A Special Article

Hepatocellular Carcinoma Secondary to Chronic Hepatitis C Virus Infection in Veterans at the VA Caribbean Healthcare System – Have Surveillance Measures Been Effective?

Read Article

The incidence of hepatocellular carcinoma (HCC) due to hepatitis C virus (HCV) infection has been rising worldwide as well as in the United States. Current American Association for the Study of Liver Disease (AASLD) guidelines recommend performing an abdominal ultrasound in cirrhotic patients every six months for early detection of HCC. The main objective of this study was to retrospectively evaluate a population diagnosed with HCC secondary to HCV in the Veterans Administration (VA) Caribbean Healthcare system and to determine if screening strategies were applied appropriately. Secondary aims were to describe certain patient characteristics upon diagnosis of HCC and determine the median survival time of this population. It was found that in 95.4% of the cases, the diagnosis of HCC was incidental, and not part of a surveillance strategy. More so, the median survival after diagnosis was only 10 months. These findings should help raise awareness of the importance of HCC surveillance in cirrhotic patients with or without HCV.

The incidence of hepatocellular carcinoma (HCC) due to hepatitis C virus (HCV) infection has been rising worldwide as well as in the United States. Current American Association for the Study of Liver Disease (AASLD) guidelines recommend performing an abdominal ultrasound in cirrhotic patients every six months for early detection of HCC. The main objective of this study was to retrospectively evaluate a population diagnosed with HCC secondary to HCV in the Veterans Administration (VA) Caribbean Healthcare system and to determine if screening strategies were applied appropriately. Secondary aims were to describe certain patient characteristics upon diagnosis of HCC and determine the median survival time of this population. It was found that in 95.4% of the cases, the diagnosis of HCC was incidental, and not part of a surveillance strategy. More so, the median survival after diagnosis was only 10 months. These findings should help raise awareness of the importance of HCC surveillance in cirrhotic patients with or without HCV.

Sheryl Rosa, Walisbeth Class, Henry DeJesus, Doris H. Toro, VA Caribbean Healthcare System, San Juan, PR

INTRODUCTION

Hepatocellular carcinoma (HCC) is one of the most feared outcomes of chronic hepatitis C virus (HCV) infection.1 The incidence of HCC varies widely within different regions of the world and concordantly differs among different racial and ethnic groups within the same country.2 This phenomenon is attributed to regional variations in exposure to the different hepatitis viruses, environmental pathogens and inheritance patterns of genetically linked liver diseases.2 Although the mechanism of carcinogenesis of the Hepatitis C virus has not been elucidated, studies which have used mouse models suggest that the development of HCC in HCV arises from a rapid cellular turnover and chronic inflammation and not from oncogene activation as is seen in hepatitis B related HCC.3

Epidemiology

There is an estimated global distribution of 185 million people who currently live with chronic HCV infection.4 Even more so, approximately 399,000 people die each year because of HCV complications, among them being HCC.5,6 Liver cancer is the fifth most frequently diagnosed cancer in men, while in women it is the ninth most frequently diagnosed cancer worldwide, and it is the fourth leading cause of cancer-related death in the world.2,7 In North America, the incidence rates in 2008 for males and females were 6.8 and 2.2 per 100,000 persons, respectively. The differences in gender expression are not clearly understood, but are suspected to be secondary to hepatitis carrier states, exposure to environmental toxins, and the effects of androgens.8

Although North and South America are considered low incidence areas for cases of HCC, the incidence in the United States has increased during the past two decades, possibly due to a large pool of people with longstanding chronic hepatitis C.9 The rate began to accelerate in the mid-1980s, most likely because of the increased incidence of cirrhosis due to chronic HCV infection and nonalcoholic fatty liver disease, combined with a large influx of immigrants from East Asia and other geographic areas with high endemic rates of hepatitis B viral infection.9 The annual incidence of HCC in the US was at least 6 per 100,000 in 2010. Recently, the incidence of HCC among the United States veteran population had been notably rising as well, likely secondary to an increase of HCV infection among this specific population.10

Risk Factors

In addition to chronic HCV there have been many risk factors associated to developing HCC which include Hepatitis B carrier state, hereditary hemochromatosis, comorbid hepatic disease, environmental toxins and cirrhosis of any cause. The most commonly seen risk factors in the United States are HCV infection, alcohol use and nonalcoholic fatty liver disease; risk factors commonly identified in the veteran population. Risks factors for HCC development among patients with HCV-related cirrhosis can be considered as host related, virus related and of external origin.9 Independent risk factors associated with progression to HCC are older age (>55 years: 2- to 4-fold increased risk) and male sex (2-to 3- fold increased risk). Several comorbid conditions are thought to increase the risk of HCC among patients with HCV-related cirrhosis, including porphyria cutanea tarda (PCT), hepatic iron overload, liver steatosis and diabetes mellitus.9

Upon review of the 2010 HCV Veterans registry in the region of Veterans Integrated Service Network (VISN 8), to which Puerto Rico belongs, based on serologic evidence of HCV infection status (HCV positive) as well as the subset of those with VA laboratory evidence of HCV viremia, there were 21,997 patients registered with HCV from which 19,649 were HCV positive and 15,587 had HCV viremia.11 Per the report from 2011, a total of 16,026 HCV viremic Veterans were registered with VHA care in VISN 8. During the same year in San Juan VA, a total of 1,567 patients were registered as being in care for HCV and 25 of them were first diagnosed with HCC during that year. The prevalence of HCV in veterans is about 3.7 times higher than in the general population, which could explain the many new cases of HCC arising in this population.

As a deadly entity, it is of benefit that HCC is detected early in its course, for management or even if possible, curative treatment. HCC, if untreated, has a mean survival of 1 to 3 months and a 5-year survival rate as low as 3%. For this important reason, surveillance measures guidelines for high-risk populations have been published by the American Association for the Study of Liver Disease (AASLD).1 Surveillance is deemed cost-effective if the expected HCC risk exceeds 1.5% per year in patients with cirrhosis. Current AASLD guidelines recommend physicians to perform an abdominal ultrasound in a screening interval of 6 months in patients with confirmed cirrhosis. It is expected that by following this strict routine sonographic surveillance of cirrhotic patients it would be possible to identify any liver lesion with malignant potential in a timely manner, and therefore treat patients at an earlier stage of disease and possibly extend these patient’s survival expectancy and improve their quality of life.1 It is imperative for primary care physicians and for gastroenterologists to keep this recommendation in mind as part of routine screening in patients with cirrhosis secondary to HCV or any other etiology. However, it is suspected that in real life practice, these recommendations are not being followed as rigorously as expected, and in turn hepatic lesions are either found at a later stage of disease or incidentally when investigating a different disease ailment.

Objective

This study consisted of a retrospective analysis using data gathered from the medical records of Latino veterans from the VA Caribbean Healthcare System who had documented diagnoses of HCV and also HCC. The principal objective of this study was to evaluate if adequate surveillance of hepatoma by imaging was performed in these patients as recommended by AASLD guidelines. The secondary objective was to identify and describe the pertinent sociodemographic and clinical characteristics of these patients. Variables accounted for were gender, comorbid diabetes mellitus, platelet count, body mass index (BMI), coexisting risk factors for HCC development, if surveillance was done appropriately, Child Pugh score upon HCC diagnosis, model for end stage liver disease (MELD) score upon HCC diagnosis, diagnostic study used, stage of diagnosis using the BCLC staging system, treatment modalities applied, and median survival time.

Materials and Methods

This study was conducted by performing a detailed review of the electronic medical charts (CPRS system) of Latino veterans from the VA Caribbean Healthcare System with confirmed chronic HCV infection and HCC diagnosed between January 1, 2001 and May 21, 2013. These cases were identified by patient encounters coded as per ICD-9 diagnosis codes 155.0, 155.2, 230.8, V10.07, 070.44, and 070.54. Inclusion criteria were ages between 22-88, confirmed HCV with HCC, and Latino origin. Exclusion criteria were absence of HCV, ages less than 22 or over 88, and any other ethnic origin which is not Latino. Statistical analysis was performed with the Statistical Package for Social Science, IBM SPSS Statistics for Windows, Version 21.0. Armonk, NY: IBM Corp.

Results

After a systematic review of all the medical records of patients with HCV and concomitant HCC, we identified a cohort of 131 patients who met the inclusion criteria. Details are in Table 1.

All the patients within our cohort were of male gender. The mean age at diagnosis was 62.2 years. The majority of the patients were of normal body mass index (BMI) (43.5%), but followed closely by patients with BMI compatible with pre-obesity (33.6%). When taking into account diabetes mellitus as a comorbid condition, there was not a substantial difference among those who were diabetic and the non-diabetics which were 45% and 55% respectively. In addition, a significant difference was neither found when considering platelet count with 56.5% of patients with platelet count 100,000 µL or above whereas 43.5% of patients were found with less than 100,000 µL upon diagnosis.

Liver biopsy was performed in 42% of our cohort and within that percentage, 22.9% were patients with histopathology findings reporting cirrhosis which is consistent with a Metavir score of F4. However, in 19.1% of the cases where biopsy was performed, the Metavir score was not described in the pathology report and the grade of fibrosis of the tissue was not specified. Since studies published in 2001, liver biopsy is not required for diagnosis of HCC after the establishment of specific diagnostic criteria by imaging studies which in turn avoids the risks associated to percutaneous biopsies which include bleeding and tumor spread along the needle track.12 Of the total cohort, 39.7% had a diagnostic imaging study (either quadruple phase computed tomography (CT) or magnetic resonance imaging (MRI)) followed by a confirmatory biopsy whereas 3.2% had a biopsy performed as the sole diagnostic study. Of those diagnosed by imaging studies, 37.4% was by quadruple phase CT and 24.4% was by dynamic MRI, however some patients had both studies done upon diagnosis and therefore are accounted for in each category.

Alfa fetoprotein (AFP) is a glycoprotein considered to be the most commonly associated serum marker with the presence of HCC. Most studies agree that levels above 500 mcg/L are highly suspicious for the presence of HCC although it has also been established that not all tumors secrete AFP and up to 40% of patients with HCC may have normal AFP levels.13 This previously described data is validated in our study where only 13.7% of the cohort was found with AFP levels above 500 mcg/L upon diagnosis but most of the cohort (77.8%) had AFP levels less than 250 mcg/L when diagnosed with HCC.

Two predictive models of the prognosis of patients with cirrhosis are the Child Pugh classification and the Model for End Stage Liver Disease (MELD) score.14 In general terms, A Child Pugh score of A means that a patient has compensated cirrhosis, a score of B is compatible with significant functional compromise, and C signifies decompensated cirrhosis. The MELD score is used to calculate an estimated 90-day mortality in patients with cirrhosis and is a crucial factor in prioritizing patients for liver transplantation. Due to the latter, a MELD exception is given to patients with HCC who are candidates for liver transplant to prioritize those patients due to the high mortality risk of HCC.15 The MELD score accounted for in this study was without the inflation of the MELD exception points to better characterize these patients upon diagnosis. The data obtained revealed that in terms of both Child Pugh score and MELD score, the majority of that patients were diagnosed within an early stage with 59.5% of patients diagnosed while still being classified within Child Pugh A and notably as well 59.5% of patients with a MELD score less than 10.

The Barcelona-Clinic Liver Cancer (BCLC) staging system is an HCC treatment algorithm which consists of multiple variables related to HCC that include tumor stage, physical status, liver functional status, and cancer-related symptoms.16 Essentially, patients at stage 0 are candidates for resection, stage A are candidates for curative therapies, stage B patients should be referred for chemoembolization, stage C can be considered for experimental therapies, and patients with stage D should undergo palliative treatment. Within our cohort, six cases (4.6%) were diagnosed at BCLC stage 0, 59 cases (45%) at BCLC stage A, 40 cases (30.5%) at BCLC stage B, 15 cases (11.5%) at BCLC stage C, and 11 cases (8.4%) at BCLC stage D. Correspondingly, the therapeutic modalities used in these patients were: surgery in three (2.3%), TACE in 72 (55%), RFA in three (2.3%), TACE + RFA in 10 (7.6%), liver transplant in two (1.5%), Sorafenib in 49 (37.4%), palliative care in 53 (40.5%), and experimental or no treatment in 12 (9.2%). It is important to point out that 52 patients received two or more treatment modalities.

Overall, the median survival was 10 months after diagnosis. Of the 131 patients diagnosed with HCC, only six (4.6%) were identified within a routine surveillance program whereas 125 (95.4%) of the cases were diagnosed incidentally. AASLD guidelines recommended screening of cirrhotic patients every 6 to 12 months until 2010 and afterwards guidelines were revised and recommended screening for HCC every 6 months. These changes were considered upon data collection.

Discussion

The data obtained from this retrospective analysis of the medical records of Latino veterans who had HCC within a background of HCV suggests that are no clear predictors indicative of which patients will develop this disease with such a low 5-year survival rate. In contrast to other studies, it was seen that in our population there was not a strong association with risk factors such as diabetes mellitus, elevated BMI, or low platelet count. The majority of patients who were diagnosed with HCC were classified within early stages of the most common prognostic models which are the Child Pugh score and MELD score. Also, this study validates previously described data noting that AFP is not a universal marker for the diagnosis of HCC with only 13.7% of our cohort with levels above 500 mcg/L upon diagnosis. More importantly, although 80.1% of the cohort was diagnosed within BCLC stages B or better, the median survival was only 10 months after diagnosis. However, the most striking point of all the data is that only 4.6% of the patients with HCC were diagnosed within a routine surveillance program.

The main limiting factors in our study were that all patients were male veterans; therefore, these results may not represent the general population of non-veteran men and any women. Another limiting factor was that RFA was not available in our institution until 2007 and therefore unknown if the lack of this treatment modality within this time period would have had an effect in the median survival time after diagnosis if patients would have had that option. It is also fair to mention that when this study was conducted, the new direct acting antiviral treatments for HCV were not available and therefore new studies will need to be performed on that specific population of patients that are treated for HCV.

In conclusion, surveillance was not conducted effectively in our study population. Patients diagnosed with cirrhosis need to be entered in a strict surveillance program due to the lack of clinical or laboratory indicators of HCC development and the poor prognosis once diagnosed. Within our study, 95.4% of the HCC cases were diagnosed incidentally and therefore it is unknown if an earlier diagnosis would have led to a better survival rate for which new studies are recommended within a cohort that has been surveilled as recommended by guidelines.These findings should help raise awareness among all physicians of the importance of HCC surveillance in all cirrhotic patients to grant patients an early diagnosis and hopefully improve their chance of survival and quality of life. With the mandatory widespread use of electronical medical records and the aid of modern technology, it may be of great help to use regulatory medical reminders in the institution of imaging surveillance in cirrhotic patients within all the healthcare facilities in our nation.

The contents of this publication do not represent the views of the VA Caribbean Healthcare System, the Department of Veterans Affairs or the United States Government.

Download Tables, Images & References