User login
The FAST score had an overall sensitivity of 89% and an overall specificity of 89% with a defined rule-out cutoff of .35 or lower and rule-in cutoff of .67 or higher, respectively, Federico Ravaioli, MD, PhD, a gastroenterologist at the University of Modena & Reggio Emilia in Italy, and colleagues, wrote in Gut.
“These results could be used in clinical screening studies to efficiently identify patients at risk of progressive NASH, who should be referred for a conclusive liver biopsy, and who might benefit from treatment with emerging pharmacotherapies,” the authors wrote.
The research team analyzed 12 observational studies with 5,835 participants with biopsy-confirmed nonalcoholic fatty liver disease (NAFLD) between February 2020 and April 2022. They included articles that reported data for the calculation of sensitivity and specificity of the FAST score for identifying adult patients with fibrotic NASH based on a defined rule-out cutoff of .35 or lower and rule-in cutoff of .67 or higher. Fibrotic NASH was defined as patients with NASH plus a NAFLD activity score of 4 or greater and fibrosis stage 2 or higher.
The pooled prevalence of fibrotic NASH was 28%. The mean age of participants ranged from 40 to 60, and the proportion of men ranged from 23% to 91%. The mean body mass index ranged from 23 kg/m2 to 41 kg/m2, with a prevalence of obesity ranging from 23% to 100% and preexisting type 2 diabetes ranging from 18% to 60%. Nine studies included patients with biopsy-proven NAFLD from tertiary care liver centers, and three studies included patients from bariatric clinics or bariatric surgery centers with available liver biopsy data.
Fibrotic NASH was ruled out in 2,723 patients (45.5%) by a FAST score of .35 or lower and ruled in 1,287 patients (21.5%) by a FAST score of .67 or higher. In addition, 1,979 patients (33%) had a FAST score in the so-called “grey” intermediate zone.
Overall, the FAST score pooled sensitivity was 89%, and the pooled specificity was 89%. By the rule-out cutoff of .35, the sensitivity was 89% and the specificity was 56%. By the rule-in cutoff of .67, the sensitivity was 46% and the specificity was 89%.
At an expected prevalence of fibrotic NASH of 30%, the negative predictive value of the .35 cutoff was 92%, and the positive predictive value of the .67 cutoff was 65%. Across the included studies, the negative predictive value ranged from 77% to 98%, and the positive predictive value ranged from 32% to 87%.
For the rule-in cutoff of .67, at a pretest probability of 10%, 20%, 26.3%, and 30%, there was an increasing likelihood of detecting fibrotic NASH by FAST score at 32%, 52%, 60%, and 65%, respectively. For the rule-out cutoff of .35, at the same pretest probability levels, the likelihood of someone not having fibrotic NASH and not being detected by FAST score was 2%, 5%, 7%, and 8%, respectively.
In subgroup analyses, the sensitivity of the rule-out cutoff was significantly affected by the study design. In addition, age and BMI above the median both affected pooled sensitivity but not pooled specificity. On the other hand, the rule-in cutoff was significantly affected by study design, BMI above the median, and presence of preexisting type 2 diabetes above the median.
“Today, we stand on the cusp of a revolutionary time to treat NASH. This is due in part to the fact that many exciting, novel precision metabolic treatments are in the pipeline to combat this disease,” said Brian DeBosch, MD, PhD, associate professor of cell biology and physiology at the Washington University in St. Louis, who was not involved with this study.
“A major barrier in clinical NASH management is a rapid, noninvasive, and precise means by which to clinically stage such patients,” Dr. DeBosch said. “We now approach as closely as ever the sensitivity and specificity required to stratify the highest-risk patients, identify candidates for advanced therapy, and meaningfully reduce biopsies through using noninvasive testing.”
Dr. DeBosch noted the importance of pretest probability and specific subpopulations when deciding whether to use the FAST score. For instance, he said, a tertiary academic liver transplant center will see a different patient population than in a primary care setting. Also, in this study, the presence or absence of diabetes and a BMI above 30 significantly altered sensitivity and specificity.
“One important remaining question stemming from these data is whether FAST can also be used as a surrogate measure to follow disease regression over time following intervention,” Dr. DeBosch said. “Even if FAST is not useful in that way, defining individuals who most need to undergo biopsy and/or those who need to undergo treatment remain important uses for this test.”
The study authors did not declare a specific funding source or report any competing interests. DeBosch reported no relevant disclosures.
The FAST score had an overall sensitivity of 89% and an overall specificity of 89% with a defined rule-out cutoff of .35 or lower and rule-in cutoff of .67 or higher, respectively, Federico Ravaioli, MD, PhD, a gastroenterologist at the University of Modena & Reggio Emilia in Italy, and colleagues, wrote in Gut.
“These results could be used in clinical screening studies to efficiently identify patients at risk of progressive NASH, who should be referred for a conclusive liver biopsy, and who might benefit from treatment with emerging pharmacotherapies,” the authors wrote.
The research team analyzed 12 observational studies with 5,835 participants with biopsy-confirmed nonalcoholic fatty liver disease (NAFLD) between February 2020 and April 2022. They included articles that reported data for the calculation of sensitivity and specificity of the FAST score for identifying adult patients with fibrotic NASH based on a defined rule-out cutoff of .35 or lower and rule-in cutoff of .67 or higher. Fibrotic NASH was defined as patients with NASH plus a NAFLD activity score of 4 or greater and fibrosis stage 2 or higher.
The pooled prevalence of fibrotic NASH was 28%. The mean age of participants ranged from 40 to 60, and the proportion of men ranged from 23% to 91%. The mean body mass index ranged from 23 kg/m2 to 41 kg/m2, with a prevalence of obesity ranging from 23% to 100% and preexisting type 2 diabetes ranging from 18% to 60%. Nine studies included patients with biopsy-proven NAFLD from tertiary care liver centers, and three studies included patients from bariatric clinics or bariatric surgery centers with available liver biopsy data.
Fibrotic NASH was ruled out in 2,723 patients (45.5%) by a FAST score of .35 or lower and ruled in 1,287 patients (21.5%) by a FAST score of .67 or higher. In addition, 1,979 patients (33%) had a FAST score in the so-called “grey” intermediate zone.
Overall, the FAST score pooled sensitivity was 89%, and the pooled specificity was 89%. By the rule-out cutoff of .35, the sensitivity was 89% and the specificity was 56%. By the rule-in cutoff of .67, the sensitivity was 46% and the specificity was 89%.
At an expected prevalence of fibrotic NASH of 30%, the negative predictive value of the .35 cutoff was 92%, and the positive predictive value of the .67 cutoff was 65%. Across the included studies, the negative predictive value ranged from 77% to 98%, and the positive predictive value ranged from 32% to 87%.
For the rule-in cutoff of .67, at a pretest probability of 10%, 20%, 26.3%, and 30%, there was an increasing likelihood of detecting fibrotic NASH by FAST score at 32%, 52%, 60%, and 65%, respectively. For the rule-out cutoff of .35, at the same pretest probability levels, the likelihood of someone not having fibrotic NASH and not being detected by FAST score was 2%, 5%, 7%, and 8%, respectively.
In subgroup analyses, the sensitivity of the rule-out cutoff was significantly affected by the study design. In addition, age and BMI above the median both affected pooled sensitivity but not pooled specificity. On the other hand, the rule-in cutoff was significantly affected by study design, BMI above the median, and presence of preexisting type 2 diabetes above the median.
“Today, we stand on the cusp of a revolutionary time to treat NASH. This is due in part to the fact that many exciting, novel precision metabolic treatments are in the pipeline to combat this disease,” said Brian DeBosch, MD, PhD, associate professor of cell biology and physiology at the Washington University in St. Louis, who was not involved with this study.
“A major barrier in clinical NASH management is a rapid, noninvasive, and precise means by which to clinically stage such patients,” Dr. DeBosch said. “We now approach as closely as ever the sensitivity and specificity required to stratify the highest-risk patients, identify candidates for advanced therapy, and meaningfully reduce biopsies through using noninvasive testing.”
Dr. DeBosch noted the importance of pretest probability and specific subpopulations when deciding whether to use the FAST score. For instance, he said, a tertiary academic liver transplant center will see a different patient population than in a primary care setting. Also, in this study, the presence or absence of diabetes and a BMI above 30 significantly altered sensitivity and specificity.
“One important remaining question stemming from these data is whether FAST can also be used as a surrogate measure to follow disease regression over time following intervention,” Dr. DeBosch said. “Even if FAST is not useful in that way, defining individuals who most need to undergo biopsy and/or those who need to undergo treatment remain important uses for this test.”
The study authors did not declare a specific funding source or report any competing interests. DeBosch reported no relevant disclosures.
The FAST score had an overall sensitivity of 89% and an overall specificity of 89% with a defined rule-out cutoff of .35 or lower and rule-in cutoff of .67 or higher, respectively, Federico Ravaioli, MD, PhD, a gastroenterologist at the University of Modena & Reggio Emilia in Italy, and colleagues, wrote in Gut.
“These results could be used in clinical screening studies to efficiently identify patients at risk of progressive NASH, who should be referred for a conclusive liver biopsy, and who might benefit from treatment with emerging pharmacotherapies,” the authors wrote.
The research team analyzed 12 observational studies with 5,835 participants with biopsy-confirmed nonalcoholic fatty liver disease (NAFLD) between February 2020 and April 2022. They included articles that reported data for the calculation of sensitivity and specificity of the FAST score for identifying adult patients with fibrotic NASH based on a defined rule-out cutoff of .35 or lower and rule-in cutoff of .67 or higher. Fibrotic NASH was defined as patients with NASH plus a NAFLD activity score of 4 or greater and fibrosis stage 2 or higher.
The pooled prevalence of fibrotic NASH was 28%. The mean age of participants ranged from 40 to 60, and the proportion of men ranged from 23% to 91%. The mean body mass index ranged from 23 kg/m2 to 41 kg/m2, with a prevalence of obesity ranging from 23% to 100% and preexisting type 2 diabetes ranging from 18% to 60%. Nine studies included patients with biopsy-proven NAFLD from tertiary care liver centers, and three studies included patients from bariatric clinics or bariatric surgery centers with available liver biopsy data.
Fibrotic NASH was ruled out in 2,723 patients (45.5%) by a FAST score of .35 or lower and ruled in 1,287 patients (21.5%) by a FAST score of .67 or higher. In addition, 1,979 patients (33%) had a FAST score in the so-called “grey” intermediate zone.
Overall, the FAST score pooled sensitivity was 89%, and the pooled specificity was 89%. By the rule-out cutoff of .35, the sensitivity was 89% and the specificity was 56%. By the rule-in cutoff of .67, the sensitivity was 46% and the specificity was 89%.
At an expected prevalence of fibrotic NASH of 30%, the negative predictive value of the .35 cutoff was 92%, and the positive predictive value of the .67 cutoff was 65%. Across the included studies, the negative predictive value ranged from 77% to 98%, and the positive predictive value ranged from 32% to 87%.
For the rule-in cutoff of .67, at a pretest probability of 10%, 20%, 26.3%, and 30%, there was an increasing likelihood of detecting fibrotic NASH by FAST score at 32%, 52%, 60%, and 65%, respectively. For the rule-out cutoff of .35, at the same pretest probability levels, the likelihood of someone not having fibrotic NASH and not being detected by FAST score was 2%, 5%, 7%, and 8%, respectively.
In subgroup analyses, the sensitivity of the rule-out cutoff was significantly affected by the study design. In addition, age and BMI above the median both affected pooled sensitivity but not pooled specificity. On the other hand, the rule-in cutoff was significantly affected by study design, BMI above the median, and presence of preexisting type 2 diabetes above the median.
“Today, we stand on the cusp of a revolutionary time to treat NASH. This is due in part to the fact that many exciting, novel precision metabolic treatments are in the pipeline to combat this disease,” said Brian DeBosch, MD, PhD, associate professor of cell biology and physiology at the Washington University in St. Louis, who was not involved with this study.
“A major barrier in clinical NASH management is a rapid, noninvasive, and precise means by which to clinically stage such patients,” Dr. DeBosch said. “We now approach as closely as ever the sensitivity and specificity required to stratify the highest-risk patients, identify candidates for advanced therapy, and meaningfully reduce biopsies through using noninvasive testing.”
Dr. DeBosch noted the importance of pretest probability and specific subpopulations when deciding whether to use the FAST score. For instance, he said, a tertiary academic liver transplant center will see a different patient population than in a primary care setting. Also, in this study, the presence or absence of diabetes and a BMI above 30 significantly altered sensitivity and specificity.
“One important remaining question stemming from these data is whether FAST can also be used as a surrogate measure to follow disease regression over time following intervention,” Dr. DeBosch said. “Even if FAST is not useful in that way, defining individuals who most need to undergo biopsy and/or those who need to undergo treatment remain important uses for this test.”
The study authors did not declare a specific funding source or report any competing interests. DeBosch reported no relevant disclosures.
FROM GUT