關於恩替卡韋（ETV）與富馬酸替諾福韋酯（TDF）治療的慢性

StephenW

Longitudinal real world study on estimated glomerular filtration rate (eGFR) changes in entecavir (ETV) versus tenofovir disoproxil fumarate (TDF)-treated chronic hepatitis B patients: a Real-B Study.

Worsening Kidney Function With TDF Than Entecavir for HBV in Big Cohort

AASLD The Liver Meeting Digital Experience, November 13-16, 2020

Mark Mascolini

A 10-year matched retrospective comparison of tenofovir disoproxil fumarate (TDF) versus entecavir (ETV) for chronic HBV infection linked TDF to worsening kidney function indicated by falling estimated glomerular filtration rate (eGFR) [1]. In a matched comparison in this 6189-person international study, TDF independently raised the risk of kidney impairment development or progression regardless of whether baseline (initial) eGFR lay above or below 60 mL/min.

Whether long-term kidney toxicity is worse with TDF or ETV in people taking those antivirals for chronic HBV infection remains controversial, according to the international team who led this Real-B Study. Retrospective studies and meta-analyses offer conflicting data on whether TDF carries a higher risk of kidney impairment in people with HBV. A 2000-2016 retrospective 410-person study concluded that TDF does not pose a greater risk of further kidney damage than ETV in people with HBV and a baseline eGFR above 60 mL/min [2].

To get a better fix on kidney function changes in people taking TDF or ETV for chronic hepatitis B infection, the Real-B team retrospectively studied 6189 adults with chronic HBV infection starting treatment for the first time with either TDF (2482 people) or ETV (3707 people).

Study participants in the United States, Hong Kong, Korea, Taiwan, Japan, and China had at least 12 months of follow-up after starting therapy at one of 22 centers. Researchers recorded CKD-EPI-calculated eGFR about every 6 months for all participants. They used propensity score matching [2] to create similar groups of TDF takers and ETV takers, matching the groups by age, gender, diabetes, hypertension, cirrhosis, initial eGFR, and length of follow-up.

Multivariable generalized linear modeling adjusted average eGFR during follow-up for cirrhosis, hypertension, and diabetes. Multivariable Cox regression identified factors independently linked to kidney function falling at least 1 level according to the following grading system: grade 1, eGFR above 90 mL/min; grade 2, eGFR 60-89; grade 3a, eGFR 45-59; grade 3b, eGFR 30-44, grade 4, eGFR 15-29; and grade 5, eGFR below 15.

Propensity score matching gave the researchers 1871 pairs of TDF and ETV takers with an initial eGFR at or above 60 mL/min and 520 pairs with an initial eGFR below 60 mL/min. For the above 60-and below 60-mL/min groups, ages averaged 47.0 with ETV and 47.5 with TDF and 54.1 with ETV and 53.7 with TDF. Respective proportions of men across those four groups were 54.9%, 56.9%, 97.3%, and 97.9%. And respective baseline eGFRs were 86.0, 85.5, 53.7, and 53.2 mL/min. Similar proportions taking ETV or TDF in the two matched groups had diabetes, hypertension, and cirrhosis. In the groups with baseline eGFR above 60, follow-up lasted 46.1 months with ETV and 46.5 months with TDF. In the groups with baseline eGFR below 60, follow-up lasted 49.0 months with ETV and 51.2 months with TDF.

For the total unmatched cohort, adjusted average eGFR through 120 months was significantly higher with ETV than TDF: 83.6 vs 82.5 mL/min (P = 0.002). In unmatched participants 50 or older at baseline, adjusted average eGFR through 120 months also proved significantly higher with ETV than with TDF (80.3 vs 76.4 mL/min, P < 0.0001). But in unmatched people younger than 50 at baseline, adjusted average eGFR was significantly lower with ETV than with TDF (86.4 vs 88.2, P = 0.001).

Average eGFRs adjusted by generalized linear modeling fell significantly more over time with TDF than with ETV in the entire unmatched study group, in the matched group starting with an eGFR above 60 mL/min, and in the matched group starting with an eGFR below 60 mL/min. The authors noted, however, that average eGFR differences between the TDF and ETV group were small, ranging from 1.2 to 3.1 mL/min.

In the propensity score-matched group of 1871 pairs starting the study with an eGFR above 60 mL/min, 5-year cumulative onset of kidney impairment was significantly greater with TDF than with ETV (49.8% vs 45.08%, P = 0.0039). Cox regression analysis determined that taking TDF rather than ETV independently raised chances of worsening kidney function by about 40% (hazard ratio [HR] 1.41). Other factors that independently boosted the risk of worsening kidney function were older age (HR 1.02), male gender (HR 2.16), having diabetes or hypertension (HR 1.4), and a higher FIB-4 fibrosis score (HR 1.02). A higher initial eGFR independently lowered chances of worsening kidney function (HR 0.99).

The investigators believe their findings indicate caution when prescribing TDF for HBV infection, especially in people with other risk factors for kidney impairment.

References
1. Mak LY, Hoang J, Jun DW, et al. Longitudinal real world study on estimated glomerular filtration rate (eGFR) changes in entecavir (ETV) versus tenofovir disoproxil fumarate (TDF)-treated chronic hepatitis B patients: a Real-B Study. AASLD The Liver Meeting Digital Experience, November 13-16, 2020. Abstract 21.
2. Trinh S, Le AK, Chang ET, et al. Changes in renal function in patients with chronic HBV infection treated with tenofovir disoproxil fumarate vs entecavir. Clin Gastroenterol Hepatol. 2019;17:948-956.e1. doi: 10.1016/j.cgh.2018.08.037.
3. Austin PC.  An introduction to propensity score methods for reducing the effects of confounding in observational studies. Multivariate Behav Res. 2011;46:399-424. doi: 10.1080/00273171.2011.568786.

StephenW

關於恩替卡韋（ETV）與富馬酸替諾福韋酯（TDF）治療的慢性乙型肝炎患者估計腎小球濾過率（eGFR）變化的縱向真實世界研究：一項Real-B研究。

與恩替卡韋相比，大劑量乙肝患者的TDF使腎臟功能惡化

AASLD肝臟會議數字體驗，2020年11月13日至16日

馬克·馬斯科利尼

替諾福韋富馬酸替諾福韋（TDF）與恩替卡韋（ETV）在慢性HBV感染中的10年回顧性比較將TDF與腎功能惡化聯繫起來，表明腎小球濾過率（eGFR）下降表明[1]。在這項有6189人進行的國際研究中的匹配比較中，無論基線（初始）eGFR高於還是低於60 mL / min，TDF均獨立增加了腎臟損害發生或發展的風險。

負責這項Real-B研究的國際團隊表示，對於那些使用這些抗病毒藥治療慢性HBV感染的人群，使用TDF或ETV進行長期腎臟毒性治療是否仍存在爭議。回顧性研究和薈萃分析提供了有關TDF是否在HBV患者中帶來更高腎臟損害風險的相互矛盾的數據。一項2000年至2016年的410人回顧性研究得出的結論是，對於HBV和基線eGFR高於60 mL / min的患者，與ETV相比，TDF不會造成更大的腎臟損害風險[2]。

為了更好地了解接受TDF或ETV治療的慢性乙型肝炎患者的腎臟功能變化，Real-B團隊回顧性研究了6189例患有慢性HBV感染的成年人，首次使用TDF（2482人）或ETV（ 3707人）。

美國，香港，韓國，台灣，日本和中國的研究參與者在22個中心之一開始治療後，至少進行了12個月的隨訪。研究人員大約每6個月記錄一次CKD-EPI計算的eGFR。他們使用傾向得分匹配[2]創建了類似的TDF使用者和ETV使用者組，並按照年齡，性別，糖尿病，高血壓，肝硬化，初始eGFR和隨訪時間對組進行了匹配。

肝硬化，高血壓和糖尿病的隨訪期間，多變量廣義線性模型調整了平均eGFR。多變量Cox回歸確定的與腎功能獨立相關的因素根據以下分級系統至少下降1級：1級，eGFR高於90 mL / min； 2級； 2級，eGFR 60-89； 3a級，eGFR 45-59； 3b級，eGFR 30-44、4級，eGFR 15-29； 5年級，eGFR低於15。

傾向得分匹配為研究人員提供了1871對TDF和ETV攝入者，其初始eGFR等於或高於60 mL / min，520對對初始eGFR低於60 mL / min。對於高於60 mL / min和低於60 mL / min的組，使用ETV的平均年齡為47.0，使用TDF的平均年齡為47.5，使用ETV的平均年齡為54.1，使用TDF的平均年齡為53.7。這四組中男性的比例分別為54.9％，56.9％，97.3％和97.9％。各自的基線eGFR為86.0、85.5、53.7和53.2 mL / min。兩組匹配的ETV或TDF服用比例相似，有糖尿病，高血壓和肝硬化。在基線eGFR高於60的組中，ETV隨訪46.1個月，TDF隨訪46.5個月。在基線eGFR低於60的組中，ETV隨訪49.0個月，TDF隨訪51.2個月。

對於總的不匹配隊列，ETV的120個月調整後平均eGFR顯著高於TDF：83.6 vs 82.5 mL / min（P = 0.002）。在基線時年齡為50歲或以上的無匹配參與者中，ETV的調整後平均eGFR至120個月也明顯高於TDF（80.3 vs 76.4 mL / min，P <0.0001）。但是在基線時年齡小於50歲的無與倫比人群中，ETV的調整後平均eGFR顯著低於TDF（86.4 vs 88.2，P = 0.001）。

在整個不匹配的研究組中，在匹配組中從eGFR高於60 mL / min開始，而在匹配組中從eGFR低於60 mL開始，通過廣義線性建模調整的平均eGFR隨著時間的流逝比ETV下降得更多。 /分鐘。然而，作者指出，TDF和ETV組之間的平均eGFR差異很小，範圍為1.2至3.1 mL / min。

在eGFR高於60 mL / min的研究中，傾向評分匹配組中有1871對，與ETV相比，TDF的5年累積腎損害發病率顯著高於ETV（49.8％vs 45.08％，P = 0.0039）。 Cox回歸分析確定，單獨服用TDF而不是ETV可使腎功能惡化的機會增加約40％（危險比[HR] 1.41）。獨立增加腎臟功能惡化風險的其他因素是年齡較大（HR 1.02），男性（HR 2.16），患有糖尿病或高血壓（HR 1.4）以及較高的FIB-4纖維化評分（HR 1.02）。較高的初始eGFR獨立降低腎功能惡化的機會（HR 0.99）。
研究人員認為，他們的發現在為TDF開具HBV感染處方時特別謹慎，特別是在有其他腎功能損害危險因素的人群中。

參考文獻
1. Mak LY，Hoang J，Jun DW等。 關於恩替卡韋（ETV）與替諾福韋富馬酸二甲吡呋酯（TDF）治療的慢性乙型肝炎患者估計腎小球濾過率（eGFR）變化的縱向真實世界研究：一項Real-B研究。 AASLD肝臟會議數字體驗，2020年11月13日至16日。摘要21。
2. Trinh S，Le AK，Chang ET等。 替諾福韋酯富馬酸vs恩替卡韋治療慢性HBV感染患者腎功能的變化。 Castro Gastroenterol Hepatol。 2019; 17：948-956.e1。 doi：10.1016 / j.cgh.2018.08.037。
3.奧斯汀PC。 傾向評分方法的簡介，以減少觀察研究中混淆的影響。 多元行為研究。 2011; 46：399-424。 doi：10.1080 / 00273171.2011.568786。