Correlation of Serum Alpha-Fetoprotein (AFP) Levels with the size of Hepatocellular carcinoma on Triphasic CT scan: A study in patients with the heterotrophic viral infection

Objective: To determine the association between serum Alpha-Fetoprotein (AFP) levels with different variables including Hepatocellular tumor size. Materials and Methods: This cross-sectional study was conducted in the Center of Liver and Digestive Diseases, Holy Family Hospital, Rawalpindi. Three hundred and thirty-four liver cirrhosis patients complicated with HCC were participants. Different variables such as diabetes, hypertension, ischemic heart disease, smoking, jaundice, fever, lethargy, weight loss, anorexia, pain right hypochondrium, abdominal distention, ascites, encephalopathy, biochemical tests (Bilirubin levels, serum albumin levels, prothrombin time(PT) (<4 /4-6 / >6 sec prolonged), child Turcotte class ( A, Band C), PST(performance status), Barcelona Clinic liver cancer staging system (0/A/B/C/D) and tumor size in 3 Groups (Group I size <3 cm, Group II size 3-5cm, and Group size III >5cm size) were compared with serum Alpha Feto Proteins (AFP) levels( Group I AFP level >20 and <200ng/mL, Group II AFP level 200-400 ng/mL, while Group III AFP level was >400ng/mL). Data were analyzed through SPSS Statistics version 23.0 (IBM, Armonk, NY), Spearman’s rank is used for P-value. Results: The AFP (Group-I/II/III )association with diabetes (P 0.121), HTN (P 0.811), IHD (P 0.546), HBsAg positive patients (all P 0.186), HCV positive patients (all P 0.131), pain right hypochondrium (P 0.599), fever (P 0.065), lethargy (P 0.388) and Encephalopathy (all P 0.075) were identified as non-significant association while biochemical levels such as Bilirubin (P 0.000), Albumin (P 0.000), Prothrombin Time (P 0.038) creatinine ( all P 0.001) , Child Turcotte Class A, B and C (P 0.000),smoking (P 0.023), weight Loss (P 0.002), Anorexia (P 0.007), Jaundice (P 0.012), Ascites (P value 0.000), ECOG (P 0.000), Barcelona Clinic for liver cancer (BCLC) staging system (P 0.000), lesion I (P 0.000) and lesions II (P 0.006) on CECT had significant association. Conclusion: We hereby, conclude that there is a noteworthy correlation between Alpha Feto Proteins with different variables including an increase in the size of the tumor. Hence Alfa Feto protein is not only helpful for the diagnosis of HCC but also has a pivotal role in predicting tumor burden and disease level.


Introduction
Today, Hepatocellular carcinoma (HCC) is a pressing health issue for the globe, due to late diagnosis, high morbidity, and mortality rate. Hepatocellular carcinoma (HCC) is the sixth most prevalent cancer worldwide and is associated with liver primary malignancy. It is one of the most frequent causes of death that is associated with cancer. Cases of HCC are constantly rising worldwide. [1][2][3] Multiple risk factors are identified for HCC such as chronic hepatitis B& C, non-alcoholic fatty liver disease, and alcohol consumption are the most common while the less common factors are hereditary hemochromatosis, primary biliary cirrhosis, autoimmune hepatitis, Wilson's disease, and alpha1-antitrypsin deficiency. In Pakistan, the prevalence of Hepatocellular carcinoma malignant tumors is 3.7%-16% while the common causes are viral hepatitis B and C-related cirrhosis. 4 The burden of disease and hospital admissions of this disease is also continuously increasing. Most HCC patients have been diagnosed at a late stage limiting the cure IJRMC Vol 1 (Issue 1) Khan, J., et al. 37 options. Delayed diagnosis is associated with disease being either asymptomatic initially or present with vague and non-specific symptoms. Patients are presented with different types of symptoms that are related to chronic liver diseases such as skin and eyes yellowish discoloration, right upper abdominal pain, abdomen swelling, weakness, fever, and weight loss. 5,6 In recent decades, more asymptomatic patients are diagnosed due to active monitoring and awareness of HCC in high-risk patients, particularly in cirrhosis. Hepatocellular carcinoma diagnostic tools such as blood alpha-fetoprotein (AFP) and imaging studies are being used for screening after every six months in cirrhotic patients. In patients with chronic liver disease (CLD) increased AFP level is considered a risk factor for HCC which is helpful for chronic liver disease subgroup identification. In this regard, Alpha-fetoprotein (AFP) level fluctuation reflects acute on chronic hepatic injury, liver disease progression, or HCC. 7,8 People with serum AFP levels > 20 ng/mL should be monitored because of their ability to predict liver cancer. Higher serum AFP level is linked with tumor size, bilobar involvement, portal vein tumor thrombus, and massive or diffuse tumor types. 9,10 In addition, a high level of AFP is linked to tumor recurrence and metastasis. 11,12 The Definitive Hepatocellular carcinoma diagnostic tool is an image-based diagnosis such as contrastenhanced Computed Tomography (CT) based on the arterial enhancement of the tumor followed by washout in venous and delayed phases. [13][14][15][16] Our study's main aim was to check the association of liver size with serum AFP protein levels, which may be helpful in the future for HCC diagnosis. Pakistan is a developing country, hence, offering expensive and advanced tools for the diagnosis of cancer is beyond our capacity.

Materials & Methods
Ethical approval of this study was taken from the Rawalpindi Medical University institutional board. A cross-sectional study was conducted at the Center of Liver and Digestive Diseases, Holy Family Hospital, Rawalpindi, which is a tertiary care facility. Subjects with known chronic liver disease secondary to chronic hepatitis B and Hepatitis C complicated by Hepatocellular Carcinoma patients were included. This study was started in January 2017 and was completed in January 2022. Demographic characteristics, Hepatitis B or C infection, size, and the number of Hepatocellular carcinoma lesions detected on ultrasound and confirmed by Triphasic CT scan were compared with serum Alpha Feto Proteins (AFP). Based on Alfa fetoprotein level, we divided patients into 3 groups; Group I patient's AFP levels were >20ng/ml but <200ng/mL, Group II AFP levels were 200-400 ng/mL, while Group III AFP levels were >400ng/mL. While based on tumor size we divided them into three Groups (such as <3 cm, 3-5cm, and >5cm size). In cases of multiple tumor nodules, the size of the largest lesion was considered. 17 Different signs and symptoms were recorded such as associated co-morbidities like diabetes, hypertension, ischemic heart disease, smoking, and symptoms at presentation (jaundice, fever, lethargy, weight loss, anorexia, pain right hypochondrium, and abdominal distention), presence of ascites and encephalopathy. In addition, biochemical tests such as serum bilirubin levels, serum albumin levels, prothrombin time(PT) (<4 /4-6 / >6 sec prolonged), child Turcotte class ( A, B, and C), PST (performance status), Barcelona Clinic liver cancer staging system (0/A/B/C/D), presence or absence of lymph nodal involvement and vascular metastasis was noted.
Data were analyzed through SPSS Statistics version 23.0 (IBM, Armonk, NY). Association of serum AFP levels with different variables was analyzed, Spearman's rank correlation is used for P-value, and a value <0.05 was considered a significant outcome.

Results
Baseline patients' characteristics: The patients sample size was 334, all patients were confirmed cases of Hepatocellular carcinoma. Among these patients, 214(64%) were males while 120 (36%) were females. On the basis of age groups; 1(0.2%) were in 1-10 years age group, 2(0.5%) were in 11-20 and 21-  Based on Alpha-fetoprotein (AFP) level we divided into 3 groups; Group I AFP levels were >20 but < 200 ng/mL, Group II AFP levels were 200-400 ng/mL while Group III AFP levels were > 400 ng/mL. In Group I men's participants were 23.6% while women were 14.1%, in group II men's participants were 15.6%, while women were 8.9%, in group III men's participants were 24.8% while women's proportion was 13%. The proportions of both sexes were not significant.
We analyzed biochemical test according to lowerhigh level; in all groups bilirubin < 2 mg/dL, 2-3 mg/dL and > 3 mg/dL (P 0.000), Albumin <2.8mg/dL, 2.8-3.5 g/dL and >3.5g/dL (P 0.000), Prothrombin Time < 4sec ,4-6 sec and >6 sec prolonged (P 0.038) child turcotte Class A, B and C (P 0.000), creatinine <1.2 mg/dL and1.2-3.0 mg/dL( all P 0.001) compared with AFP groups were identified as significant association .Other characteristics such as smoking (P 0.023), Weight Loss (P 0.002), Anorexia (P 0.007), Jaundice (P 0.012), abdominal Mass (P 0.000), Ascites (P value 0.000) were also identified as significant association with AFP level mentioned in Table: 1. Association between AFP levels with ECOG, BCLC, lesion number, and size: The association of AFP levels with ECOG Performance Status Scale Grade 0, 1, 2, 3, 4 (P 0.000) and Barcelona Clinic liver cancer (BCLC) staging system (P 0.000) was highly significant. The association of AFP levels with lesion size and the number of nodules was analyzed via ultrasonography and contrast-enhanced computed tomography (Triphasic), a total of 334 patients were included in both techniques. On ultrasonography lesion, I/II/III/IV in all groups (P 0.784), on CECT I/II/III/IV in all groups (P 0.767) revealed no significant association. We divided lesion size into 3 groups such as < 3 cm, 3-5 cm, and > 5 cm, lesion I (P 0.000) on USG was significant while lesion II (P 0.356) was not significant. On CECT lesions I (P 0.000) and lesions II (P 0.006) had a highly significant association with AFP mentioned in Table 2    Khan, J., et al. 41

Discussion
HCC being the leading primary hepatic malignancy is the 4 th most fatal tumor worldwide. 18,19 Although the etiology of HCC varies globally the increasing burden of Hepatitis C in Asian countries is becoming the leading cause of this tumor. Pakistan the 2 nd most prevalent country as far as Hepatitis C is concerned is putting on a huge health burden due to HCC [20]. Our study also suggests a significant contribution of HCV (109 patients) as compared to HBV (19 patients) in our part of the world which is contrary to the global association where still HBV leads as a causative agent. 21 One likely explanation for this disparity can be the major mode of transmission of HBV in our part of the world. As globally it is the vertical transmission of HBV that not only increases the risk of chronicity of the disease as well as the chances of developing HCC, but in Pakistan, the major likely mode of transmission can still be horizontal. 22 We also don't find any significant difference between AFP levels or the size of the tumor between these two viral etiologies. Spiros P Hiotis et al. in a study had shown much higher AFP levels (median level of AFP: HBV-1000 ng/ml vs. HCV-37 ng/ml; p = 0.002) as well as tumor size (HBV-78% >5 cm vs HCV-28 % >5 cm; p < 0.001) amongst patients with HBV as an etiology as compared to HCV. 23 AFP is a tumor marker of HCC, being widely in use for several decades. Multiple studies have been conducted in the past to establish its correlation with Tumor size, number, and disease burden. Our study shows a significant association between AFP levels and Tumor size on CECT of not only the 1st lesion (p = 0.000) but also with the 2nd lesion (p = 0.006) if identified on a CT scan. A similar significant association can be seen with the size of 1st lesion on ultrasound and AFP (p = 0.000000) however the significance is not well established for the 2nd lesion on ultrasound ( P = 0.356). S. Sharieff et al. 24 Khawar Shabbir et al 25  A study was conducted by Laura et al 26 , there is a weak but insignificant relationship between tumor size and HCC and the author calls this an accidental association which is contrary to our study and several other studies 27,28 where it seems to be a pathologic correlation between AFP and tumor size. AFP is a 591 amino acid glycoprotein produced by the fetal yolk sac, liver, and intestine and is associated with several tumors including HCC. A larger size of tumor means more cells producing the glycoprotein and hence a positive correlation is well anticipated. 29 There have been additional statistically significant findings with AFP levels. There was a statistically significant association of AFP groups with various child Turcotte Pugh classes as well as with bilirubin, albumin, and prothrombin time. Of note, all these abnormalities serve as prognostic indicators for liver dysfunction in various scoring systems like the Model for End-Stage Liver Disease (MELD) Score, etc.
So, a positive association of AFP groups likely indicates worsening liver disease that is expected with the increasing size of the tumor. This appears in synchronization with our hypothesis that higher AFP levels are an indicator of worsening disease burden. This is further evident by the significant association of AFP levels with ECOG Performance Status Scale grades and Barcelona Clinic for liver cancer (BCLC) staging. These evaluations are used to assess stages as well as the impact of HCC relating to the quality of life of the patients before making treatment decisions. So, a positive association of AFP groups again shows that rising AFP levels are an indicator of worsening oncologic burden including tumor size.
Certain limitations of our study must be acknowledged when we consider its results in implications. Due to the cross-sectional study design, temporality and residual confounding could be considered study limitations. Additionally, it is a single-center study with small data set so generalizability may be limited. The strength of our study is that the database is from a tertiary care referral center so the results can be representative of a larger population across various sociodemographic groups. The study does raise the need for further studies in the area to further establish the link in more rigorous clinical trials.

Conclusion
Alfa Feto Protein is a well-recognized tumor marker for Hepatocellular Carcinoma. Based on our study findings, we hereby, conclude that there is a noteworthy correlation between Alpha Feto Proteins and the increase in the size of the tumor. Hence Alfa Feto protein is not only helpful for the diagnosis of HCC but also has a pivotal role in predicting tumor burden and disease level. Though more tumor markers including PIVKA-II have been introduced in the market for the diagnosis of Hepatocellular Carcinoma, Alfa Feto protein, being the old one, still is being extensively used throughout the world as a diagnostic and prognostic indicator of stage progression of the disease and survival.