HCC is one of the leading causes of cancer-related mortality. Despite strict selection criteria, recurrence occurs in 6%-18% transplanted HCC patients (15). The main reason for this is the presence of residual micrometastases formed even before transplantation, or the spread of tumour cells from the original tumour into the bloodstream during surgical manipulation (16). Sometimes patients experience extrahepatic metastases even when no primary lesion is found after surgical resection or liver transplantation. In addition, extrahepatic metastases may occur after locoregional therapies for early stage HCC (17). Serum markers for both early diagnosis of patients at high risk for HCC and early detection of recurrence after transplantation are of great importance as they offer the opportunity to decrease patient mortality and reduce medical costs. Although the role of PIVKA-II has been widely studied in the diagnosis and prognosis of HCC, as well as in the evaluation of surgical treatments such as TACE, there are hardly any studies in the literature that have explored the role of PIVKA-II in the follow-up of patients after LT for HCC.
Regarding the different underlying aetiologies of liver disease in the patients included in our study, HCV is the predominant aetiology, accounting for 34.8% of patients. Overall, taking into account also the combined aetiologies, more than half of the patients had HCV, with this aetiology accounting for 56.5% of the cases. Our results are similar to those of other studies in HCC patients in which HCV is also the predominant aetiology (18–20). Unlike hepatitis B infection, there is no vaccine for HCV (21); however, the prevalence of both factors is decreasing due to vaccination of newborns against HBV and the existence of effective treatments for carriers of HBV and HCV (22). Although some studies have suggested that PIVKA-II and AFP levels are associated with the aetiology of liver disease (23, 24), in our study no significant association between these two biomarkers and the aetiologies of HCC was found, making the levels of these biomarkers independent of the aetiology of the underlying liver disease, in agreement with the findings of Sharman et al (25).
When studying the association of pre-transplant PIVKA-II levels with Child-Pugh classification, in our study no statistically significant differences between the different groups were found, so that PIVKA-II levels are independent of the Child-Pugh class to which the patient belongs. Our results are in concordance with those of Saito et al. who also found no significant association between serum PIVKA-II level and Child-Pugh score before treatment with TACE (26). In contrast, we did find significant differences in serum AFP levels between the different groups of Child-Pugh patients, with Child-Pugh class C patients having significantly higher AFP levels than those in classes A and B, in agreement with previous studies (27). This finding confirms the ability of AFP to reflect the severity of liver dysfunction in patients with HCC.
After evaluating the association between PIVKA-II and the different clinicopathological parameters of the patients in our study, it was found that patients with larger tumor size had higher levels of PIVKA-II, which suggests that the serum concentration of this biomarker may play an important role in predicting disease severity, such that a higher PIVKA-II concentration may indicate larger tumour volume and worse clinical stage. This correlation has also been observed in other studies in patients diagnosed with HCC, but not on the OLT waiting list (28–31). AFP levels were also significantly associated with larger tumour size, in a way that patients with tumour size > 3 cm had significantly higher AFP levels than patients with tumour size ≤ 3 cm. Our findings are closely related to those of Si et al., who found that serum levels of AFP and PIVKA-II in patients diagnosed with HCC due to HBV, but not selected for OLT, were significantly associated with tumor size (32). However, in our study, neither serum PIVKA-II nor AFP levels correlated with the number of tumours, in line with other studies such as that of Feng et al. who found no correlation between serum PIVKA-II levels and the number of tumours (30), or Lapinski et al. who also found no significant association between serum AFP levels and the number of tumour nodules (33).
Most of the patients enrolled in the study underwent at least one TACE before liver transplantation. TACE is an effective treatment in patients with unresectable HCC, however most patients treated with TACE will need repeat therapy due to partial response or tumor recurrence. Treatment outcomes after TACE depend on both the severity of underlying liver dysfunction and tumor burden (34). In our study, pre-transplant PIVKA-II levels correlated with the number of TACEs performed prior to pre-transplant sampling. This result suggests that those patients who received multiple treatments with the intention of reducing the tumor to carry out the transplant had more advanced HCC and a worse clinical stage. Today, the benefits of repeating this treatment are unclear, since on the one hand it does reduce the size of the tumour, but at the same time it could favour tumour spread and local inflammation. Li et al. showed that residual liver cancer cells and normal liver tissue undergo changes in gene expression after TACE, so that the possibility of recurrence and metastasis of residual HCC cells increases after TACE (35). Furthermore, remnants of normal liver tissue can lead to HCC recurrence, due to cirrhosis and increased compensatory hyperplasia and proliferative activity after TACE. In addition, several previous studies have shown that the establishment of collateral circulation after TACE is an important factor for HCC recurrence and metastasis, which can lead to tumour cell growth with high metastatic potential (36). Serum AFP levels, in contrast, did not correlate significantly with the number of pre-transplant TACES.
After studying the association between serum PIVKA-II and AFP levels of the patients included in the study, they did not correlate with each other, in agreement with other studies in which, even dealing with HCC patients, they were not candidates for OLT (37–39). The found results reflect that these two biomarkers are independent of each other, which could be explained by the different synthesis pathways of the two markers in hepatoma cells (40). Both biomarkers may reflect the tumour burden of HCC patients (41), and may be complementary markers in terms of their clinical utility (42).
Ideally, tumour marker levels should fall within a reference range after effective treatment, and rise before imaging studies detect tumour relapse (37). The exact mechanism by which PIVKA-II is produced by the tumour remains unclear, but the normalisation of PIVKA-II levels after curative treatment of HCC, such as liver resection, clearly indicates that the tumour is the source of its production (43). In our study, we observed that both PIVKA-II and AFP decreased significantly between pre-transplant and post-transplant values, in agreement with previous studies, such as that of Feng et al., in which serum levels of PIVKA-II and AFP decreased significantly after liver resection (30). Specifically, in the patients in our study, PIVKA-II shows a statistically significant decrease at 1 month, 6 months and 1 year post-transplantation. In contrast, at 2 years and 3 years post-transplant, we did not find a significant decrease in the levels of this marker. It is important to note that PIVKA-II levels may be elevated in certain situations other than HCC, for example in cases of biliary obstruction (11). Also the level of PIVKA-II can be strongly influenced by drugs such as rifampicin, vitamin K deficiency (e.g. due to malnutrition in cirrhotic patients) and antivitamin K drugs (e.g. anticoagulants such as acenocoumarol and warfarin) (43). However, PIVKA-II elevations in patients in our study who did not have post-transplant tumour recurrence were not associated with any specific cause, as none of these patients were found to be on treatment with the aforementioned drugs or to have vitamin K deficiency or obstructive jaundice. AFP levels were significantly lower at one month, one year and two years post-transplantation. Considering that tumour growth patterns are highly variable among individuals, there is probably no single perfect biomarker for HCC monitoring after transplantation; therefore, the combination of biomarkers might be more informative than any single biomarker alone (44).
Patient 13 was found to have recurrence 6 months after transplantation. PIVKA-II levels in this patient increased at 1 month after OLT, with respect to pretransplant levels, and decreased at 6 months post-transplant, coinciding approximately with the time at which the recurrence was detected by imaging. This patient probably developed tumor recurrence around one month after OLT, so that PIVKA-II would be indicating post-transplant recurrence before imaging tests, such as computed tomography (CT) or magnetic resonance imaging (MRI). However, these levels decreased again at 6 months after transplantation, when HCC recurrence and extrahepatic metastases were detected by imaging, so we cannot affirm that the elevation of PIVKA-II at one month post-transplant was caused by tumor recurrence, even though other causes of PIVKA elevation, already mentioned above, were ruled out, such as the use of anticoagulant antivitamin K drugs. In patients 7 and 19 in our study, the evolution of PIVKA II levels after OLT follows an ideal pattern, in which these levels decreased at 1 month post-transplant with respect to pre-transplant values, reflecting that the diseased liver of HCC patients is the source of production of this abnormal prothrombin molecule, and increased considerably at 6 months post-transplant, coinciding approximately with the detection of tumor recurrence by imaging in both patients. In the case of patient 23, PIVKA-II levels increased abruptly, even before the detection of tumour recurrence by imaging, while AFP levels remained within the reference range. This is in line with the study by Nanashima et al., who showed that changes in PIVKA-II levels after treatment, including hepatectomy and ablation therapy, tended to be more reflective of tumour recurrence than changes in AFP level (45). Therefore, according to our findings, and despite its limitations, PIVKA-II could be a good marker in the study of progression after liver transplantation, complementing AFP.
After evaluating the predictive ability of pre-transplant AFP and PIVKA-II levels to predict post-transplant recurrence, we found that PIVKA-II was a better predictor of recurrence than AFP, with a PIVKA-II AUC of 0.613 versus AFP AUC of 0.544. Our results are consistent with previous studies (18, 46–48), in which PIVKA-II was also a better predictor of post-transplant recurrence than AFP. This is biologically feasible because PIVKA-II binds to vascular endothelial growth factor receptor 2 and induces autophosphorylation of the receptor and downstream effectors, including phospholipase C-Y and mitogen-activated protein kinase, promoting endothelial cell proliferation and migration. However, AFP is a product of tumour cells and has no biological effect on promoting tumour growth (49). Combining pre-transplant levels of PIVKA-II and AFP for prediction of post-transplant recurrence increased the AUC to 0.742 with increased sensitivity (100%) but decreased specificity (45.2%). Our results are in partial agreement with other authors such as Chon et al. in whose study sensitivity, but also specificity, were simultaneously improved when both AFP and PIVKA-II levels were measured (sensitivity, 66.7%; specificity, 47.9%) compared to those obtained when only AFP levels were measured (sensitivity 60.1% and specificity 45.2%) or PIVKA-II levels alone (sensitivity 62.9% and specificity 47.9%) (50). Lee et al. also demonstrated that the combination of AFP and PIVKA-II was a better predictor of recurrence after transplantation in advanced HCC than either marker alone (51). Thus, this combination of markers could improve the ability of each marker alone to identify patients most likely to have post-transplant recurrence, although the results were also not statistically significant (p > 0.05). The number of patients was a limitation of these studies due to two factors: on the one hand, patients must meet a series of strict requirements to be on the waiting list and be candidates for transplantation, and in addition, there must be a donor with an organ compatible with the recipient. The other limitation is the moderate post-transplant recurrence rate of 6–18%, both of which contribute to severely limit the cases under study. In this context, the results of this study may be useful for the development of new lines of research that provide new evidence for PIVKA-II to be incorporated as a useful biomarker in the management of HCC in the clinical laboratory, as occurs in countries such as Japan, where together with ultrasound and lens culinaris agglutinin-reactive fraction of AFP (AFP-L3), it is used in the screening of at-risk patients every 6 months (52).