Screening for Hepatoma and an Introduction to LIRADS

Transcription

1 Screening for Hepatoma and an Introduction to LIRADS Helena Gabriel, MD Associate Professor of Radiology Director, School of Ultrasound rthwestern University Feinberg School of Medicine Chicago, IL Overview - the problem Principles of screening-is screening efficacious? Evidence for screening Findings of on US, CT/ MR, US contrast based on molecular and vascular tumor features LIRADS: explanation and image examples How US fits into LIRADS and screening Hepatocellular Carcinoma, A World-Wide Problem GLOBOCAN, World Heath Organization Global Hepatoma Incidence Map 212 Incidence varies strikingly by geography Global variations in incidence reflect variation in risk factors for, namely HBV and HCV GLOBOCAN, WHO. Global Hepatoma Incidence Map: Hepatitis B Prevalence Overlay GLOBOCAN, WHO. Global Hepatoma Incidence Map: Hepatitis C Prevalence Overlay 1

2 Age-standardized incidence and mortality rates of liver cancer (per 1, persons) in different world regions Hepatoma incidence and mortality Liver cancer incidence (Per 1) China South EastAsia European Union Africa US Cases Deaths Fifth most common cancer in men, seventh in women. Mortality: Second highest in males, sixth highest in females, overall third most common cause of death worldwide 211: incidence- 782,451, mortality- 745,533 (WHO, 212) Incidence/mortality ratio of 1.7 suggests it is a lethal disease, most people die of disease if no intervention : a global view. Nat Rev Gastroenterol, Hepatol. 7, El-Serag HB. Epidemiology of viral hepatitis and hepatocellular carcinoma. Grastorenterology. 212; 142(6): e1. Altekruse SF, McGlynn KA, Reichman ME> Hepatocellular carcinoma incidence, mortality, and survival trends in the US. J Clin Oncol. 29:27(9): Khanasekaran R, Limave A, Cabrera R. Hepat Med. Evidence and research. 212; 4: Age-adjusted incidence rates of liver cancer (per 1,) in the United States mortality: Barcelona clinic liver cancer (BCLC) staging vs. survival Liver cancer incidence in the USA (Per 1) All races Whites Blacks BCLC is most commonly used staging system for. BCLC takes into account tumor stage, liver function, physical status and symptoms and links the stages with treatment algorithm. Natural history of disease/ prognosis is dismal if there is no intervention/ treatment. End Result database from years Llovet JM, Fuster J, Bruix J. The Barcelona approach: diagnosis, staging and treatment of hepatocellular carcinoma. Liver transplantation, vol 1, no 2, supple 1, 24:S Hepat Med. 212; 4: Survival, according to tumour stage, in untreated Hepatoma Survival probability (Percent) Intermediate stage Advanced stage There is a world-wide epidemic which has geographical variation based on risk factors Incidence continues to increase Mortality if untreated is dismal Hepatoma is a major world-wide health concern Time (Months) Llovet JM, et al. Hepatocellular carcinoma. Lancet 23; 362:

3 Should there be world-wide screening, surveillance for hepatoma? AASLD (American Association for Study of Liver Disease) EASLD (European Association for Study of Liver Disease) AASLD guideline provides an excellent data, evidence-based approach to diagnosis, staging and treatment of patient with. Provides data-supported recommendations AASLD, EASLD very similar except for a few small points Principles of effective population screening, World Health Organization The disease should be common with significant mortality and morbidity The disease should have a latent or early symptomatic stage There should be a target population for screening The test should have high sensitivity and specificity and be acceptable to patient There should standardized recall procedures There should be facilities available for recall procedures and tx The disease must have effective therapy that can decrease mortality if the disease is detected early enough The test should decrease disease-specific mortality Cost of screening should be economically balanced Wilson JMB, Jungner G. Principles and practices of screening for disease. Geneva, Switzerland: World Health Organization; Report.: Public Health Papers. 34. Screening for Hepatoma-Identifying an at risk population Risk factors for developing hepatocellular carcinoma At risk population is dependent on incidence From prior graphs, distribution of worldwide mirrors Hepatitis B, C infection Prospective trials have shown that if the incidence of is 1.5%/year or greater, surveillance helpful, Sarasin et al. and Lin et al. NAFLD Other causes of cirrhosis Hepatitis B Chronic hepatitis / cirrhosis Hepatitis C Alcoholic liver disease Aflatoxin Sarasin FP, Giostra E, Hadengue A. Cost-effectiveness of screening for detection of small in western patients with Child-Pugh class A cirrhosis. AM J Med 1996; 11: Lin OS, Keeffe EB, Sanders GD, et al. Cost-effectiveneess of screening for inpatients with cirrhosis du to chronic hepatitis.c. Aliment Pharmacol Ther 24; 19: Hepat Med. 212; 4: Screening for Hepatoma: Identification of an at-risk patient population Risk factors for hepatoma, 75% due to hepatitis B, C infection: Risk with Hepatitis B Most common cause of in low resource countries. Annual incidence of in patients with hepatitis B.5-1.% in Asia Relative risk of in patients with hepatitis B-1 x Range in US varies Risk increases with onset of cirrhosis Risk with Hepatitis C Most common cause of in Western countries Relative risk-2 x Highest risk with cirrhosis-inc of 2-8%/year Screening for Hepatoma-Decreased Mortality? Disease that is potentially treatable/ curable Screening has shown decrease in mortality SEER data has shown other endpoints such as downgrading in stage or 5 year mortality rates not valid endpoints due to lead and length time bias One RCT (randomized controlled trial) in Shanghai, Zhang et al, evaluated the use of screening/surveillance US and AFP. 18,816 patients who underwent 6 mo AFP and US surveillance. 37% reduction in related mortality even with poor compliance (6% adherence to screening protocol) 3

4 Study profile Cumulative mortality from in screening and control groups Refuse to participate (384) 192 subjects aged years with HBVM(+) or chronic hepatitis Randomization Screening group (9757) Agree to participate (9373) First round screening December, patients with 69 patients with Control group (9443) t told, no screening December, patients with Survival Stage I screening Stage II screening Stage II control Stage III screening Stage III control December 31, died from December 31, died from Months Zhang BH, et al. J Cancer Res Clin Oncol 24 Zhang BH, et al. J Cancer Res Clin Oncol 24 Screening for : decrease in mortality Second China study did not show a decrease in mortality but smaller and not as well regarded Other prospective trials needed but will likely not be performed since most feel surveillance is helpful and not ethical to not provide surveillance Screening in -decrease in mortality Other cohort studies show improved survival with screening. McMahon B. Hepatology.2 Wong, et al. Transplant, 2. Surrogate endpoint-detection of at an earlier stage when curative treatment possible. Gaba RC et al. Ann Hepatol, 213 El-Serag HB, et al. Gut, 211 Wong GL, et al. Liver Int, 28 How Well does the Screening Modality Perform? Adequate Sensitivity, Specificity, Accuracy Does the test have adequate sensitivity, specificity, accuracy (positive and negative predictive values) and performance characteristics (ROC curves) Serological AFP DCP Lectin-bound AFP TIE2 Radiological screening US CT MR US contrast Surveillance Tests-AFP AFP is a glycoprotein secreted by fetal hepatocytes and (poorly differentiated) Pitfalls of AFP: AFP can be high in chronic liver disease and not all s secrete AFP ROC analysis of AFP suggest 2 ng/ml is the best cut off At this level, sensitivity of AFP for, 6% HALT-C study- studied efficacy of interferon and looked at AFP levels. AFP levels inadequate for surveillance detection of AFP inadequate screening test for Better as a diagnostic rather than screening test? Other serologic tests-dcp and others- ineffective 4

5 Surveillance test: AFP Original AASLD guidelines suggested US with AFP as per the China study New updated AASLD guideline, APF determination lacks adequate sensitivity and specificity for effective surveillance (and for diagnosis). Thus, surveillance has to be based on ultrasound examination. Bruix J, Sherman M. AASLD practice guideline. Management of : An update. Surveillance Test: US Sensitivity for ranges from 3-1%, Specificity 73-1% Singal et al. Aliment Pharmacol Ther 29 Meta-analysis of surveillance US for in patients with cirrhosis 94% any stage 63 % early stage Addition of AFP not helpful in detection, 63%--69%, not statistically significant Screening at 6 month rather than 1 year intervals improved sensitivity (5.1% vs. 7.1%) Screening interval should be based on tumor growth rate and doubling times- median doubling time is 17 days. AASLD recommends a screening interval of 6 months Performance characteristics of ultrasound for the detection of hepatocellular carcinoma Sensitivity of ultrasound with or without alphafetoprotein for the detection of early-stage First author and study. pts. Sensitivity (95% CI) Specificity (95% CI) Odds ratio (95% CI) Kobayashi (.35.97).98 (.92 1.) 128 (15 1) Pateron (.49.95).96 (.9.99) 91 (18 463) Bolondi (.84.98).95 (.91.97) 262 (82 833) Caturelli (.99 1.).99 (.98.99) Sangiovanni (.94 1.).85 (.81.89) 317 (76 1) Sangiovanni (.92 1.).9 (.84.95) 632 (81 1) Pooled estimates (.89.98).91 (.76.97) ( ) Pooled estimates * (.83.98).94 (.89.97) ( ) Pts, patients;, hepatocellular carcinoma; AFP, alpha-fetoprotein. * Pooled estimates after exclusion of outlier studies (Caturelli et al.27). Confidence interval could not be determined given high upper limit. Meta-analysis: surveillance with ultrasound for early-stage hepatocellular carcinoma in patients with cirrhosis First author and study. pts.. early US sensitivity (95% CI) US with AFP sensitivity (95% CI) Kobayashi (.16.84).5 (.16.84) Arrigoni (.41.89).75 (.48.93) Oka (.51.81).8 (.64.91) Pateron (.5.54).38 (.14.68) Cottone (.69.96).87 (.69.96) Zoli (.76.98).91 (.76.98) Tradati (.4.78).33 (.4.78) Henrion ( (.54 1.) Bolondi (.7.91).82 (.7.91) Caturelli (.96.99).99 (.96.99) Santagostino (.3.65).25 (.3.65) Sangiovanni (.41.6).5 (.41.6) Sangiovanni (.38.62).5 (.38.62) Pooled estimates (.5.83).74 (.56.87) Pooled estimates * (.49.76).69 (.53.81) Pts, patients;, hepatocellular carcinoma; AFP, alpha-fetoprotein. * Pooled estimates after exclusion of outlier studies (Caturelli et al.27). Meta-analysis: surveillance with ultrasound for early-stage hepatocellular carcinoma in patients with cirrhosis Screening with CT/MR Is screening modality cost effective? Higher sensitivity for absolute lesion detection than US Very few studies evaluating in the context of screening Pocha et al. Found US was marginally more sensitive than CT, (71.4% vs. 66.7%) and more cost effective. Two studies found CT cost effective for screening, Lin et al. 24, and Arguedas MR, et al. 23. real screening data with MRI Yet, many institutions in USA, screen with CT, MR Three guidelines permit CT, MR for surveillance of patients in whom US is limited by obesity, other factors or those of very high risk of. Kudo M. Japan Society of Hepatology 29 Kobe Congress EASLD Asian Pacific Association of the Study of the Liver consensus recommendation, 21. Level considered cost effective is interventions achieved at a cost of < $5,/year of life gained Cost effectiveness dependent on incidence Screening for is cost effective if risk of is 1.5%/year or greater, AASLD. Lin, et al found US screening for cost effective despite incidence Lin OS, et al. Cost-effectives of screening for in patients with cirrhosis due to chronic hepatitis C. Aliment Pharmacol Ther 24;19:

6 Is there effective treatment for? BCLC (Barcelona Clinic Liver Cancer): Staging & treatment allocation algorithm Screening for Hepatoma Conclusions: Stage PST, Child-Pugh A Very early stage () Single < 2 cm, carcinoma in-situ Single Portal pressure / bilirubin Increased rmal Early stage (A) Single or 3 nodules 3 cm, PS 3 nodules 3 cm Associated diseases Yes Stage A C PST 2, Child-Pugh A - B Intermediate stage (B) Multinodular, PS Advanced stage (C) Portal invasion, N1, M1, PS 1-2 Stage D PST > 2, Child-Pugh C Terminal stage (D) Can be performed in at risk patients with cirrhosis Has been shown to decrease mortality US has been shown to be effective, CT/MR not validated in RCTs Can be cost-effective with US Natural course of disease can be altered by effective tx Is routinely performed Recommended by the AASLD, EASLD and Barcelona guidelines Resection Liver transplantation RF / PEI TACE Sorafenib Best supportive care RF: Radiofrequency ablation PEI: Percutaneous Ethanol Injection TACE: Transarterial chemoembolization PST: Per single tumor PS: Physical status Adapted from EASL EORTC Clinical Practice Guidelines. Journal of Hepatology 212;56: AASLD guidelines Imaging Findings of : Patients at high risk for developing should be entered into surveillance programs Patients on a transplant waiting list should be screened for Surveillance for should be performed with ultrasonography AFP is an inadequate screening test for and is not recommended Patients should be screened at 6 month intervals Surveillance interval does not need to be shortened for patients at higher risk of Imaging Findings of Hepatoma: Ultrasound Single vs multiple vs infiltrating Echogenicity varies. Ignee, et al. Z Gastroenterol, 25, 1 histo-proven cases: Hypoechoic (48%), isoechoic (9%), hyperechoic (19%), mixture (25%) Well-defined or poorly defined Hypoechoic halo may represent a fibrous capsule Additional finding: PV invasion, stellate scar, calcification Vascularity: Hypervascular with penetrating branching appearance. Contrast increases vascularity Limitations in cirrhotic liver Findings of Hepatoma on CT, MR, contrast US Findings seen with contrast-enhanced imaging reflect some of the changes seen with hepatocarcinogenesis of hepatic nodules. Hepatocarcinogenesis is the transformation of normal hepatocytes to malignant nodules, that involves many factors: Molecular level-genetic alterations Structural level-genetic mutations, transpositions, deletions Pathologically, dedifferentiation of hepatocytes to abnormal nodules Angiogenesis-unpaired arteries and sinusoidal capillarization, decrease in portal tracts Venous drainage-changes from HVs to PVs and sinusoids Tumor capsule and fibrous septa Fat deposition Decrease in OATP expression Iron resistance 6

7 Hemodynamic and OATP expression changes during multistep hepatocarcinogenesis Imaging correlates to hepatocarcinogenesis: cirrhotic nodule OATP expression Adapted from Choi et al, CT and MR Imaging Diagnosis and Staging of Hepatocellular Carcinoma: Part I. Radiology 214 Most cirrhotic nodules are imperceptible and blend with parenchymal background T1W isointense, rarely hyperintense (mechanism unknown) T2W iso to hypointense Postcontrast-iso to hypointense (relative to adjacent delayed enhancing fibrosis)-rmal portal triads and OATP expression is preserved-similar signal intensity to liver in hepatobiliary phase. Slightly atypical features: T1 hyperintensity, delayed relative hypoenhcancement, hepatobiliary phase hyperintesity Imaging correlates to hepatocarcinogenesis: Dysplastic nodule Hyperintense on T1W- Iron or Copper Hypo or Iso on T2W (never hyperintense)-may contain iron Post contrast-iso (most have portal triads), some have neovascularity and are hyperintense on contrast. washout, no capsular appearance-drainage patterns have not changed. Imaging correlates to hepatocarcinogenesis: Early Fat deposition-fewer arteries and veins may cause ischemia in the nodule and fatty reactive change n-iron containing May have arterial hypervascularity, may have washout Imaging correlates to hepatocarcinogenesis: AASLD: Diagnostic algorithm for suspected Progressed Arterial hypervascularity- unpaired arteries and sinusoidal capillarization Portal venous and delayed phase washout-drainage patterns changed to portal veins and sinusoids T2W hyperintensity, may be iso or hypo as well Tumor capsule and fibrous septae Corona radiata dules <1 cm should be followed with ultrasound every 3-6 months. If no growth for 2 years, revert to routine surveillance < 1 cm Repeat US at 3 months Growing / changing Stable character Investigate according to size Liver nodule detected For nodules >1cm, typical imaging appearance is diagnostic. Otherwise, a second study with the other imaging modality should be performed, or the lesion biopsied > 1 cm 4 phase CT or dynamic contrast enhanced MRI Arterial hypervascularity AND venous or delayed phase washout Other contrast enhanced study (CT or Yes MRI) Arterial hypervascularity AND venous or delayed phase washout Biopsy Yes Adapted from AASLD Clinical Practice Guidelines. Hepatology 211;53:

8 Hepatoma distant metastases Lung, intra-abdominal lymph nodes, bone, adrenal glands Porta hepatis nodes often seen and benign Pelvic bones often the site of bone metastases LI-RADS (v. 213) Observation Benign entity Definite Probable Neither definite nor probable n- malignancy? Mass? Yes Yes OM LR1 LR2 Tumor in vein? Yes Diameter (mm): Arterial phase hypo- or iso-enhancement Arterial phase hyper-enhancement < 2 2 < Washout Capsule Threshold growth ne: One: Two: LR4A LR4B LR4B LR4A LR4A LR4A LR5A LR4B LR5B LR5B LR5V LR5 Treated Apply ancillary features and then tie-breaking rules to adjust category Reproduced with permission from ACR LI-RADS (v. 214) Conclusions Observation in high-risk patient Treated observation Untreated observation Definitely benign Probably benign Neither definite nor probable benign n malignancy LR - M LR - Treated LR - 1 LR - 2 Tumor in vein LR - 5V Arterial phase hypo- or iso-enhancement Arterial phase hyper-enhancement Diameter (mm): < 2 2 < ne: LR 4 Washout Capsule One: / LR - 5 LR - 5 Threshold growth Two: LR - 5 LR - 5 Apply ancillary features and then tie-breaking rules to adjust category Observations in this cell are categorized LR 4 except as follows: / LR - 5 LR 5G if there is 5% diameter increase in 6 months equivalent to OPTN 5A - G LR 5US if there is both washout and visibility as discrete nodules at antecedent surveillance ultrasound, per AASLD criteria Hepatoma is an epidemic with geographic variations that reflect variations in risk factors, namely Hepatitis B, C Screening for hepatoma decreases mortality and should be performed according to the tenets of screening Ultrasound is the only screening/surveillance modality which has been proven to decrease mortality in a large RCT. Reproduced with permission from ACR Conclusions Findings of hepatoma on CT, MR, US contrast can reflect the process of hepatocarcinogenesis and the molecular, structural, and vascular changes that occur with the development of. LIRADS classification system, although complicated, is a precise means of categorizing and communicating lesions US has a very small role in LIRADS but this may increase with US contrast Conclusions Detection, diagnosis and treatment of hepatoma involves many interdisciplinary groups world-wide. Alphabet soup includes: AASLD, EASLD Barcelona staging system Childs-Pugh, MELD criteria Milan criteria, Unos LIRADS 8

9 9