Alimentary Pharmacology & Therapeutics Early View, Version of Record online: 16 OCT 2017
Short-term risk of hepatocellular carcinoma after hepatitis C virus eradication following direct-acting anti-viral treatment
E. Ogawa, N. Furusyo, H. Nomura, K. Dohmen, N. Higashi, K. Takahashi, A. Kawano, K. Azuma, T. Satoh, M. Nakamuta, T. Koyanagi, M. Kato, S. Shimoda, E. Kajiwara, J. Hayashi, The Kyushu University Liver Disease Study (KULDS) Group
First published: 16 October 2017
Summary
Background
With the development of direct-acting anti-virals (DAAs), almost all patients with chronic hepatitis C virus (HCV) infection can achieve sustained viral response (SVR).
Short-term risk of hepatocellular carcinoma after hepatitis C virus eradication following direct-acting anti-viral treatment
E. Ogawa, N. Furusyo, H. Nomura, K. Dohmen, N. Higashi, K. Takahashi, A. Kawano, K. Azuma, T. Satoh, M. Nakamuta, T. Koyanagi, M. Kato, S. Shimoda, E. Kajiwara, J. Hayashi, The Kyushu University Liver Disease Study (KULDS) Group
First published: 16 October 2017
Summary
Background
With the development of direct-acting anti-virals (DAAs), almost all patients with chronic hepatitis C virus (HCV) infection can achieve sustained viral response (SVR).
Aim
To evaluate the short-term risk of HCC among patients with SVR by DAAs, including those with cirrhosis or previous HCC.
To evaluate the short-term risk of HCC among patients with SVR by DAAs, including those with cirrhosis or previous HCC.
Methods
This large-scale, multicentre cohort study included 1,675 consecutive patients who achieved SVR by treatment with interferon-free sofosbuvir-based regimens, divided into groups with (n = 152) or without previous HCC (n = 1,523). The Kaplan-Meier method and Cox proportional hazard analysis were used to calculate the cumulative HCC incidence and related factors of HCC.
This large-scale, multicentre cohort study included 1,675 consecutive patients who achieved SVR by treatment with interferon-free sofosbuvir-based regimens, divided into groups with (n = 152) or without previous HCC (n = 1,523). The Kaplan-Meier method and Cox proportional hazard analysis were used to calculate the cumulative HCC incidence and related factors of HCC.
Results
During the follow-up period (median: 17 months), 46 (2.7%) patients developed HCC. The 1-year cumulative rates of de novo HCC were 0.4% and 4.9% for the noncirrhosis and cirrhosis groups respectively (log-rank test: P < 0.001). For cirrhotic patients, serum α-fetoprotein level at the end of treatment (EOT-AFP) was the strongest predictor of de novo HCC. The 1-year cumulative de novo HCC rates were 1.4% and 13.1% in the EOT-AFP < 9.0 ng/mL and ≥ 9.0 ng/mL groups (cut-off value) respectively (log-rank test: P < 0.001). The 1-year cumulative rates of HCC recurrence were 6.5% and 23.1% for the noncirrhosis and cirrhosis groups respectively (log-rank test: P = 0.023). For cirrhotic patients, previous HCC characteristics were significantly associated with HCC recurrence. In contrast, sex, age and metabolic features did not influence de novo HCC or recurrence.
During the follow-up period (median: 17 months), 46 (2.7%) patients developed HCC. The 1-year cumulative rates of de novo HCC were 0.4% and 4.9% for the noncirrhosis and cirrhosis groups respectively (log-rank test: P < 0.001). For cirrhotic patients, serum α-fetoprotein level at the end of treatment (EOT-AFP) was the strongest predictor of de novo HCC. The 1-year cumulative de novo HCC rates were 1.4% and 13.1% in the EOT-AFP < 9.0 ng/mL and ≥ 9.0 ng/mL groups (cut-off value) respectively (log-rank test: P < 0.001). The 1-year cumulative rates of HCC recurrence were 6.5% and 23.1% for the noncirrhosis and cirrhosis groups respectively (log-rank test: P = 0.023). For cirrhotic patients, previous HCC characteristics were significantly associated with HCC recurrence. In contrast, sex, age and metabolic features did not influence de novo HCC or recurrence.
Conclusions
For cirrhotic patients after elimination of HCV, serum EOT-AFP level and previous HCC characteristics would be useful markers for predicting de novo HCC or recurrence.
For cirrhotic patients after elimination of HCV, serum EOT-AFP level and previous HCC characteristics would be useful markers for predicting de novo HCC or recurrence.
DISCUSSION
Full Text Available Online
Download PDF
This large-scale, multicentre cohort study provides a great amount of data regarding the frequency of and demographic factors related to the development of HCC by SVR patients following interferon-free SOF-based treatment for 12 weeks. As would be expected, the cumulative incidence of HCC was significantly higher for cirrhotic patients than for noncirrhotic patients, irrespective of a past history of HCC. It is worth noting that the serum EOT-AFP level was the strongest predictor of de novo HCC for cirrhotic patients in the short-term after the initiation of DAA. The risk of early de novo HCC or recurrence for patients with chronic HCV treated with DAAs has received considerable attention in recent years. A better understanding of HCC development by patients treated with DAAs who achieve SVR is necessary to provide better care in real-life practice. Advantages of this study include that we divided the patents into groups based on cirrhosis status and focused on longitudinal biochemical parameters, such as serum albumin, ALT and AFP levels, which allowed us to better evaluate the risk of HCC development by SVR patients during and after interferon-free treatment with DAAs.
According to recent reports from Europe,[14, 15] SVR patients with previous HCC have high risk of HCC recurrence in the short-term, which has been hypothesised to be because the rapid HCV load decline brings about a misbalance of HCV-stimulated immune control that allows tumour clones to emerge.[23] On the other hand, multicentre studies that included different study designs, baseline characteristics and treatment options for prior HCC do not support this finding.[16, 24, 25] In this study, the overall 1-year cumulative HCC recurrence rate was 16.5%, with a lower annual recurrence rate after resection or percutaneous ablation.[26] However, the 1-year cumulative HCC recurrence rate was 23.1% when limited to cirrhotic patients. Previously, we reported on HCC recurrence after treatment with pegylated-interferon alpha plus RBV, finding a 1-year cumulative HCC recurrence rate of 22.2%,[27] with no significant difference between treatment with interferon and DAAs. We were not able to provide a new perspective because of differences in the age-composition of the treatment option groups (interferon median age, 57: interferon-free DAA median age, 66), which made it difficult to analyse the data with propensity-score matching. Moreover, selection bias was an important consideration because of the fact that DAAs allow patients who in the past would never qualify for interferon therapy to be treated. A prospective cohort in comparison with untreated patients showed no evidence of an increased risk of early HCC recurrence in DAA treatment.[16] Although it is conceivable that the anti-viral treatment options are not associated with the HCC recurrence seen in our study, further studies will be necessary to better elucidate the relation.
The strength of this study is the evaluation of the risk factors for the development of HCC. In this era of interferon-based treatment, including combination therapy with DAA, the risk of HCC development for SVR patients is modulated by host factors such as older age, abnormalities of glucose metabolism such as diabetes mellitus and insulin resistance, metabolic features, high serum AFP level and the severity of fibrosis.[28-31] Interestingly, in the DAA-only treatment of this study, neither age nor metabolic features were significantly associated with de novo HCC and recurrence; the long-term risk factors for HCC should be elucidated in future study. Instead, the serum EOT-AFP level was the most important predictor of de novo HCC for cirrhotic patients. Although the AFP level of almost all patients improves during DAA treatment, the longitudinal AFP level may play a role in early de novo HCC. In the group with previous HCC, cirrhotic patients were significantly more prone to experience HCC recurrence, but the AFP levels at baseline and EOT were not significant factors. This finding was similar to a previous report by Conti et al. that reported that liver stiffness by transient elastography was associated with HCC recurrence.[15] In fact, among our 21 patients who had HCC recurrence after DAA treatment, 15 (71.4%) had decreased AFP <10 ng/mL at EOT. The expression of several biochemical markers has been used to predict poor prognosis of HCC recurrence. On the other hand, characteristics of previous HCC, such as a short time from previous HCC treatment to initiation of DAA, within 1 year, and noncurative HCC treatment procedures, were significantly associated with early HCC recurrence, despite our exclusion of the patients with potential HCC at the time of screening, as stated in the Methods section. Therefore, close HCC screening should be required for all patients with previous HCC, during and after viral clearance.
In contrast to patients with advanced fibrosis or cirrhosis, the risk of HCC development is very low for patients with no/mild fibrosis who achieved SVR. Thus, most HCV guidelines have little mention of HCC surveillance for these patients, except for those with other liver disease, such as from alcohol consumption, non-alcoholic fatty liver disease/steatohepatitis, or metabolic disease including obesity (body mass index >25 kg/m2) and/or diabetes. In our study, among the 903 patients with no/mild fibrosis and no history of HCC who achieved SVR, 565 (62.6%) fit these criteria and none developed de novo HCC during the short-term follow-up period.
Six patients (0.4%) died during the follow-up period, including 3 from liver-related complications. It must be emphasised that only 1 of the patients without previous HCC died of liver-related complications. In this regard, the benefits of treatment with DAAs before the development of HCC include a decreased risk of death, reduced need for liver transplantation and less development of HCC over the long term. These benefits apply to patients with previous HCC because all patients, with or without a history of HCC, have short-term restoration of the liver functional reserve. Continued follow-up will be necessary to determine the potential long-term benefits related to the development of HCC and mortality.
The limitations of our study include factors related to its observational, real-life design, including potential physician prescribing bias and incomplete patient records. However, over 99% of the data for all items, except for EOT-AFP level (91.3%) and IL28B genotype (77.8%), were available in this study, and there were only 3 patients lost to follow-up. Taken together, our analysis was done from highly precise data that gives important insights into the risk of HCC development in a diverse patient population managed in clinical practice. Another limitation is the lack of data on decompensated cirrhosis patients because they are not covered by public insurance in Japan, which may have contributed to the relatively low rate of HCC incidence. An advantage is that many elderly patients were included in this study because the age of HCV-infected patients is increasing in Japan. Our findings will be useful for HCV management in other countries, with a similar trend forecast in Europe and the USA.[32]
In conclusion, our data showed no evidence of an increased risk of de novo HCC or recurrence by patients treated with interferon-free SOF-based regimens. However, cirrhosis was strongly associated with the short-term development of HCC after HCV eradication. Moreover, our findings underscore the fact that the serum EOT-AFP level is a useful marker for predicting de novo HCC for cirrhotic patients and that close HCC surveillance should be required for patients with a past history of HCC, especially for patients treated with noncurative procedures. Going forward, further studies will be required to elucidate the risk of HCC development over the long-term.
Continue to full text article - http://onlinelibrary.wiley.com/doi/10.1111/apt.14380/full
Early View
(Online Version of Record published before inclusion in an issue)
Full Text Available Online
Download PDF
This large-scale, multicentre cohort study provides a great amount of data regarding the frequency of and demographic factors related to the development of HCC by SVR patients following interferon-free SOF-based treatment for 12 weeks. As would be expected, the cumulative incidence of HCC was significantly higher for cirrhotic patients than for noncirrhotic patients, irrespective of a past history of HCC. It is worth noting that the serum EOT-AFP level was the strongest predictor of de novo HCC for cirrhotic patients in the short-term after the initiation of DAA. The risk of early de novo HCC or recurrence for patients with chronic HCV treated with DAAs has received considerable attention in recent years. A better understanding of HCC development by patients treated with DAAs who achieve SVR is necessary to provide better care in real-life practice. Advantages of this study include that we divided the patents into groups based on cirrhosis status and focused on longitudinal biochemical parameters, such as serum albumin, ALT and AFP levels, which allowed us to better evaluate the risk of HCC development by SVR patients during and after interferon-free treatment with DAAs.
According to recent reports from Europe,[14, 15] SVR patients with previous HCC have high risk of HCC recurrence in the short-term, which has been hypothesised to be because the rapid HCV load decline brings about a misbalance of HCV-stimulated immune control that allows tumour clones to emerge.[23] On the other hand, multicentre studies that included different study designs, baseline characteristics and treatment options for prior HCC do not support this finding.[16, 24, 25] In this study, the overall 1-year cumulative HCC recurrence rate was 16.5%, with a lower annual recurrence rate after resection or percutaneous ablation.[26] However, the 1-year cumulative HCC recurrence rate was 23.1% when limited to cirrhotic patients. Previously, we reported on HCC recurrence after treatment with pegylated-interferon alpha plus RBV, finding a 1-year cumulative HCC recurrence rate of 22.2%,[27] with no significant difference between treatment with interferon and DAAs. We were not able to provide a new perspective because of differences in the age-composition of the treatment option groups (interferon median age, 57: interferon-free DAA median age, 66), which made it difficult to analyse the data with propensity-score matching. Moreover, selection bias was an important consideration because of the fact that DAAs allow patients who in the past would never qualify for interferon therapy to be treated. A prospective cohort in comparison with untreated patients showed no evidence of an increased risk of early HCC recurrence in DAA treatment.[16] Although it is conceivable that the anti-viral treatment options are not associated with the HCC recurrence seen in our study, further studies will be necessary to better elucidate the relation.
The strength of this study is the evaluation of the risk factors for the development of HCC. In this era of interferon-based treatment, including combination therapy with DAA, the risk of HCC development for SVR patients is modulated by host factors such as older age, abnormalities of glucose metabolism such as diabetes mellitus and insulin resistance, metabolic features, high serum AFP level and the severity of fibrosis.[28-31] Interestingly, in the DAA-only treatment of this study, neither age nor metabolic features were significantly associated with de novo HCC and recurrence; the long-term risk factors for HCC should be elucidated in future study. Instead, the serum EOT-AFP level was the most important predictor of de novo HCC for cirrhotic patients. Although the AFP level of almost all patients improves during DAA treatment, the longitudinal AFP level may play a role in early de novo HCC. In the group with previous HCC, cirrhotic patients were significantly more prone to experience HCC recurrence, but the AFP levels at baseline and EOT were not significant factors. This finding was similar to a previous report by Conti et al. that reported that liver stiffness by transient elastography was associated with HCC recurrence.[15] In fact, among our 21 patients who had HCC recurrence after DAA treatment, 15 (71.4%) had decreased AFP <10 ng/mL at EOT. The expression of several biochemical markers has been used to predict poor prognosis of HCC recurrence. On the other hand, characteristics of previous HCC, such as a short time from previous HCC treatment to initiation of DAA, within 1 year, and noncurative HCC treatment procedures, were significantly associated with early HCC recurrence, despite our exclusion of the patients with potential HCC at the time of screening, as stated in the Methods section. Therefore, close HCC screening should be required for all patients with previous HCC, during and after viral clearance.
In contrast to patients with advanced fibrosis or cirrhosis, the risk of HCC development is very low for patients with no/mild fibrosis who achieved SVR. Thus, most HCV guidelines have little mention of HCC surveillance for these patients, except for those with other liver disease, such as from alcohol consumption, non-alcoholic fatty liver disease/steatohepatitis, or metabolic disease including obesity (body mass index >25 kg/m2) and/or diabetes. In our study, among the 903 patients with no/mild fibrosis and no history of HCC who achieved SVR, 565 (62.6%) fit these criteria and none developed de novo HCC during the short-term follow-up period.
Six patients (0.4%) died during the follow-up period, including 3 from liver-related complications. It must be emphasised that only 1 of the patients without previous HCC died of liver-related complications. In this regard, the benefits of treatment with DAAs before the development of HCC include a decreased risk of death, reduced need for liver transplantation and less development of HCC over the long term. These benefits apply to patients with previous HCC because all patients, with or without a history of HCC, have short-term restoration of the liver functional reserve. Continued follow-up will be necessary to determine the potential long-term benefits related to the development of HCC and mortality.
The limitations of our study include factors related to its observational, real-life design, including potential physician prescribing bias and incomplete patient records. However, over 99% of the data for all items, except for EOT-AFP level (91.3%) and IL28B genotype (77.8%), were available in this study, and there were only 3 patients lost to follow-up. Taken together, our analysis was done from highly precise data that gives important insights into the risk of HCC development in a diverse patient population managed in clinical practice. Another limitation is the lack of data on decompensated cirrhosis patients because they are not covered by public insurance in Japan, which may have contributed to the relatively low rate of HCC incidence. An advantage is that many elderly patients were included in this study because the age of HCV-infected patients is increasing in Japan. Our findings will be useful for HCV management in other countries, with a similar trend forecast in Europe and the USA.[32]
In conclusion, our data showed no evidence of an increased risk of de novo HCC or recurrence by patients treated with interferon-free SOF-based regimens. However, cirrhosis was strongly associated with the short-term development of HCC after HCV eradication. Moreover, our findings underscore the fact that the serum EOT-AFP level is a useful marker for predicting de novo HCC for cirrhotic patients and that close HCC surveillance should be required for patients with a past history of HCC, especially for patients treated with noncurative procedures. Going forward, further studies will be required to elucidate the risk of HCC development over the long-term.
Continue to full text article - http://onlinelibrary.wiley.com/doi/10.1111/apt.14380/full
Early View
(Online Version of Record published before inclusion in an issue)
No comments:
Post a Comment