Saturday, October 21, 2017

Editorial: hepatitis C virus (HCV) disease progression – HCV cure and the elimination of the “ethnic slope”

November 2017 Volume 46, Issue 10 Pages 911–1028

Editorial: hepatitis C virus (HCV) disease progression – HCV cure and the elimination of the “ethnic slope”
Tau, S. C. Pappas

First published: 20 October 2017
DOI: 10.1111/apt.14283

Linked Content
This article is linked to Le et al papers.

Ethnic disparities in health outcomes demand attention and explanation, as they themselves are plights to be abolished. We read with great interest the study by Le et al., which confirms that ethnic disparities exist in hepatitis C virus (HCV)-related liver disease progression and overall survival, and suggests that achieving SVR abolishes the ethnic disparities.[1] But what is the root cause of apparent baseline disparities—biology or bias in observation?

Imagine HCV as a ball rolling down different “ethnic slopes,” each representing different rates of progression to, and of, cirrhosis. Does HCV progress more rapidly in Hispanics and Asians, or are they simply presenting later in their disease course? Is it possible the balls started down the same slope but at different times?

If Asian and Hispanic patients had been infected significantly earlier than White and African American patients, their outcomes at baseline and follow-up would be worse. In this study, Asians had fewer co morbidities, and Asians and Hispanics tend to be immigrants. Both are thus presumably less likely to seek or have access to care and be diagnosed. The progression to, and of, cirrhosis seems susceptible to this bias of earlier infection as Lu et al. recognise but unfortunately were unable to study.

There is another hypothesis: racial disparities in fact exist, both before and after SVR (the balls roll on at different slopes, albeit slower), but the SVR cohort is, by nature of treatment selection, for example with interferon, a selectively healthier and smaller cohort, potentially reducing the ability to detect a difference in outcomes between ethnicities when there may be one. Furthermore, by definition, a patient in the SVR cohort could not have died or developed cirrhosis or HCC before achieving SVR, or else he would be excluded in the analysis of that outcome (immortal time bias).[2] Thus, the authors may have selected for HCV patients who are earlier in their disease and the time to detect differences in progression to cirrhosis or HCC is prolonged and not captured by the study—irrespective of achieving SVR. It would be important to see if ethnic disparity remains specifically in the cohort who were treated but did not achieve SVR, excluding those subjects with SVR and subjects not treated. If, in this cohort, ethnic disparities are abolished, then the authors’ hypothesis that abolition of disparities is related to achieving SVR is contradicted.

While further studies are needed to confirm the findings of Lu et al., it is quite plausible that once SVR is achieved, poor outcomes are delayed or stopped equally in all races. Once one removes the ball altogether (SVR), why does the slope downward matter? The force driving disease progression is eliminated. Indeed, studies show that curing HCV halts, or at least slows down, progression of liver disease.[3, 4] So cure may be colour blind, but is disease? The question matters in healthcare settings where triaging patients is warranted due to the expense of HCV anti-viral medications. A controversial implication is that if HCV behaves biologically differently among different ethnicities, then prioritisation should be a function of ethnicity. However, if it were simply that Hispanics and Asians present later in their disease, then the focus should be early diagnosis by expanding access to care.

Le et al. are to be commended for bringing our attention back to ethnic disparities in HCV-related outcomes and providing data that suggests SVR eliminates these disparities. As the populations of the United States and Canada or other countries swing to an ethnic mix of increasing Hispanic and Asian populations, further understanding of the cause of the disparities is key to developing diagnostic and treatment strategies to abolish them.
Editorial: hepatitis C virus (HCV) disease progression – HCV cure and the elimination of the “ethnic slope”
K. Le,G. Garcia,M. H. Nguyen
First published: 20 October 2017
We thank Drs. Tau and Pappas for their insightful comments on our paper.[1, 2] We agree that understanding the cause of ethnic disparities in health outcomes is crucial in developing strategies to effectively eliminate them.

While it remains to be seen whether there is a biological component to the increased risk for advanced liver disease observed in Hispanic and Asian patients in our study, there have been studies that have called to attention the barriers to care that impact ethnic minorities with chronic hepatitis C.

For Asian and Hispanic immigrant groups in particular, acculturation plays a major role in modifying behavioural, social and health characteristics of immigrants which lead to a decline in health over time.[3] Immigrants have higher unemployment and poverty rates and lower rates of health insurance coverage, especially in Hispanic and Southeast Asian groups.[4] Furthermore, in a recent population based study, Vutien et al found racial minority status to be independently associated with not receiving anti-viral treatment for chronic hepatitis C even after adjustment for socioeconomic status and medical/psychiatric comorbidities, indicating that additional ethnicity-related factors affecting treatment rates were present.[5] Therefore, while earlier age of infection in Hispanic and Asian patients may be a contributing factor to the disparities in disease progression observed in our study, there are additional socioeconomic, cultural and insurance barriers that exist for minorities which further exacerbate the disparity and should be addressed.

To address concerns over immortal time bias, we conducted further analysis on the patients who received treatment but did not achieve SVR. There continued to be significant differences in the 10-year cumulative incidence of cirrhosis among the ethnic groups with Hispanics (75.6%) and Asians (70.8%) having the highest incidence, followed by African Americans (54.6%) and Caucasians (53.6%) (log-rank test P-value: .0025). On multivariate analysis (Table 1), Hispanic patients in this group had a 52% increased risk of developing cirrhosis (adjusted HR 1.52, CI 1.09-2.12, = .014) compared to Caucasians. Significant differences also continued to persist in 10-year cumulative HCC incidence with Asians (36.3%) having the highest incidence, followed by Hispanics (29.7%), Caucasians (19.2%) and lowest in African Americans (13.4%). Asians in this group had a 63% increased risk (adjusted HR 1.63, CI 1.04-2.54, = .032) of HCC compared to Caucasians. Overall, ethnic disparities persisted in patients who received treatment but failed to achieve SVR. These findings confirm our conclusion that the abolition of ethnic disparities is related to achieving SVR.

Table 1. Cox proportional hazards regression analyses for progression to cirrhosis and HCC in patients who have received anti-viral therapy but did not achieve SVR

Unadjusted HR (95% CI)P-valueAdjusted HR (95% CI)P-value
  1. HR, hazard ratio; CI, confidence interval; HBV, hepatitis B virus; HCC, hepatocellular carcinoma; MELD, model for end-stage liver disease.
A. Cirrhosis
Caucasian1 (Referent)1 (Referent)
Hispanic1.61 (1.21-2.14).0011.52 (1.09-2.12).014
Asian1.58 (1.09-2.29).0161.25 (0.81-1.92).32
African American1.07 (0.65-1.77).781.52 (1.09-2.12).97
Age (5-yr intervals)1.20 (1.12-1.27) <.0011.15 (1.07-1.24) <.001
Male1.07 (0.85-1.35).571.02 (0.78-1.34).9
Diabetes1.63 (1.28-2.09) <.0011.21 (0.90-1.63).21
HBV coinfection1.45 (0.90-2.34).131.26 (0.73-2.19).41
Albumin0.77 (0.63-0.94).0120.92 (0.74-1.15).45
Platelets0.993 (0.992-0.995) <.0010.995 (0.993-0.997) <.001
Caucasian1 (Referent)1 (Referent)
Hispanic1.38 (0.97-1.96).071.37 (0.93-2.03).11
Asian2.44 (1.67-3.57) <.0011.63 (1.04-2.54).032
African American0.97 (0.47-2.00).940.99 (0.43-2.31).99
Age (5-yr intervals)1.39 (1.29-1.51) <.0011.39 (1.25-1.54) <.001
Male2.50 (1.80-3.49) <.0012.61 (1.78-3.81) <.001
Diabetes2.61 (1.96-3.47) <.0011.52 (1.10-2.11).012
HBV coinfection2.30 (1.36-3.90).0021.74 (0.93-3.27).083
Cirrhosis20.21 (8.31-49.14) <.00111.10 (3.52-35.03) <.001
MELD1.03 (1.01-1.05).0021.01 (0.98-1.03).54

We believe the root cause of ethnic disparities in HCV disease progression is multifactorial and can include longer duration of infection and poorer linkage to care besides potential unknown race-specific genetic factors. However, whether the ball is rolling down the same “ethnic slope” at different times or different “ethnic slopes” at the same time, it appears that once the “ball” (the disease driving force) is removed (by SVR), the slope downward no longer matters. The results should encourage and drive additional efforts in early screening, diagnosis and linkage to appropriate care of chronic hepatitis C patients of all ethnic backgrounds.

No comments:

Post a Comment