You are on page 1of 4

AFP-L3%, AFP, DCP combination testing:

A Bibliography

AFP-L3%, AFP, DCP combination testing: A Bibliography Wako Diagnostics

Wako Diagnostics

(1) Clinical Evaluation of Lens culinaris Agglutinin-Reactive α-Fetoprotein and Des-γ-Carboxy Prothrombin in Histologically Proven Hepatocellular Carcinoma in the United States. Carr B, et al., Dig Dis Sci. 2007; 52:776-782. Summary Carr et al (2007) studied the DCP, AFP-L3% and AFP levels in 99 United States HCC patients who were histologically diagnosed with unresectable HCC. The cutoff values were: DCP (200 mAU/mL), AFP-L3% (10%), AFP (25 ng/mL). The sensitivity of AFP-L3%, DCP, and AFP was 61.6%, 72.7%, and 67.7%, respectively. The highest sensitivity of 85.9% was achieved when all three markers were used in concert. AFP-L3% level was significantly associated with portal vein invasion (p = 0.0059) and presented a strong statistical relationship to patient outcome. Metastasis of HCC was significantly associated with DCP levels (p = 0.0368) but DCP alone was not associated with the cumulative survival rate (p = 0.9581). The authors conclude that the combination of AFP-L3%, DCP, and AFP was demonstrated to be superior in the detection of HCC compared with each marker alone. In addition, since AFP-L3% was significant- ly related to portal vein invasion and patient outcome, AFP-L3% has the potential to be a prognostic marker for HCC.

(2) Clinical utility of Lens culinaris agglutinin-reactive fraction of alpha-fetoprotein (AFP-L3) and des-γ-carboxy prothrombin (DCP) alone or in combination as biomarkers for hepatocellular carcinoma (HCC). Sterling R, et al., Poster presentation at DDW, May 2007. Summary Sterling et al prospectively evaluated 332 patients with HCV cirrhosis, of whom 34 developed HCC. Of the 298 cirrhosis patients who did not develop HCC during the course of this study, 86 (29%) had AFP > 20 ng/mL. In this false positive population, the speci- ficity of L3% and DCP, respectively, was 87 and 91%, with the combination of L3% and DCP having a specificity of 80%. The authors conclude that the combination of elevated AFP-L3% and DCP has greater clinical utility than AFP alone in detecting HCC. “Impor- tantly, a negative AFP-L3% and DCP have high specificity and NPV in excluding HCC in those with an elevated AFP.”

(3) Risk factors for hepatocellular carcinoma may impair the performance of biomarkers: A comparison of AFP, DCP, and AFP-L3. Volk M, et al., Cancer Biomarkers. 2007; 3:79-87. Summary The aim of this case-controlled study was to compare the performance characteristics of the biomarkers AFP, DCP, and AFP-L3% in the diagnosis of HCC, and to determine the effect of risk factors for HCC on test performance. Out of 253 patients 84 were diagnosed with HCC (52 had early stage tumor) and 169 had cirrhosis without HCC. Cutoff values were determined by their ROC analysis:

DCP (150 mAU/mL), AFP-L3% (3%), and AFP (23 ng/mL). DCP was the best in differentiating early HCC from cirrhosis, with a sensitivity of 92% and specificity of 93%. In addition, DCP identified 15 (88%) of 17 patients with early stage HCC that would have been missed with total AFP measurement at 20 ng/mL cutoff. The combination of all three markers yielded slightly lower values for sensitivity (88%) and specificity (91%). When the patients were divided into two groups by their risk for HCC, all three markers had a lower sensitivity and area under ROC curve in the high-risk group compared to the low-risk group. The authors conclude that “the combination of markers had excellent performance but the improvement over DCP alone was relatively modest.” The authors further suggest that the affect of risk factors for HCC on the performance of biomarkers should be considered and included in the analysis of future studies.

(4) Prognostic significance of simultaneous measurement of three tumor markers in patients with hepatocellular carcinoma. Toyoda H, et al., Clinical Gastroenterology and Hepatology. 2006; 4:111-117. Summary

Toyoda et al (2006) studied the DCP, L3% and AFP levels in 685 HCC patients. 23% were negative for all 3 markers. 56% were AFP+,

33% were L3+ and 54% were DCP+.

Combination of L3 and DCP recognized 63%. L3% uniquely added only 14 patients (2.0%), DCP

added 110 (16.1%) and AFP added 96 (14%). Patients positive for DCP alone (size, portal vein thrombosis) or L3 alone (number of tu-

mors) differed from those patients having no markers. There was significant association of number of positive markers with signs of advanced HCC and with decreased longevity. This study shows that tumor markers AFP-L3 and DCP appear to represent different features of tumor progression in patients with HCC. The authors conclude “the number of tumor markers present could be useful for the evaluation of tumor progression, prediction of patient outcome and treatment efficacy.”

(5) A simultaneous monitoring of Lens culinaris agglutinin A-reactive alpha-fetoprotein and des-gamma-carboxy prothrombin as an early diagnosis of hepatocellular carcinoma in the follow-up of cirrhotic patients. Shimauchi Y, et al., Oncol Rep. 2000; 7(2):249-56. Summary Shimauchi et al (2000) surveyed 78 cirrhosis patients by biomarkers, US and CT. HCC was diagnosed in 21 of 78 (27%) during the follow-up period. Mean follow-up was 42 months. At the time of diagnosis, the positive rates were: AFP (>200 ng/mL) 14%, AFP- L3% (>10%) 33% and DCP (>40 AU/mL) 43%. Evidence that the HCC were detected “early” comes from the high percentage unifocal (81%) and small tumor size (67%<2.1 cm, 95% < 3.1 cm). Of the 21 HCC patients, 9 were DCP+, 7 were L3+ and 7 were negative for both. The sensitivity for both (i.e. at least one positive) was 67%. This study demonstrates that there is complementarity between the DCP and AFP-L3. The authors therefore conclude that, “the simultaneous determination of these tumor markers is necessary for the

follow-up of LC patients. When either of the two tumor markers is persistently positive through the follow-up period, it is necessary to observe the patient closely in expectation of the occurrence of HCC in the near future.”

(6) Clinicopathologic features of patients with primary malignant hepatic tumors seropositive for α-fetoprotein-L3 alone in comparison with other patients seropositive for α-fetoprotein-L3. Okuda H, et al., J Gastroenterology and Hepatology. 2005; 20:759-764. Summary Okuda et al (2005) studied a pool of 220 primary malignant hepatic tumor patients of whom 84 (38.2%) were L3+ before hepatectomy. Of the 84, 76 had HCC and the remaining 8, seven had carcinomas derived from cholangiocytes. They employed different criteria for AFP (>100 ng/mL) and L3 (>15%). Curiously, six of the 7 non-HCC liver carcinomas were in the group positive only for L3, causing the authors to claim that in patients with primary malignant hepatic tumors, doctors should measure L3. 53 of 84 (63%) were also DCP+. The authors conclude that patients with primary malignant hepatic tumors seropositive for AFP-L3 alone with low AFP concentrations have unique clinicopathologic features. Therefore, AFP-L3 levels should be measured at least once, even in those seronegative for both AFP and DCP.

(7) A combined proteomic and serologic approach to diagnosis of hepatocellular carcinoma. Zinkin T, et al., Hepatology. 2005; 42:239A. Summary Zinkin et al (2005) reported in a poster at AASLD’s TLM that a study of 39 HCC patients and 51 cirrhosis patients, DCP (cutoff 125 AU/mL) had a sensitivity of 84% and specificity of 69%. AFP (20 ng/mL) gave 74% sensitivity and 71% specificity. AFP-L3% was able to rule out 9 of 12 false positives by AFP in the cirrhosis group. There were no L3 specificity/sensitivity values in the abstract. The combination data are difficult to interpret since they included proteomic results into the combination of the 3 biomarkers; combined, the values were sensitivity (95%) and specificity (73%). The authors conclude that a combination of SELDI-TOF MS with AFP, AFP-L3 and DCP can significantly improve test sensitivity and that this combination measurement should be applied to a prospective cohort of at risk patients to see if this approach can detect preclinical HCC.

(8) Sterling R, et al., unpublished data Summary Wako DCP original Package Insert shows the study results of the full 387 prospectively evaluated patients who were HCC-free at the outset of the study. Of the 39 patients developing HCC, 28 (71.2%) were positive for one or both biomarkers. L3 alone detected 20 of 39 (51.3%) whereas DCP alone detected 19 of 39 (48.7%). Thus, only 9 were L3 positive and DCP negative (23.1%). Of the 262 not develop- ing HCC, 216 (82.4%) were negative for both biomarkers throughout the study. This is using the 7.5 ng/mL cutoff for DCP.

(9) Prediction of recurrence of hepatocellular carcinoma after curative ablation using three tumor markers. Tateishi R, et al., Hepatology. 2006; 44(6):1518-1527. Summary Tateishi et al (2006) studied a pool of 416 patients with HCC (80.8% were HCV-positive) who received percutaneous ablation as the initial treatment to elucidate the predictability of three tumor markers (AFP, AFP-L3, DCP) on HCC recurrence after curative ablation. Cutoff values were: AFP (100 ng/mL), AFP-L3% (15%), DCP (100 mAU/mL). Before ablation, the sensitivities for AFP, DCP, and AFP-L3 were 23.3%, 16.3%, and 13.7%, respectively. At least one marker was positive in 151 (36.3%) patients. After curative ablation, the value decreased below the cutoff level in 67 of 97 (69.1%), 61 of 68 (89.7%), and 46 of 57 (80.7%) patients for AFP, DCP, and AFP-L3, respective- ly. Tumor recurrence was identified in 277 patients. Univariate analysis revealed that platelet count, size and number of tumors, AFP, AFP-L3, and DCP preablation; AFP and AFP-L3 postablation were significant for recurrence. Multivariate analysis indicated that AFP >100 ng/mL and AFP-L3 >15%, both pre-and postablation, were significant predictors. The authors report that AFP-L3 positivity after ablation was the strongest predictor of recurrence. The presence of AFP-L3 postablation is a strong indication of residual cancer that cannot be detected with imaging techniques. The authors conclude that tumor markers pre- and post ablation were significant predic- tors for HCC recurrence and can complement imaging modalities in the evaluation of treatment efficacy.

(10) Sequential fluctuation pattern of serum des-γ-carboxy prothrombin levels detected by high-sensitive electrochemiluminescence system as an early predictive marker for hepatocellular carcinoma in patients with cirrhosis. Shimizu A, et al., International Journal of Molecular Medicine. 2002; 9:245-250. Summary Shimizu et al studied the usefulness of high-sensitive DCP, AFP, and AFP-L3% for the early diagnosis of HCC in 90 patients with viral cirrhosis. Out of 90 patients, 56 patients developed HCC and 34 patients did not develop HCC. Cut off values were: AFP (20 ng/mL), AFP-L3% (10%), DCP (40 mAU/mL). The combination of all three markers gave the highest accuracy overall with 82.1%, 82.4%, 82.2% for sensitivity, specificity, accuracy, respectively. Additionally, Shimizu et al. followed the serial changes of high-sensitive DCP, AFP and AFP-L3% levels in 90 patients over a period of two years. Authors found that the rate of HCC development in patients with fluctuat- ing DCP and AFP-L3% was significantly higher than that in patients who showed non-fluctuating type (93% vs. 56% for DCP and 91%

Wako Diagnostics

Wako Diagnostics

vs. 58% for AFP-L3%). The rate of HCC development was not significantly different for fluctuating and non-fluctuating AFP (69% vs. 59%). Nineteen of 75 patients with non-fluctuating DCP level who had progressively elevated DCP all developed HCC. Importantly,

  • 13 out of these 19 patients showed the elevation of DCP 3 months before the detection of HCC. Of 79 patients with non-fluctuating

AFP-L3%, 15 showed continuously elevated levels until HCC detection. The authors conclude that periodic measurement of the serum

DCP levels using high sensitive method is very useful for HCC screening in patients with viral cirrhosis.

(11) Clinical significance of Lens culinaris agglutinin-reactive fraction of serum α-fetoprotein in patients with hepatocellular carcinoma. Yoshida S, et al., International Journal of Oncology. 2002; 20:305-309. Summary Yoshida et al measured serum levels of AFP, AFP-L3%, AFP-L3 absolute value (AV), and DCP in 80 histologically diagnosed HCC patients. The cutoff values were: DCP (40 mAU/mL), AFP (10 ng/mL), and AFP-L3% (10%). The sensitivities for AFP-L3%, AFP, and DCP were 44%, 84%, and 50%, respectively. Only 5 out of 80 (6%) were negative for all three markers which suggests clinical usefulness of combination assay using all three markers. In 41 patients who were below the AFP-L3% cutoff, 28 patients had abnormal AFP level and 17 had abnormal DCP level. AFP-L3%, AFP-L3-AV and DCP all showed statistical significance (AFP-L3%: p=0.0089, AFP-L3-AV:

p=0.0067, DCP: p=0.0069) for predicting the grade of differentiation of HCC i.e. malignant potential for both solitary and multiple type HCCs. AFP-L3% had a significant correlation to predicting the size of solitary type HCC (p=0.0003). DCP showed a significant correla- tion (p=0.0011) for differentiating nodular and diffuse type HCC but AFP was the best (p<0.0001) in this category. The authors conclude that AFP, AFP-L3%, and DCP may reflect the different biological characteristics of HCC and that all three markers may be necessary for evaluating the clinical states of HCC.

(12) Clinical utility of simultaneous measurement of serum high-sensitivity of des-γ-carboxy prothrombin and Lens culinaris agglutinin A-reactive α-fetoprotein in patients with small hepatocellular carcinoma. Sassa T, et al., Eur J Gastroenterol & Hepatol. 1999; 11:1387-



Sassa et al. conducted a study to evaluate the diagnostic efficacy of simultaneous measurements of high-sensitive DCP (H-DCP) and AFP-L3%. Simultaneous measurements of serum DCP and AFP-L3% were made in 61 patients diagnosed (56 patients diagnosed histo- logically and 5 patients clinically diagnosed) with small HCCs (<or = 2cm in diameter) and 134 controls (chronic hepatitis: 59 cases; cir- rhosis: 75 cases). Cut off values were: AFP-L3% (10%) and DCP (40 mAU/mL). Out of 61 patients, 27 (44.3%) were positive for H-DCP and 14 (23.0%) were positive for AFP-L3%. Authors found no correlation between AFP-L3% and H-DCP. When measured separately,

  • 19 patients (31.1%) were positive for H-DCP alone and 6 patients were positive for AFP-L3%. Eight patients were positive for both

markers. When measured simultaneously, 33 of 61 patients (54.1%) of patients were positive for either marker. The specificity and ac- curacy were 97.8% and 84.1%, respectively. Positive and negative predictive values were 97.8% and 82.4%, respectively. In comparison, AFP alone yielded sensitivity of 8% with 100% positive and 70.5% negative predictive values. While there was no correlation of DCP to tumor size, type and total numbers AFP-L3% had significant correlation to multiple tumors (p=0.0316) and to differentiation of HCC

(p=0.0564). The authors conclude that simultaneous measurement of DCP and AFP-L3% improves the sensitivity for detection of HCC < 2cm in diameter and that this combination testing could be complementary and useful for the diagnosis of small HCC.


(1) Serum tumor markers for detection of hepatocellular carcinoma. Zhou L, et al., World J Gastroenterol. 2006 Feb; 12(8):1175-1181 This review lists and reviews serum HCC biomarkers and categorizes them as: 1. Oncofetal antigens and glycoprotein antigens, 2. en- zymes and isoenzymes, 3. genes, and 4. cytokines. All listed biomarkers are described in their clinical potential in clinical diagnosis and surveillance of patients who are at risk of developing HCC. The authors conclude that “…AFP, AFP-L3 and DCP are the most useful serum tumor markers for the detection of HCC, and simultaneous determination of these markers could improve the accuracy, espe- cially in differentiating HCC from nonmalignant hepatopathy.”

(2) Serum markers of hepatocellular carcinoma. Spangenberg H.C., et al., Seminars in Liver Disease. 2006; 26(4):385-390. Spangenberg et al reviewed and evaluated serum markers for early detection of HCC. Studies show that AFP-L3 is demonstrated to be associated with reduced liver function, poorer differentiation, and biologically malignant characteristics. Therefore, authors state that “AFP-L3 appears to be a more reliable and better HCC marker than total AFP and an excellent prognostic parameter in patients with HCC.” Since there is no correlation between AFP and DCP, DCP is more specific for HCC. Additionally, DCP correlates with the HCC stage as well as survival. There are a few studies that demonstrated increased specificity and sensitivity by combining both AFP and DCP. In summary, the authors conclude that “the determination of AFP, AFP-L3, and DCP levels are excellent to monitor tumor progression as well as the efficacy of treatment in marker-positive patients.”

(3) Serological markers of liver cancer. Yuen M-F, Lai C-L, Clinical Gastroenterology. 2005 Feb; 19(1):91-99. AFP-L3 correlates with the staging of HCC and overall is superior to the total AFP in the accuracy of diagnosing HCC, and in the cor- relation with the stage and the prognosis of the disease. Since DCP is independent of AFP and is rarely observed in other liver diseases it is more specific for HCC than total AFP and AFP-L3. In addition, DCP correlate with the stages of the HCC as well as survival. In the “practice points” of conclusion, the authors state that “combination of two or three markers increases the sensitivity and diagnostic accuracy for hepatocellular carcinoma”.

(4) National Academy of Clinical Biochemistry Guidelines for the Use of Tumor Markers in Primary Liver Cancer (section 3D of NACB:

Practice Guidelines and Recommendations for Use of Tumor Markers in the Clinic Liver Cancer). Lamerz R, et al., Published in 2006 as part of the NACB’s Laboratory Medicine Practice Guidelines (LMPG). This monograph reviews consensus recommendations of the National Academy of Clinical Biochemistry (NACB) in the United States and the European Group on Tumor Markers (EGTM) in Europe and focuses on the appropriate use of tumor markers for specific can- cers. The guidelines describe clinical utility of DCP and AFP-L3. It lists DCP as a “promising emerging marker of considerable poten- tial”. Though the NACB does not recommend any HCC-related biomarkers except AFP for the routine surveillance of patients with or at risk of HCC, it supports further evaluation. It also has a comprehensive list of biomarkers and their potential utility and status.

(5) The prognostic significance of clinical and pathological features in hepatocellular carcinoma. Qin L-X, Tang Z-Y, World J Gastroen- terol. 2002; 8(2):193-199. AFP-L3 was found to be useful indicator of distant metastasis and a poor prognosis for HCC. Patients with AFP-L3 positive had worse liver function and larger tumors compared to the negative group. Often, HCC was more advanced with poor tumor histology and distant metastasis was diagnosed significantly more often in AFP-L3% positive group. DCP is not an independent prognostic factor. However, DCP measurement was effective in predicting HCC recurrence and had the advantage of pre-surgical assessment.