cancer Study Group of Japan, Department of Gastroenterology and Hepatology, Kinki University School of Medicine, Osaka-Sayama, Japan The Response Evaluation Criteria in Solid Tumors (RECIST) is inappropriate to assess the direct effects of treatment on hepatocellular carcinoma (HCC) by locoregional therapies such as radiofrequency ablation (RFA) and transarterial chemoembolization (TACE). Therefore, establishment of response evaluation criteria solely devoted to HCC is needed urgently in clinical practice as well as in clinical trials of HCC treatment, such as molecular-targeted therapies, which cause necrosis of the tumor. The Response Evaluation Criteria in Cancer of the Liver (RECICL) was revised in 2015 by the Liver Cancer Study Group of Japan based on the 2009 version of RECICL, which was commonly used in Japan. Major revised points of the RECICL 2015 is to define the target lesions of twolesionsperorganorthreelesionsperliver,uptoamaximum of five lesions. The second revised point is that setting the timing at which the overall treatment response has been changed. The third point is that the definition of treatment effect 1 has been changed to more than 50% tumor enlargement, excluding the area of necrosis after treatment. Overall evaluation of treatment response has been amended to make it possible to evaluate the overall response including extrahepatic lesions by systemic therapy, which is similar to RECIST or modified RECIST. We hope this new treatment response criteria, RECICL, proposed by the Liver Cancer Study Group of Japan will benefit HCC treatment response evaluation in the setting of daily clinical practice and clinical trials, not only in Japan, but also internationally. Key words: hepatocellular carcinoma, Liver Cancer Study Group of Japan, modified Response Evaluation Criteria in Solid Tumors, Response Evaluation Criteria in Cancer of the Liver, Response Evaluation Criteria in Solid Tumors INTRODUCTION THE FIRST VERSION of the Response Evaluation Criteria in Cancer of the Liver (RECICL) was compiled in October 1993 and published in a 1994 issue of Kanzo. 1 The revised versions of RECICL were subsequently published in Kanzo in and and have been used widely in routine clinical practice. As described in the 2004 RECICL and the version published in Hepatology Research, 4 the basic concepts of RECICL are that it: (i) is simple and sufficiently applicable in daily clinical practice; (ii) is internationally acceptable; (iii) is primarily for locoregional treatments (ethanol injection, microwave coagulation and radiofrequency ablation [RFA]) and Correspondence: Dr Masatoshi Kudo, Liver Cancer Study Group of Japan, Department of Gastroenterology and Hepatology, Kinki University School of Medicine, Ohno-Higashi, Osaka-Sayama, Osaka , Japan. m-kudo@med.kindai.ac.jp Received 11 March 2015; revision 26 May 2015; accepted 31 May 2015 transcatheter arterial therapy; (iv) is also applicable in radiation therapy and systemic chemotherapy as additional treatment methods; (v) provides separate criteria for the assessment of direct treatment effects on intrahepatic target lesions and overall response; and (vi) complies with the sixth edition of the General Rules for the Clinical and Pathological Study of Primary Liver Cancer. 10 However, because the 2009 version was not applicable in all clinical cases following the introduction of molecular-targeted therapy as a systemic chemotherapy for hepatocellular carcinoma (HCC), we recently revised RECICL for the fourth time. The revision process this time is in line with the concepts used for the 2009 version but includes relatively extensive revision of the definition of treatment effect 1 (TE1) and the assessment of overall response. In Europe, the World Health Organization (WHO) Handbook for Reporting Results of Cancer Treatment in Solid Tumors (hereinafter the WHO criteria) 5 and Response Evaluation Criteria in Solid Tumors (RECIST, version 1.0) 6 have been used conventionally to assess treatment response in 3

2 4 M. Kudo et al. Hepatology Research 2016; 46: 3 9 HCC. RECIST 1.0 was simplified further, and version is used widely today. However, the WHO criteria and RECIST are both inappropriate for the assessment of direct treatment effects on the target HCC lesion because they use only tumor size reduction as the criterion for treatment effect and do not recognize tumor necrosis as a treatment effect even though the necrosis of HCC is frequently caused by various locoregional treatments. For example, it has been shown that pathological necrosis does not correlate with tumor size reduction assessed by the WHO criteria and RECIST after transarterial chemoembolization (TACE) using lipiodol. 8 For this reason, establishing response evaluation criteria that take into consideration the necrosis of HCC is urgently needed. In 2010, Lencioni et al. proposed a modified version of RECIST (mrecist) that incorporates the necrosis of tumors as a treatment effect into RECIST 1.0 to make the appropriate criteria for HCC. 9 However, the criteria used to evaluate necrosis lack accuracy because unidirectional measurements were used. On the other hand, since the 1990s, tumor necrosis has been assessed using bidirectional measurements to improve the accuracy of treatment response evaluation in RECICL. Therefore, RECICL may be more specific to HCC. In recent years, treatment methods for HCC have changed greatly because of the adoption of molecular-targeted agents. Unlike cytotoxic anticancer drugs, molecular-targeted agents cause not only size reduction but also necrosis in tumors due to their antiangiogenic property. In other words, there is a greater need to establish proper evaluation criteria for tumor necrosis. In the 2009 version, the response evaluation criteria for systemic chemotherapy were supplemented by RECIST. In this revision, RECICL was modified to be all-in-one evaluation criteria that are applicable to all treatment methods by incorporating the overall evaluation criteria of RECIST. RECICL is now applicable to extrahepatic lesions and its features that are specific tohccremain. The high incidence of multicentric tumors and of intrahepatic recurrence and metastasis is a biological feature specific to HCC and not to other types of cancer. Consequently, it is not always appropriate to interpret the appearance of a new intrahepatic lesion in the untreated area in the liver as progressive disease (PD). The major goal of RECICL is that it focuses only on the efficacy of various treatment methods (locoregional treatments) for target lesions, and also correlates the treatment effects with prognosis as much as possible in assessing the overall response. This is different from RECIST, which is used to assess the efficacy of one particular therapy. With RECIST, when current therapy is considered ineffective when PD is diagnosed, changing to another treatment method may be considered. Unlike systemic chemotherapy, treatment by transcatheter arterial therapy or ablation therapy is not performed throughout the liver and therefore does not affect newly developed lesions in untreated areas of the liver. In other words, when the latter therapies are repeated for new target lesions, the treatment response of the new lesions is expected to be the same as that observed in previous target lesions. Therefore, even if a new target lesion (i.e. multicentric lesion) appears in a different area, it should neither represent an indication for changing the treatment method nor be used to determine prognosis. This point should be considered when evaluating treatment response in HCC, and RECICL provides the evaluation criteria that fully consider these specific characteristicsofhcc. MAJOR REVISED POINTS IN THE 2015 VERSION OF RECICL (TABLE 1) 1 Up until the 2009 version of RECICL, tumor enlargement of 25% or more was defined as TE1 and was treated as PD. However, a bidirectional enlargement of 25% is roughly equivalent to a unidirectional enlargement of 11% and is a stricter criterion than the definition of 20% enlargement for PD in RECIST 1.1. This 20% unidirectional enlargement in RECIST 1.1 would be equal to a 44% bidirectional enlargement ( = 1.44 times). In this revision, we used an enlargement of 50% as the criterion for TE1 because of its consistency with RECIST and its usability in clinical settings (Table 2). 2 Up until the 2009 version, lesions subject to evaluation (target lesions) were intrahepatic lesions, and when multiple intrahepatic lesions are present, five lesions from largest to smallest were selected as target lesions. However, in accordance with RECIST 1.1, two lesions per organ and a maximum of five lesions in total will be determined as target lesions in the 2015 version. If three or more intrahepatic lesions are present, a maximum of three lesions should be counted as target lesions to comply with the criterion of three or less lesions that are 3 cm or less in the Milan criteria and as an indication for ablation therapy. 3 Inthe2015version,itwasclearlystatedthatresponseto locoregional treatments should be assessed 1 3months after completion of therapy or at the last treatment if a series of treatments were given. In chemotherapy (including hepatic arterial infusion), treatment response should be assessed 1 3 months after the first administration of the anticancer agents. If the therapy is ongoing, the assessment should be repeated every 1 3months. The efficacy of radiation therapy is

3 Hepatology Research 2016; 46: 3 9 RECICL (2015 revised version) 5 Table 1 Differences between Response Evaluation Criteria in Cancer of the Liver (RECICL) 2009 and 2015 versions RECICL 2009 version RECICL 2015 version Target lesions Intrahepatic lesions only Intrahepatic and extrahepatic lesions Maximum of five intrahepatic lesions Two lesions per organ or three lesions per liver, up to a maximum of five lesions Measurement direction Bidirectional (same as 2009 version) Assess the cross-sectional area of the (same as 2009 version) Measurement method tumor by multiplying the major axis of the (same as 2009 version) maximum cross-section and the maximum diameter crossing the major axis at a right angle. However, the unstained region and the region of lipiodol accumulation without washout are regarded as necrotized regions. Time for evaluation Immediately after ablation therapy One to three months after ablation therapy One month or more after TACE or hepatic arterial infusion chemotherapy One to three months after TACE or hepatic arterial infusion chemotherapy Six months or less after radiation therapy (same as 2009 version) Treatment response of target lesions (treatment effect [TE]) TE4 Tumor necrosis of 100% or 100% reduction (same as 2009 version) in tumor size TE4a Necrotized area larger than an original tumor (same as 2009 version) (enough ablative margin) TE4b Necrotized area similar in size to an original (same as 2009 version) tumor (insufficient ablative margin) TE3 Tumor necrosis of % or % (same as 2009 version) reduction in tumor size TE2 Effect other than TE3 or TE1 (same as 2009 version) TE1 Tumor enlargement of >25% regardless of necrosis Tumor enlargement of >50% (excluding the area of necrosis after treatment) assessed based on the best treatment response observed during the 6 months of therapy after treatment initiation. 4 Although the evaluation criteria for overall treatment response were supplemented by RECIST in the RECICL 2009 version, we developed new evaluation criteria in line with RECIST 1.1 for the 2015 version. For new lesions that developed after ablation therapy or transarterial chemoembolization, it is important to describe whether the new lesion is (i) inside or (ii) outside the previously treated area to make the information useful in future revisions of the current criteria. FULL TEXT OF RESPONSE EVALUATION CRITERIA IN CANCER OF THE LIVER Subjects SUBJECTS ARE PATIENTS who are treated for the primary lesion and for recurrence. As a rule, contrastenhanced computed tomography (CT) is performed to assess treatment response in RECICL, the principle targets in the assessment should be clearly visualized on the images, and should be hypervascular in intrahepatic and extrahepatic lesions. Among all measurable lesions, two lesions per organ and a maximum of five lesions in total are defined as target lesions. If there are three or more lesions in the liver, three nodules should be included in the target lesions. The area of individual target lesions is calculated by multiplying the length of the major axis by the maximum diameter crossing the major axis at a right angle, and the sum of the areas in all target lesions is used as the baseline area. All the remaining lesions are regarded as non-target lesions. Detailed description Description of past medical history 1 Methods and date when the definitive diagnosis of HCC was made.

4 6 M. Kudo et al. Hepatology Research 2016; 46: 3 9 Table 2 Assessment of direct treatment effect on target lesions Treatment effect (TE) TE4 Tumor necrosis of 100% or 100% reduction in tumor size TE4a: necrotized area larger than the original tumor (sufficient ablative margin) TE4b: necrotized area similar in size to the original tumor (insufficient ablative margin) TE3 Tumor necrosis of % or % reduction in tumor size TE2 TE1 Effect other than TE3 or TE1 Tumor enlargement of 50% (excluding the area of necrosis after treatment) A unidirectional enlargement of 20%, which is defined as progressive disease in RECIST 1.1, is equivalent to a bidirectional enlargement of 44% ( = 1.44 times). Therefore, in the new version, the enlargement of a tumor by 50% is defined as TE1 to be consistent with internationally accepted criteria, RECIST and to be easy to use in the clinical setting. 2 Previous treatment modality (as described in Description of treatment modalities ). 3 Dates of initiation and completion of previous treatment. 4 Methods and date when recurrence was diagnosed. Descriptions of HCC at the time of treatment initiation In accordance with the sixth edition of the General Rules for the Clinical and Pathological Study of Primary Liver Cancer (edited by Liver Cancer Study Group of Japan), 10 the descriptions should include the following items: 1 Tumor location. 2 Tumor size, number and vascular invasion. The tumor size is presented as the major axis and maximum diameter crossing the major axis at a right angle. 3 Macroscopic classification. 4 Tumor stage (tumor node metastasis [TNM]). Even for tumors that are only assessable by imaging, TNM staging should be described similar to surgical findings and those of the resected specimen. 5 Histological classification or differentiation. Description of treatment modalities 1 Treatment names: transarterial therapy (hepatic arterial infusion chemotherapy, transarterial embolization [TAE], and transarterial chemoembolization [TACE], and local ablation treatments such as percutaneous ethanol injection, microwave coagulation, and radiofrequency ablation [RFA]); radiation therapy; and systemic chemotherapy (including molecular targeted therapy) 2 Treatment details: for treatments using drugs, the name of the drugs* (anticancer drugs, lipiodol, etc.), route of administration, treatment interval, dose, total number of administrations, and total dose should be described. For other treatment methods, the details should be described appropriately. When the treatment is discontinued, the reason for discontinuation and the presence or absence of adverse effects should be described. *Chemotherapeutic agents, any agent directly injected into the tumor to induce necrosis such as ethanol, and/or embolization material used in transarterial chemoembolization should be described. 3 Dates of initiation and completion or termination of each treatment. Assessment of direct treatment effect on the target lesion 1 To assess the treatment effect on the target nodule, the tumor-necrotizing effect or the rate of tumor size reduction are calculated based on the size reduction or disappearance of arterial enhancement of the nodule on dynamic contrast-enhanced CT. In the assessment, the findings of dynamic magnetic resonance imaging (MRI) or contrast-enhanced ultrasonography (US) can substitute the findings of dynamic contrast-enhanced CT. 2 The necrotizing effect is assessed by diagnostic imaging. The percent ratio of the necrotized area to the cross-sectional area of the tumor should be calculated.* *When various cross-sections are obtained for a single tumor, the total sum of the necrotic areas should be used; however, when the maximum cross-section represents the entire findings of the tumor, assessment could be based on the maximum cross-sectional area. 3 The size reduction rate is calculated using the equation below after calculating the size by multiplying the major axis of the maximum cross-section by the maximum diameter crossing the major axis at a right angle: Size reduction rate ¼ðsize ½ before treatmentš ½size after treatmentšþ= ðsize before treatmentþ100 4 Target nodule treatment response (treatment effect [TE]). Effects on individual lesions are categorized into four categories based on tumor size reduction observed within a

5 Hepatology Research 2016; 46: 3 9 RECICL (2015 revised version) 7 predetermined period* after initiation of treatment or the maximum tumor-necrotizing effect.** As shown in Table 2, TE4 is equivalent to a complete response (CR) and is defined as 100% tumor-necrotizing effect or 100% tumor size reduction. TE3 is equivalent to partial response (PR) and defined as % tumornecrotizing effect or % tumor size reduction. TE2 is regarded as stable disease (SD) if the effect is neither PR nor PD. TE1 corresponds to PD with an increase in tumor size of 50% or more,*** excluding the area of treatment-induced necrosis. *Treatment effects are assessed 1 3months after local ablation treatments (percutaneous ethanol injection, microwave coagulation and RFA), hepatic arterial infusion chemotherapy with or without lipiodol, TAE, TACE or systemic chemotherapy. When a series of treatments are performed, the effects should be also assessed 1 3 months after completion of the last treatment. The effect of radiation therapy is assessed using the best response observed during the 6 months after treatment initiation. **For local ablation treatments, when the nonstained, low-density area is wider across the entire circumference than the low-density area in the late phase of pretreatment CT scan (sufficient ablative margin), the lesion is regarded as 100% necrotized (TE4a). Although the ideal sufficient ablative margin is more than 5 mm from an edge around the tumor, presence of a smaller sized non-stained area of 5 mm can be regarded as a sufficientmargininthis criteria. The disappearance of hypervascularity alone with the lack of a wider non-stained region compared with the low-density area on CT scan is judged as TE4b (insufficient ablative margin). The effect of TAE or TACE is judged as 100% tumor-necrotizing effect (TE4) when the following findings are present: (i) complete tumor disappearance; (ii) lack of tumor hypervascularity on contrast-enhanced CT; (iii) strongly dense lipiodol accumulations; or (iv) shrinkage of lipiodol-accumulated areas across the lesion when lipiodol is used. The effect of radiation therapy is also evaluated using the criteria for necrotizing effects when tumor necrosis is observed. ***Up until the 2009 version of RECICL, tumor enlargement of 25% or more was defined as TE1 and was treated as PD. However, a bidirectional enlargement of 25% is roughly equivalent to a unidirectional enlargement of 11%, and is stricter than the PD criterion of 20% enlargement in RECIST 1.1. To ensure consistency with RECIST 1.1, a unidirectional enlargement of 20% will be equivalent to a bidirectional enlargement of 44% ( = 1.44 times).therefore,inthe2015versionofrecicl,a 50% enlargement in tumor size is used as the criterion for TE1 because treatment is continued in most patients with PD based on RECIST and because of the usability in clinical settings. 5 When multiple intrahepatic lesions are present, the treatment effect on target nodules is assessed individually for up to a maximum of three lesions. Overall evaluation of treatment response 1 From the perspective of intrahepatic and extrahepatic treatment effects and the duration of treatment effects, overall treatment response is assessed based on the four-grade system using CR, PR, SD and PD, as shown in Table 3. 2 To use this method to predict the prognosis, TE is determined and recorded at 1 3 months after treatment initiation to assess overall response. However, the effect of radiation therapy is assessed based on the maximum effect observed within 6 months of treatment initiation. 3 Definition of new lesions. In dynamic CT, dynamic MRI (including gadoxetic acid-enhanced magnetic resonance imaging [Gd-EOB-MRI]) and Sonazoid-enhanced US, new intrahepatic lesions are typically observed as hypervascularity in the early phase and washout in the late phase (low-intense nodules on Gd-EOB-MRI hepatocyte phase images or Kupffer defect on Sonazoidenhanced US images). As a rule, new intrahepatic lesions should be nodules of 10 mm or more. When a new intrahepatic lesion appears after ablation therapy or TACE, it is important to determine whether the location is inside or outside the previously treated area. Extrahepatic lesions should be nodules of 10 mm or more, and lymph nodes with a minor axis of 15 mm or more are regarded as lymph node metastasis. Table 3 Overall response (evaluation based on the maximum response obtained over a period of 1 3 months or within 6 months in case of radiation therapy) Target lesions Non-target lesions New lesions Overall response TE4 TE4 No CR TE4 TE3, TE2 No PR TE3 Non-TE1 No PR TE2 Non-TE1 No SD TE1 Any Yes or no PD Any TE1 Yes or no PD Any Any Yes PD CR, complete response; PD, progressive disease; PR, partial response; SD, stable disease; TE, treatment effect.

6 8 M. Kudo et al. Hepatology Research 2016; 46: 3 9 Table 4 Reference table for the assessment of treatment effect in liver cancer AFP, α-fetoprotein; AFP-L3, Lens culinaris agglutinin-reactive fraction of AFP; CR, complete response; DCP, des-γ-carboxyprothrombin; PD, progressive disease; PIVKA-II, protein induced by vitamin K absence/antagonist-ii; PR, partial response; SD, stable disease. Detailed criteria Necrotizing effects are assessed in accordance with the response evaluation criteria for the treatment of a target nodule. 1 The presence of a non-enhancing area corresponding to the original nodule after treatment on dynamic contrast-enhanced CT using an i.v. bolus injection is regarded as a necrotizing effect. Such non-stained, low-density areas are enhanced less intensely than

7 Hepatology Research 2016; 46: 3 9 RECICL (2015 revised version) 9 the surrounding liver parenchyma in the early and late phases.* In other words, the CT attenuation value of a non-stained, low-density area does not change due to contrast enhancement. ***The early phase of contrast-enhanced CT using a bolus injection represents the hepatic arterial dominant phase, whereas the late phase represents the equilibrium phase. 2 When lipiodol is used, the presence of a region retaining lipiodol homogeneously and densely in the tumor on CT images taken 1 month after therapy is regarded as a necrotizing area. Contrast-enhanced MRI or contrast-enhanced US can be used as an alternative. 3 The lowest levels of three tumor markers (α-fetoprotein [AFP], protein induced by vitamin K absence/ antagonist-ii and Lens culinaris agglutinin-reactive fraction of AFP) within 3 months (6 months in the case of radiation therapy) described in the sixth revised edition of the General Rules for the Clinical and Pathological Study of Primary Liver Cancer 10 should be recorded as reference values for the overall response evaluation (Table 4). REFERENCES 1 Liver Cancer Study Group of Japan. Response evaluation criteria in cancer of the liver. Kanzo 1994; 35 (2): (in Japanese) 2 Liver Cancer Study Group of Japan. Response evaluation criteria in cancer of the liver. Kanzo 2004; 45 (7): (in Japanese) 3 Kudo M, Kubo S, Takayasu K et al. Liver cancer study group of Japan, response evaluation criteria in cancer of the liver (2009 revised version). Kanzo 2010; 51 (5): (in Japanese) 4 Kudo M, Kubo S, Takayasu K, et al. Response evaluation criteria in cancer of the liver (RECICL) proposed by the Liver Cancer Study Group of Japan (2009 revised version). Hepatol Res (7): WHO. Handbook for Reporting Results of Cancer Treatment.: Geneva, SwitzerlandWorld Health Organization Offset Publication, Therasse P, Arbuck SG, Eisenhauer EA, et al. New guidelines to evaluate the response to treatment in solid tumors. European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada. J Natl Cancer Inst (3): Eisenhauer EA, Therasse P, Bogaerts J, et al. New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer (2): Takayasu K, Arii S, Matsuo N, et al. Comparison of CT findings with resected specimens after chemoembolization with iodized oil for hepatocellular carcinoma. AJR Am J Roentgenol (3): Lencioni R, Llovet JM. Modified RECIST (mrecist) assessment for hepatocellular carcinoma. Semin Liver Dis (1): Liver Cancer Study Group of Japan. General Rules for the Clinical and Pathological Study of Primary Liver Cancer, 6th revised edn. Tokyo: Liver Cancer Study Group of Japan, Kanehara & Co., (in Japanese)