Editor's note: China is a major country with liver cancer, and the incidence rate and mortality rate of liver cancer account for about half of the world's total. Early identification of high-risk populations and early diagnosis of liver cancer are key to improving the overall survival rate of patients. Since the discovery of vitamin K in 1929, researchers have gradually confirmed the important role of serum abnormal prothrombin caused by its deficiency in the diagnosis of liver cancer. In 2019, Dr. Feng Shen and his team from Eastern Hepatobiliary Surgery Hospital, the third affiliated hospital of the Naval Medical University, collaborated with other top hospital teams in China to establish and validate the ASAP risk prediction model. Since then, the ASAP model has continuously broken through the prediction limitations of hepatitis B-related liver cancer, and has achieved frequent success in predicting hepatitis C-related liver cancer, non-alcoholic fatty liver disease-related liver cancer, and early recurrence diagnosis of liver cancer.In June 2023, the ASAP model was honored with the “UNIVANTS Excellence in Medical Care Award” for its significant contributions to clinical practice in China. In addition, Chinese Taiwan scholars released the ASAP model in the same month to predict the high performance of the risk of liver cancer in untreated chronic hepatitis B patients within one year, with an AUROC of up to 90%, further exploring the clinical value of ASAP and broadening its scope of application. So, there have been many good news recently. For this reason, Hepatology Digest invited Dr. Feng Shen to review the development of PIVKA-II and ASAP, comment on the latest progress, especially in the exploration of Chinese population, and look forward to its future clinical application.
Overview of the development of the ASAP model
In 1929, Danish biologist Henrik Dam discovered vitamin K and won the Nobel Prize in 1943.
In 1968, Ganro PO and Niléhn JE discovered that vitamin K deficiency would produce serum abnormal prothrombin PIVKA, Protein Induced by Vitamin K Absence or Antagonist, DCP, Des-gamma-carboxy [abnormal] prothrombin, des-gamma-carboxy-prothrombin.
In 1984, Liebman HA first discovered that the level of PIVKA-II increased in most patients with hepatocellular carcinoma (HCC), and could be used as a biomarker for the diagnosis of HCC.
In 1994, Suehiro T first confirmed that PIVKA-II could be used as a biomarker for predicting the prognosis of HCC.
In 2011, the Ministry of Health of China issued the “Diagnostic and Treatment Standards for Primary Liver Cancer 2011 Edition”, recommending PIVKA-II as one of the auxiliary diagnostic tumor markers.
In 2014, the Japanese Liver Disease Society recommended that the combined imaging examination of alpha-fetoprotein (AFP), PIVKA-II, and alpha-fetoprotein variant (AFP-L3) be used as a monitoring program for extremely high-risk liver cancer patients.
In 2016, Abbott launched the PIVKA-II detection marker in China.
In 2018, the reference interval for the PIVKA-II test marker in the Chinese population was released, with a 95th percentile upper limit of 40.38 mAU/ml.
In 2019, Dr. Feng Shen’s team worked with 11 well-known top-three hospitals in China to jointly research and successfully develop the ASAP liver cancer risk assessment model, which included four factors: age Age, gender Sex, serum alpha-fetoprotein AFP level, and serum abnormal prothrombin PIVKA-II level. The relevant content was published in the top journal of clinical chemistry, Clinical Chemistry, with an IF of 11.13.
In 2020, with the permission of Dr. Feng Shen and Dr. Tian Yang, the ASAP online computing applet was launched, which promoted this major scientific research achievement to clinical practice for free.
In 2020, China’s “Guidelines for Stratified Screening and Monitoring of Primary Liver Cancer 2020 Edition” recommended the combined use of PIVKA-II testing to improve the detection rate of early liver cancer, with a B2 level recommendation.
In 2022, the China’s National Health Commission issued the “2022 Edition of the Diagnostic and Treatment Standards for Primary Liver Cancer”, recommending PIVKA-II as a marker for early diagnosis of liver cancer for clinical Grade 1A recommendation.
In 2022, the “Chinese Population Liver Cancer Screening Guidelines 2022” published by academician He Jie and others suggested that serum markers such as PIVKA-II could be used as a complementary screening technique, but it could not completely replace ultrasound US combined with AFP testing. The quality of evidence was: medium; Recommendation intensity: strong.
In 2022, Dr. Feng Shen and Dr. Tian Yang published articles in SCI journals, gradually promoting ASAP to screening applications for all liver cancer risk groups.
In June 2023, the ASAP liver cancer risk assessment model was honored with the “Univants Excellence in Healthcare Award” for its exceptional practices in China. This award is jointly selected and presented by seven globally leading medical associations: IFCC, AACC, EHMC, Modern Healthcare, HIMSS, NAHQ, and IHE.
Establish an ASAP model tailored to the Chinese population
China is a big country with hepatitis B and liver cancer, and the number of new liver cancer patients in China accounts for about half of the world’s total. For HCC patients in China, about 84% are caused by HBV infection. The prevalence of HBsAg in the general population in China is 6.1%, and there are 86 million chronic HBV infected patients. Therefore, it is very important to establish an effective evaluation model to monitor the risk of disease occurrence and development in HBV-infected patients, achieve early screening, diagnosis, and treatment of the disease, and thus avoid the occurrence of HCC.
In 2019, Dr. Feng Shen’s team led 11 large-scale top-three hospitals in China to create the first hepatitis B-related liver cancer diagnosis model suitable for the majority of Chinese people – the ASAP model, which uses CHB patients as the monitoring population and incorporates four independent risk factors, including patient age Age, gender Sex, serum alpha-fetoprotein AFP level, and serum abnormal prothrombin PIVKA-II level, as predictive variables to predict the risk of primary hepatocellular carcinoma in CHB patients. The relevant content was published in the top journal of clinical chemistry, Clincal Chemistry, as an editorial assignment, with an IF:11.13.
The 11 centers involved in the modeling included 2,925 subjects, and all blood samples were sent to the same testing center for testing. All patients with liver cancer were diagnosed by surgery or pathological biopsy. Among them, 2198 cases participated in the training cohort to establish the model, and 727 cases participated in the validation cohort to validate the model. Through multi-factor analysis, age, gender, AFP, and PIVKA-II were identified as independent risk factors, and an ASAP nomogram model was constructed. The validation results showed that the sensitivity and specificity of the ASAP model in the training cohort were 76.1% and 90.4%, respectively, while in the validation cohort they were 73.8% and 90.0%, respectively, which were much higher than the diagnostic efficacy of AFP or PIVKA-II alone. The ASAP model, as the first predictive model specifically for countrymen, showed extremely high predictive performance and great application value. The following year, the ASAP online mini-program was launched, providing great convenience for clinical applications.
ASAP frontier trend: continue to explore the depth and breadth of application of Chinese people
- AFP+PIVKA-II predicts hepatitis C virus (HCV)-related HCC
The 5-year risk of progression to HCC in HCV-infected patients is 7% to 14%. The overall survival rate of early treatment can reach 50% to 80% in 5 years, but it is less than 10% if it progresses to advanced stage. Therefore, early screening and diagnosis of HCV-HCC can bring great benefits to patients. Currently, the most commonly used biomarker in clinical practice is AFP, with a sensitivity of 41% to 65% and a specificity of 80% to 94% for the diagnosis of HCV-HCC. In addition, PIVKA-II and AFP-L3 are also popular detection schemes in clinical research. Dr. Feng Shen and Dr. Tian Yang’s team conducted relevant research to find suitable solutions among the three detection methods mentioned above to improve the sensitivity of detection.
The study included 172 HCV controls and 105 HCV-HCC patients. The levels of AFP, AFP-L3, and PIVKA-II in patients with HCV-HCC were significantly higher than those in patients with HCV (P<0.001). In patients with early HCC defined by BCLC and TNM staging, PIVKA-II showed the best predictive performance AUC when tested alone: PIVKA-II vs. AFP vs. AFP-L3=0.90 vs. 0.80 vs. 0.69, P<0.001; In the collocation test group, PIVKA-II+AFP showed the best predictive performance AUC: PIVKA-II+AFP vs. AFP+AFP-L3 vs. PIVKA-II+AFP-L3=0.93 vs. 0.78 vs. 0.89, P<0.001, which was not different from the predictive performance AUC of the combination of the three: AFP+AFP-L3+PIVKA-II 0.93, P=0.277. Therefore, adding PIVKA-II detection to routine AFP testing may be a more suitable detection method.
2. ASAP model can be used as a predictor of HCV-HCC in countrymen
In the prediction of HCV-HCC, the well-known GALAD model, which consists of gender, age, alpha-fetoprotein variant AFP-L3, alpha-fetoprotein AFP, and abnormal prothrombin DCP, or PVIKA-II, has one more variable AFP-L3 compared to the ASAP model. The team of Feng Shen and Tian Yang analyzed the specific efficacy of the ASAP model without the AFP-L3 variable in diagnosing HCV-HCC.
The study included 193 HCV controls and 168 HCV-HCC patients. There were significant differences between the HCV-HCC group and the HCV control group in terms of age, gender, Child-Pugh classification, and cirrhosis, as well as in terms of single biomarkers such as AFP, PIVKA-II, and AFP-L3, as well as scores for ASAP and GALAD. Compared with a single tumor marker, both the ASAP score and GALAD score showed better diagnostic performance; In both the general population and the cirrhosis population, the diagnostic efficacy of the ASAP score had a trend of superiority over the GALAD score, with an AUC of 0.917 vs. 0.894 and a P-value of 0.057; In further analysis of early HCC subgroups and AFP-negative subgroups, ASAP score also demonstrated diagnostic efficacy that was not inferior to or even superior to GALAD. Therefore, the ASAP score may be more suitable for screening and monitoring liver cancer in Chinese patients with chronic hepatitis C, and it has greater health and economic benefits.
3. AFP+PIVKA-II predicts HCC associated with non-alcoholic fatty liver disease (NAFLD)
NAFLD has become one of the main causes of HCC worldwide, but the current ultrasound-based liver cancer surveillance strategy is still not ideal for screening high-risk populations such as hepatitis, cirrhosis, and NAFLD. The widespread application of liver cancer biomarker detection in clinical practice has greatly benefited risk detection. Professor Yang Tian’s team evaluated the diagnostic performance of three biomarkers, AFP, PIVKA-II, and AFP-L3, used alone or in combination to diagnose NAFLD-HCC in Chinese patients.
The study included 345 NAFLD controls and 139 NAFLD-HCC patients. Research shows that the median levels of AFP, PIVKA-II, and AFP-L3 in patients with NAFLD-HCC are higher than those in patients with NAFLD, P<0.001. In the detection of a single biomarker, the diagnostic performance of PIVKA-II was superior to that of AFP AUC: 0.869 vs. 0.763, P<0.001 and AFP-L3 AUC: 0.869 vs. 0.689, P<0.001; In the combined detection, the best AUC of AFP+PIVKA-II was: AFP+PIVKA-II vs. AFP+PIVKA-II+AFP-L3 vs. PIVKA-II+AFP-L3 vs. AFP+AFP-L3 0.906 vs. 0.904 vs. 0.881 vs. 0.759, and the addition of AFP-L3 did not improve the diagnostic performance AUC: 0.904, P=0.086. In addition, similar results were also obtained in the analysis of non-cirrhotic subgroup and early stage liver cancer subgroup, which filled the gap in Chinese population data and provided new detection combinations for clinical practice.
In addition, the team of Feng Shen and Tian Yang also compared the diagnostic efficacy of the ASAP model and the GALAD model for NAFLD-HCC, including sensitivity, specificity, positive predictive value, negative predictive value, diagnostic accuracy, and other indicators. At the 31st Annual Meeting of the Asia Pacific Association for the Study of the Liver (APASL) in 2022, Professor Yang Tian was invited to present the research in the form of an oral report.
The results of this study showed that the two combined diagnostic models for NAFLD-HCC in the Chinese population had higher diagnostic efficacy than any single serum marker for liver cancer, such as AFP, AFP-L3, and PIVKA-II, for both all stages and early stages of diagnosis; More importantly, the ASAP model is superior or non-inferior to the GALAD model in the diagnostic efficacy of NAFLD-HCC in Chinese population. Given that the ASAP model can detect one less AFP-L3 than the GALAD model, which greatly reduces the cost of detection, and the diagnostic efficacy of the ASAP model is as excellent as that of the GALAD model, the ASAP model is more suitable for screening and monitoring liver cancer in Chinese NAFLD patients.
4. PIVKA-II and ASAP predict the risk of HCC in untreated CHB patients within 1 year
The ASAP model is increasingly being accepted and used outside of Chinese Mainland. In June 2023, Am J Cancer Res reported a retrospective case-control study completed by scholars from Chinese Taiwan, which showed that both PIVKA-II levels and ASAP scores had high predictive performance for HCC in patients with untreated non-cirrhotic CHB within 1 year, especially in patients with normal alpha-fetoprotein levels. The area under the receiver operating characteristic curve (AUROC) of the ASAP score was as high as 90%, providing evidence for the ASAP model to benefit more patients.
The researchers included untreated non-cirrhotic CHB patients who were followed up at National Taiwan University Hospital, China, including 69 HCC patients in the HCC group and 102 non-HCC patients in the control group. The PIVKA-II levels in serum samples were measured before the occurrence of HCC, at the time of HCC diagnosis, or at the last follow-up. The baseline PIVKA-II level in the HCC group was significantly higher than that in the control group (36 vs. 24 mAU/mL, P=0.013), and the ASAP score in the HCC group was also significantly higher than that in the control group (0.48 vs. 0.25, P<0.001). Both PIVKA-II level and ASAP score can predict the occurrence of HCC within 1 year. The AUROC of PIVKA-II is 0.7695%CI:0.68-0.94, and the AUROC of ASAP score is 0.9095%CI:0.85-0.95. The AUROC of ASAP score in patients with AFP level <20 ng/mL is 0.8995%CI:0.83-0.95, as shown in Figure 5. The optimal critical value for predicting the occurrence of HCC within 1 year using PIVKA-II is 31 mAU/mL, with a sensitivity of 65% and a specificity of 83%; The optimal cutoff value for predicting the occurrence of HCC within 1 year using the ASAP score was 0.31, with a sensitivity of 92.5% and a specificity of 71.7%.
5. PIVKA-II predicts the risk of postoperative recurrence
For early HCC patients, about 50% of them will have recurrence after surgery, so it is very important to effectively assess the risk of recurrence and provide timely diagnosis and treatment. Currently, AFP is the most widely used biomarker for predicting the risk of HCC recurrence, but its sensitivity threshold of 20 ng/ml is only 60%, which is slightly insufficient. The emerging PIVKA-II, with a sensitivity of over 90% using a threshold of 40 mAU/ml, has gradually become a standard biomarker for HCC screening and diagnosis.
The team of Feng Shen and Tian Yang further explored the possibility of PIVKA-II predicting the risk of postoperative recurrence, and included patients undergoing HCC surgery. The levels of serum AFP and PIVKA-II were measured within one week before surgery or when recurrence was detected. After long-term follow-up, it was found that the proportion of PIVKA-II-positive patients among early recurrence patients was higher than that of AFP-positive patients, and the difference between the two was statistically significant. Multivariate analysis showed that preoperative PIVKA-II positivity, rather than preoperative AFP positivity, was an independent high-risk factor for early postoperative recurrence of HCC. Therefore, the positive PIVKA-II before surgery has clinical practical significance in predicting the recurrence of HCC after surgery.
Summary and outlook
In the past 100 years of continuous exploration of vitamin K and its PIVKA-II, the field of liver cancer prediction has shown its remarkable vitality. Based on the application of PIVKA-II and the development and maturity of the ASAP model, the prediction and diagnosis of HCC caused by HBV, HCV, NAFLD and other causes in Chinese population are becoming more and more perfect. At present, the ASAP model has been widely used throughout China and has become one of the routine testing items for physical examination screening and serological diagnosis for high-risk groups of liver cancer, receiving high praise from clinicians and laboratory doctors.
In addition, with the intelligent information solution AlinIQ AMS designed based on the ASAP model, ASAP further moved towards the international stage and won the “Univants Excellence in Healthcare Award”. The ASAP model, which is based on an intelligent information system, can not only predict the risk of early liver cancer, but also optimize the diagnosis and treatment path of liver cancer, which greatly improves the effectiveness of clinical diagnosis and treatment. It is an effective and promising cancer prevention strategy and multidisciplinary collaboration tool.
In the future, we hope that PIVKA-II and ASAP models will scale new heights, continue to contribute to the early screening, diagnosis and treatment of liver cancer in China, actively promote the optimization of early liver cancer diagnosis pathways, promote close collaboration among multiple disciplines, help China’s excellent medical practice to re-enter the world stage, and continuously explore the depth and breadth of early screening and diagnosis, benefiting more people at risk of liver cancer worldwide!
