LIVER TRANSPLANTATION 21:1259–1269, 2015
ORIGINAL ARTICLE
Transarterial Chemoembolization With DrugEluting Beads Is Effective for the Maintenance of the Milan-In Status in Patients With a Small Hepatocellular Carcinoma ,1 Matteo Angelo Manini,1 Angelo Sangiovanni,1 Laura Martinetti,2 Davide Vigano 1 1 1 2 Vincenzo La Mura, Alessio Aghemo, Massimo Iavarone, Silvia Crespi, Antonio Nicolini,2 and Massimo Colombo1 Divisions of 1Gastroenterology and Hepatology and 2Radiology, Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico, University of Milan, Milan, Italy
Transarterial chemoembolization (TACE) is the standard of care for the treatment of patients with an intermediate (Barcelona Clinic Liver Cancer [BCLC] B) hepatocellular carcinoma and to bridge patients with an early cancer to liver transplantation (LT). We explored the efficacy of TACE with drug-eluting beads (DEB) in BCLC A patients. Included are all BCLC A patients unsuitable for resection or locoregional ablation who underwent a DEB TACE between 2006 and 2012. Treatment was carried out “a la demande” until complete tumor devascularization or progression beyond Milan criteria. In patients with a complete response (CR), a contrast computed tomography (CT) scan was repeated at 3-month intervals during the first 2 years and then every 6 months alternating with abdominal ultrasound in the subsequent 3 years. Fifty-five patients had 79 tumor nodules ranging 7 to 50 mm; 32 (58%) achieved a CR that was maintained up to 4 and 7 months in 21 (38%) and 17 (31%) patients, respectively. The 24- and 36-month tumor-free survivals were 21% and 9%, respectively. The overall cumulative progression beyond Milan criteria at 3, 6, 12, and 24 months was 2%, 5%, 30%, and 54%. LT eligibility was maintained for a median of 19 months (range, 2-63 months). CR to first TACE was the strongest independent predictor of Milan-in maintenance. In conclusion, DEB TACE may effectively bridge patients with an early cancer to LT, and a CR to the first procedure may guide patient prioritization during the waiting list. Liver Transpl 21:1259-1269, C 2015 AASLD. 2015. V Received February 23, 2015; accepted May 31, 2015.
Additional supporting information may be found in the online version of this article. Abbreviations: AFP, alpha-fetoprotein; BCLC, Barcelona Clinic Liver Cancer; CI, confidence interval; CT, computed tomography; cTACE, conventional transarterial chemoembolization; CTCAE, Common Terminology Criteria for Adverse Events; CR, complete response; DEB, drug-eluting bead; HBV, hepatitis B virus; HCC, hepatocellular carcinoma; HCV, hepatitis C virus; HR, hazard ratio; IHG, intrahepatic growth; IRMA, immunoradiometric assay; LT, liver transplantation; MELD, Model for End-Stage Liver Disease; mRECIST, modified response evaluation criteria in solid tumors; MRI, magnetic resonance imaging; NIH, new intrahepatic lesion; OLT, orthotopic liver transplantation; OS, overall survival; PEI, percutaneous ethanol injection; PS, performance status; PVT, portal venous thrombosis; RFA, radiofrequency ablation; SD, standard deviation; TACE, transarterial chemoembolization; TAE, transarterial embolization; TTP, time to progression; US, ultrasound. Conflicts of interest: Massimo Colombo: Grant and research support—BMS, Gilead Science; Advisory committees—Merck, Roche, Novartis, Bayer, BMS, Gilead Science, Tibotec, Vertex, Janssen Cilag, Achillion, Lundbeck, GSK, GenSpera, AbbVie, AlfaWasserman, Jennerex; Speaking and teaching—Tibotec, Roche, Novartis, Bayer, BMS, Gilead Science, Vertex, Merck, Janssen, Sanofi. Alessio Aghemo: Advisory Board—AbbVie, Gilead Sciences; Speaker Bureau—AbbVie, Gilead Sciences, Janssen, Merck. Address reprint requests to Massimo Colombo, M.D., Division of Gastroenterology and Hepatology, Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico, University of Milan, Via Francesco Sforza 35, 20122 Milan, Italy. Telephone: 39-0255035432; FAX: 39-0250320410; E-mail:
[email protected] DOI 10.1002/lt.24196 View this article online at wileyonlinelibrary.com. LIVER TRANSPLANTATION.DOI 10.1002/lt. Published on behalf of the American Association for the Study of Liver Diseases
C 2015 American Association for the Study of Liver Diseases. V
1260 MANINI ET AL.
Treatment of hepatocellular carcinoma (HCC) in patients with cirrhosis is guided by a disease stage classification (Barcelona Clinic Liver Cancer [BCLC] classification) based on tumor burden, liver impairment, and performance status (PS).1-5 Potentially curative treatments like liver transplantation (LT), surgical resection, and local ablation are the standard of care for patients with an early detected tumor in the BCLC A stage, whereas patients with a HCC in the intermediate BCLC B stage, who would not benefit from radical therapies, are more appropriately forwarded to palliative treatments such as conventional transarterial chemoembolization (cTACE) or transarterial chemoembolization (TACE) with drug-eluting beads (DEBs), both representing useful options to improve survival.6,7 In real-life practice, however, a number of factors may cause this therapeutic algorithm to be violated to the point that up to half of the patients with an early HCC ultimately will not have access to potentially curative treatments as a consequence of the scarcity of liver donors for LT or conditions like liver failure, strategic tumor localization, or presence of comorbidities which unbalance the risk benefit ratio of most radical treatments.8,9 Not surprisingly, therefore, TACE has gained popularity in the treatment of patients with an early HCC who are unfit or unwilling to receive radical therapies. The reverse is also true, however, as a number of BCLC B patients have undergone repeat TACE courses with the aim of downsizing the tumor burden to reach an operable stage or to prevent progression of an early tumor to an inoperable stage while waiting for a LT. The application of superselective TACE with DEB loaded with doxorubicin (Biocompatibles UK, Farnham, Surrey, UK) has recently emerged as an effective treatment modality for intermediate stage HCC because of its ability of inducing a significant locoregional ischemic injury, secondary to a prolonged occlusion of the vessels feeding the target tumor, combined with a slow and prolonged release of chemotherapeutic agents, leading to high locoregional drug concentration.10-13 Consequently, DEB TACE results in a longer-lasting tumor necrosis effect as well as an attenuation of liver and systemic drug-related toxicity,10,13-17 without, however, providing increased survival benefits or better radiological response rates compared to cTACE.13,16-18 With the exception of few retrospective studies,19,20 most reports on TACE, focusing on rates of radiological response and overall survival (OS),11-13,21,22 did overlook such important clinical outcomes in cancer therapy such as rates of posttreatment tumor progression and recurrence after tumor eradication. These are meaningful clinical predictors for patients listed for LT in need of bridge treatments to maintain the tumor burden within the criteria of operability. With this topic as a background, we aimed to evaluate the efficacy of DEB TACE as a therapy to maintain patients with HCC in the BCLC A stage within the internationally defined Milan criteria, using the time to progression (TTP) beyond Milan as the primary endpoint.
LIVER TRANSPLANTATION, October 2015
PATIENTS AND METHODS Study Design This is a single-center cohort study of patients with compensated cirrhosis with a first diagnosis of an early HCC, who between January 2006 and January 2012 consecutively underwent DEB TACE as first-line treatment. Criteria for treatment were the presence of a viable tumor detected as hyperenhancement at the arterial phase by computed tomography (CT) scan or magnetic resonance imaging (MRI), whereas outcomes were evaluated with a time-scheduled imaging and clinical follow-up. DEB TACE cycle was continued until achievement of the complete ablation of the tumor, progression beyond Milan criteria, or whenever vascular contraindications, severe adverse events, hepatic decompensation, or liver function impairment occurred. Response to treatment was assessed 1 month after each DEB TACE procedure by CT scan, followed by MRI in the case of equivocal results. After achieving a complete necrosis of the nodules, a CT scan was scheduled at 3-month intervals during the first 2 years and then every 6 months alternating with ultrasound (US) for the subsequent 3 years. Thereafter, tumor-free patients were reallocated to a standard US surveillance program. Patients were referred to repeat DEB TACE whenever hyperenhancement at the arterial phase, suggestive of viable tumor tissue, or a new HCC nodule was detected, following an “a la demande” standard protocol. Censored from the TTP analysis were all patients who received a LT and those who were submitted to other than DEB TACE treatment for either local tumor recurrence or a de novo HCC, whereas survival analysis included patients who underwent second- and third-line approaches too. A written informed consent was obtained before each radiological diagnostic and interventional procedure. The study was approved by the Ca’ Granda Ospedale Maggiore Policlinico institutional review board and complied with the provisions of the Good Clinical Practice guidelines and the Declaration of Helsinki.
Endpoints The primary endpoint was TTP beyond Milan criteria. Secondary outcomes included treatment response at the first month imaging; rates of maintenance of the radiological complete response (CR) during follow-up; safety, patient tolerance, and liver/systemic toxicity; and OS.
Selection Criteria All Child-Pugh A or B patients with cirrhosis with a first diagnosis of a single HCC 5 cm in diameter or up to 3 HCCs 3 cm (Milan-in status) who were unfit for resection or locoregional ablation with radiofrequency ablation (RFA) or percutaneous ethanol injection (PEI) were consecutively evaluated for enrollment. Contraindications to locoregional treatment were the absence of target lesion or a multinodular liver at US,
LIVER TRANSPLANTATION, Vol. 21, No. 10, 2015
ascites, nodules not easily accessible or considered at high risk of complications due to location, or lesions close to the main portal or hepatic vein branches. DEB TACE was the bridge treatment of choice in patients with a multifocal HCC who were listed for LT. Exclusion criteria were age more than 75 years, arterial hypovascularized HCC, hepatic encephalopathy >1 grade, refractory ascites, serum total bilirubin level > 3 mg/dL, serum albumin level < 2.8 g/L, serum creatinine level > 2.0 mg/dL, a variceal bleeding in the preceding 3 months, previously treated HCC, history of heart or kidney impairment, vascular contraindications like hepatic artery thrombosis, complete thrombosis of the main portal branches, relevant portosystemic shunts, active infection or sepsis, hemoglobin level < 8 g/dL, and Eastern Cooperative Oncology Group PS grade 1.23 Patients with partial thromboplastin time or a prothrombin time ratio greater than 1.6 and a platelet count lower than 50 3 106/L received fresh frozen plasma and platelet concentrate to balance hemostasis.
Preliminary Investigations A detailed medical history was recorded for all patients. Routine blood chemistries, including liver tests and serum alpha-fetoprotein (AFP) assay by IRMA (Abbott, Chicago, IL; normal values, 7 ng/mL), were performed within 2 weeks before each DEB TACE to meet with the inclusion/exclusion criteria and generate a database to monitor liver function and treatmentrelated toxicity. Liver function was assessed by ChildPugh classification.24
HCC Diagnosis and Staging HCC was diagnosed according to American Association for the Study of Liver Diseases guidelines, depending on the time period of study enrollment.3,25 In each patient, the tumor was staged by abdominal/chest CT scan and bone scintigraphy and classified according to BCLC.1 Each scan was performed with a 64multidetector row CT (Definition Siemens, Erlangen, Germany). Images of the liver were acquired as previously described26 and were read blindly by 2 expert radiologists. Whenever required, a fine-needle biopsy was performed using a 21-gauge trenchant needle (Biomol, HS Hospital Service, Rome, Italy) for histological examination of both intranodular and extranodular tissue. Histological diagnosis was made according to the International Working Party criteria.27
Embolization Technique DEB TACE procedures were performed by 2 expert interventional radiologists with 20 and 10 years of experience, respectively (A.N. and S.C.). Through a right femoral artery access and after placement of a 5F cobra catheter (Terumo, Tokyo, Japan), the celiac trunk was examined through the arterial and venous phases to define hepatic artery anatomy and to assess
MANINI ET AL. 1261
portal vein patency. Then the hepatic artery was selectively catheterized to study the arterial supply of the target lesions. A coaxial microcatheter (Progreat, Terumo, Tokyo, Japan) was advanced in every feeding artery to allow embolization of the lesion with drugeluting microspheres (DC beads, Biosphere, 100-300 micrometers in diameter, 1 vial loaded with 50 mg of epirubicin), until a complete stop of blood flow was obtained. Multifocal tumors were simultaneously treated in the same session. A final study of the proper hepatic artery was performed.
Assessment of Outcome and Follow-up All patients were prospectively followed up to January 2015, until death or LT with clinical examination, serum chemistry, CT scan, and examinations qualified to diagnose extrahepatic metastases. A radiological response to treatment was assessed by CT scan 1 month after each DEB TACE procedure and afterward retrospectively classified according to modified response evaluation criteria in solid tumors (mRECIST).28 The causes of tumor progression were analyzed through the identification of 3 distinct patterns: increase of the sum of the diameters of viable target lesions (intrahepatic growth [IHG]), newly detected intrahepatic focal liver lesion, new extrahepatic lesion, and/or vascular invasion.29 All the new intrahepatic lesions (NIHs) within 1 cm of an original HCC nodule were classified as satellites. In patients undergoing LT, a response to treatment was confirmed by histological examination of the explanted liver, as previously described.30 Following each procedure, patients were hospitalized for 24 hours and then reevaluated with physical examination and blood test at day 14 and 30. Safety was monitored recording postprocedure clinical complications and liver/renal function. Afterward, the severity of complications were retrospectively graded according to the Common Terminology Criteria for Adverse Events (CTCAE), version 4.0 (v4.03; June 14, 2010).31 An adverse event was considered treatment related if it occurred within 30 days from DEB TACE.
Statistical Analysis Patients’ baseline characteristics were expressed as median. Categorical variables were compared by the Yates’ v2 test and continuous variables by the MannWhitney test. A P value of less than 0.05 was considered statistically significant. Cumulative probability curves were computed according to the Kaplan-Meier method and compared by the log-rank test.32 Univariate analysis was performed for 11 variables at baseline and for 4 evolutionary parameters at first month after DEB TACE procedure to identify predictors of HCC progression beyond Milan criteria. As recommended in the European Association for the Study of the Liver–European Organisation for Research and Treatment of Cancer clinical practice guidelines for the management of HCC, AFP was fixed at a predefined
1262 MANINI ET AL.
cutoff at 200 ng/mL because this value has been shown to have high specificity.4,33 Similarly, a predefined cutoff level of 65 years was also included in the univariate analysis because it was the maximum age considered in our hospital for the admission to the waiting list for LT. All baseline variables with a P 0.10 at univariate analysis were included in the multivariate Cox regression analysis,34 to assess their value as independent predictors. Finally, a Cox regression analysis with time-dependent covariates was performed to adjust the influence of evolutionary parameters on the progression beyond Milan criteria. Statistical analysis was performed with SPSS 19.0 statistical package (IBM, Armonk, NY).
RESULTS Population The initial cohort was composed of 277 patients with newly diagnosed HCC fulfilling Milan criteria. Figure 1 shows the flow chart leading to the selection of the study population. Fifty-five patients with 79 nodules ultimately underwent DEB TACE as first-line treatment modality. The reasons for unfeasibility of locoregional ablation with RFA or PEI are shown in Supporting
Figure 1.
Disposition of the 277 patients with a small HCC.
LIVER TRANSPLANTATION, October 2015
Table 1. Patient demographics, tumor characteristics, and laboratory findings are summarized in Table 1. HCC was histologically diagnosed in 25 (45%) patients; it was monofocal in 33 (60%) patients, with a median diameter of 24 mm (range, 20-50 mm), and multifocal (20 bifocal and 2 trifocal) in 22 (40%) patients, with a median size of 19 mm (range, 7-30 mm). The median interval between HCC diagnosis and the first DEB TACE procedure was 42 days (range, 14-67 days; 95% confidence interval [CI], 37-46 days). Eighteen (33%) patients were listed for LT, whereas 37 (67%) were excluded due to age > 65 years (n 5 32), a previous cancer (n 5 1), active alcohol abuse (n 5 1), methadone treatment (n 5 1), or refusal (n 5 2).
Chemoembolization A total of 129 DEB TACE procedures were performed, with each patient receiving a median of 2 procedures (range, 1-7). Twenty-one patients (38%) underwent a single DEB TACE, whereas the remaining 34 (62%) received at least 2 procedures. The average number of treatment procedures was 2.48 (61.44 SD; range, 1-7) in BCLC A1-3 patients and 2.18 (61.33 SD; range, 1-5) in BCLC A4 HCCs. No statistical differences were observed in the number of DEB TACE
LIVER TRANSPLANTATION, Vol. 21, No. 10, 2015
MANINI ET AL. 1263
TABLE 1. Demographics of BCLC A Patients With a De Novo HCC Characteristic Number of patients Number of nodules Male sex, n (%) Age, years Median (range) >65 years, n (%) Etiology of cirrhosis, n (%) HBV HCV Alcohol Mixed causes Other causes* Child-Pugh-Turcotte class, n (%) A B Decompensated patients, n (%) Ascites (present or previous) Hepatic encephalopathy Platelet count, median (range), 3103/mmc Serum AFP value Median (range), ng/mL 200 ng/mL, n (%) US surveillance, n (%) BCLC class, n (%) A1-3 A4 Nodules Diameter, median (range), mm 30 mm in size, n (%) On waiting list for LT, n (%)
Value 55 79 41 (75) 67 (47-75) 32 (58) 4 (7) 35 (64) 6 (11) 7 (13) 3 (5)
38 (69) 17 (31) 13 (24) 0 (0) 106 (30-257)
21 (2-1194) 17 (31) 41 (75) 33 (60) 22 (40) 22 (7-50) 15 (19) 18 (33)
*Hemochromatosis, nonalcoholic steatohepatitis, hepatitis D virus.
sessions between patients with a single tumor and multiple tumors (P 5 0.24).
Treatment Response One-month treatment response rates are reported in Table 2. After the first DEB TACE procedure tumor progressed, according to mRECIST, in 4 patients, all BCLC A4 at baseline: in 2 patients (both with a baseline AFP 200 ng/mL), the tumor progressed beyond Milan criteria due to NIH whereas 2 showed IHG still within Milan (1 with a baseline AFP 200 ng/mL and the other < 200 ng/mL). A CR was achieved in 29 (53%) patients after the first DEB TACE, in 2 (4%) additional patients after 2 procedures, and in 1 (2%) after 3 procedures. Overall 32 (58%) patients achieved a CR, which was maintained up to 4 and 7 months in 21 (38%) and 17 (31%) patients, respectively. The corresponding HCC recurrence rates were 34% and 47%, respectively. None of the recurrences within the seventh month exceeded Milan criteria. Eight recurrences were IHG of
a necrotic nodule, 3 NIH and 4 both IHG and NIH. In CR patients, none of the baseline variables predicted HCC recurrence within 7 months from DEB TACE. Twelve-, 24- and 36-month tumor-free survival of the whole cohort was 29%, 21%, and 9%.
Follow-up During a median trial time of 28 months (range, 2-63 months), 27 patients progressed beyond Milan criteria: 5 (18%) due to IHG, 18 (67%) due to NIH, and 4 (15%) due to NEH. No patient had tumor progression that was associated with development of satellites from an original HCC nodule. Twenty-three patients were censored, 3 died, whereas none were lost to follow-up. At the end of follow-up, 2 (4%) patients were still within Milan criteria, both with a sustained CR. Supporting Table 2 summarizes the causes for treatment discontinuation. No cycle was interrupted for progression to PS 1. Sixteen patients underwent transplantation after a median time of 9 months (range, 3-21 months) from the first DEB TACE and after receiving a median of 2 (range, 1-3) pretransplant treatments. During a median of 56 months (range, 2-104 months) of posttransplant surveillance, 3 (19%) patients died and none showed HCC recurrence. Of the remaining 2 patients enlisted, 1 with a sustained CR died from acute variceal bleeding, and 1 was delisted for intrahepatic tumor spread. All the 5 radiological CRs observed in patients who subsequently underwent LT were histologically confirmed on the explanted livers. The remaining 11 patients were either radiologically or histologically proven to remain within Milan criteria (6 partial response, 2 stable disease and 3 progressive disease because of NIH). Following the initial treatment with DEB TACE, 7 (13%) patients with a tumor still within Milan criteria received additional treatment with RFA for de novo HCC nodules (n 5 5) and cTACE for the unfeasibility of superselective catheterizations (n 5 2). Progression to BCLC B or C led to secondary-line treatment in 23 (42%) patients, 15 of which underwent cTACE, 2 Yttrium-90 radioembolization, and 6 received sorafenib. None of these patients were then re-treated with DEB TACE.
Safety There was no TACE-related mortality. No major adverse events were recorded (Table 3). Within 24 hours after DEB TACE, a majority of patients showed a mild transient increase in liver enzymes and symptoms of postembolization syndrome, defined as fever (37.58C), abdominal pain, nausea and/or vomiting, that were successfully managed with conservative therapies. Median hospital stay after each course of treatment was 1 day (range, 1-5 days). All patients were asymptomatic upon discharge from the hospital.
1264 MANINI ET AL.
LIVER TRANSPLANTATION, October 2015
TABLE 2. Treatment Response Evaluated by CT Scan 1 Month After the First DEB TACE Patients All Patients
BCLC A1-3
BCLC 4
(n 5 55), n (%)
(n 5 33), n (%)
(n 5 22), n (%)
29 (53) 17 (31) 5 (9) 4 (7)
21 (64) 10 (30) 2 (6) 0 (0)
Outcome Complete response Partial response Stable disease Progressive disease
8 7 3 4
(36) (32) (14) (18)
P Value* 0.09 0.86 0.63 0.04
*Yates’ v2 test.
TABLE 3. Adverse Events in 55 Patients After 129 DEB TACE Procedures
Fever 378C Abdominal pain Nausea and/or vomiting Catheterization-site bleeding† Transient renal insufficiency Groin hematoma‡ Liver decompensation§ Hepatic abscess Intratumor bleeding At least 1 complication
Grade
Grade
1-2*
3-4*
33 (60%) 21 (38%) 17 (31%) 6 (11%) 4 (7%) 2 (4%) 2 (4%) 0 (0) 0 (0) 40 (73%)
0 0 0 0 0 0 0 0 0 0
NOTE: Data are given as n (%). *According to the CTCAE, version 4.0 (v4.03). † Controlled with local compressive medication. ‡ Both self-limiting and resolved within 2 weeks of the procedure. § Mild ascites treated successfully with oral aldosterone antagonists.
TTP Beyond Milan Criteria The overall probability of progression beyond Milan criteria is reported in Fig. 2A. The cumulative 3-, 6-, 12-, and 24-month rates were 2%, 5%, 30%, and 54%, with a median TTP beyond Milan of 19.3 months (range, 1.9-39.3 months; 95% CI, 14.5-39.3). By univariate analysis, 2 baseline (BCLC A4 and AFP 200 ng/mL) and 1 evolutionary variable (CR) emerged as prognostic factors of progression beyond Milan (Table 4). By multivariate analysis including baseline variables only, progression beyond Milan was predicted by multifocal tumor and AFP 200 ng/mL (Table 4), with a median expected TTP beyond Milan of 13.9 months (95% CI, 11.2-19.3) and 12.5 months (95% CI, 9.6-16.7), respectively (Fig. 2B,C). By Cox regression analysis including both baseline and evolutionary variables, CR emerged as the strongest independent prognostic factor associated with progression beyond Milan (P < 0.001) even if BCLC did maintain
its independent prognostic value (P 5 0.04; Table 4). Median expected TTP beyond Milan in CR patients was 39.3 months (95% CI, 24.6-39.3; Fig. 2D), with a median of 39.3 months (95% CI, 38.0-39.3) and 17.1 months (95% CI, 11.2-27.3) in BCLC A1-3 and BCLC A4 patients, respectively (P 5 0.04). The corresponding features at 3, 6, and 12 months are reported in Fig. 3. At the time of analysis, 35 patients (10 still within Milan criteria) had died. The cumulative 1-, 3- and 5-year rates of survival of the whole cohort were 96%, 58%, and 37%, with a median of 50.8 months (95% CI, 30.1-58.5; Supporting Fig. 1A), whereas the same figures after censoring follow-up at the time of transplant were 93%, 45%, and 16%, with a median of 30.1 months (95% CI, 23.1-50.8; Supporting Fig. 1B).
DISCUSSION Tumor progression beyond Milan criteria is the most common cause of delisting for patients with a HCC waiting for a LT, starting from a predicted 12% probability of 6-month dropout for patients in whom the tumor is left untreated during the wait period.35,36 An additional risk to be computed is the rate of tumor recurrence after LT which increased from 12% for patients remaining within Milan criteria, either spontaneously or following bridge therapy, to 45% for those who had a tumor progression beyond Milan.37,38 The present study provides insights on the effectiveness of DEB TACE to maintain patients with an early HCC within Milan criteria, highlighting the potential value of this treatment as bridge therapy to LT. This is not a trivial point given that the application of radical treatment to patients with an early detected HCC is not the rule, as indicated by 2 European multicenter cohort studies reporting 36% to 60% rates of curative therapies being offered to BCLC A patients only.8,9 This reflects the real-life scenario, representing the so-called “treatment stage migration” concept.21 In our cohort, BCLC A patients showed a favorable rate of 58% radiological CR to DEB TACE; that, however, was accompanied by significant rates of tumor recurrence at 6 months (47%), shrinking the overall CR rate to 31%. These figures are consistent with a
LIVER TRANSPLANTATION, Vol. 21, No. 10, 2015
MANINI ET AL. 1265
Figure 2. Cumulative probability of tumor progression beyond Milan criteria: (A) overall; (B) according to baseline BCLC status; (C) according to baseline AFP level; and (D) according to the outcome of DEB TACE. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
previous report of a 6-month recurrence rate of 37% and 40% after 1 and 2 cycles of TACE respectively,19 and favorably compare with 12%-27% rates of CR reported in previous studies which, however, included patients with more advanced HCC.10-12,17 All in all, our results suggest DEB TACE to be a potentially radical treatment, even if in a restricted population, given that all 5 radiological CRs observed in patients who underwent LT were histologically confirmed on the explanted liver and that in 21% of the patients a CR was maintained for up to 2 years following TACE. The finding of a median TTP beyond Milan of 19.3 months, whereas 98% and 95% of patients could maintain the Milan-in status for 3 and 6 months, respectively, does confirm that DEB TACE is effective in attenuating tumor progression beyond Milan criteria. Our results are in line with previous studies showing a pretransplant treatment with cTACE to
result in delayed HCC progression during the waiting list time, lowered tumor recurrence rate after LT, and good survival outcomes.39-41 Our data strengthen the efficacy of DEB TACE to bridge patients to LT and also highlight predictors of tumor progression that can serve the purpose of patient prioritization for LT. Not surprisingly, we found that baseline serum AFP and multiple tumor nodules were independent predictors of progression beyond Milan, a finding largely confirmatory of the biological aggressiveness of this pattern of tumors that also overlaps with previous observations in studies where OS was the primary endpoint.42 However, when the radiological response to the first procedure was included in the Cox regression analysis, a CR emerged as the strongest independent prognostic predictor for the maintenance of the Milan-in status. This finding, in line with 3 previous studies which, however, included patients treated with different locoregional therapies,43-45 could have a
1266 MANINI ET AL.
LIVER TRANSPLANTATION, October 2015
TABLE 4. Pretreatment and Posttreatment Features Associated With the Likelihood of Progression Beyond Milan Criteria Progression Beyond Milan Criteria (27 events) Multivariate Analysis Univariate Analysis
Baseline variables Male versus female sex
HR
95% CI
P Value
0.72
(0.321.64) (0.961.07) (0.402.03) (0.582.73) (0.261.45) (0.281.91) (0.991.00) (1.076.35) (0.502.63) (1.467.49) (0.563.26) (0.221.24)
0.40
Age (per each year of increase) Age > 65 years
1.01 0.90
HCV (yes versus no)
1.26
Child-Pugh B versus A
0.62
Ascites (yes versus no)
0.73
Platelets (per each 103 unit of increase) AFP 200 ng/mL
0.99 2.61
US surveillance (yes versus no) BCLC A4 versus A1-3
1.14
Nodule diameter 30 mm
1.34
On waiting list for LT (yes versus no) Evolutionary variables* CR (no versus yes)
3.30
0.52
4.62
Child-Pugh B/C† versus A
0.84
Platelets (per each 103 unit of increase) AFP 200 ng/mL
1.00 2.32
(1.9810.83) (0.391.84) (0.991.00) (0.4212.87)
Multivariate Analysis at First Month After DEB
at Baseline
TACE
HR
95% CI
P Value
HR
95% CI
P Value
2.38
(1.085.22)
0.03
1.92
(0.844.41)
0.13
3.21
(1.447.14)
0.004
2.64
(1.116.27)
0.04
4.70
(1.7412.66)
