Page 8 - Read Online

Tong et al. Hepatoma Res 2019;5:36 I http://dx.doi.org/10.20517/2394-5079.2019.005 Page 3 of 13

HCC treatments, and current status. Amongst the 357 patients, 24 individuals were excluded from this study
due to diffuse appearing tumor in which the size could not be determined (22 patients) or due to an HCC
diagnosis made within six months of final patient entry into the database (2 patients). Of the remaining 333
patients, 169 who began HCC treatment prior to a second tumor size measurement were also excluded. The
remaining 164 patients had two consecutive imaging studies prior to HCC treatment and are the subjects in
the present study. HCC lesions were detected via surveillance in 113 patients with alpha-fetoprotein (AFP)
testing and US scans. The remaining 51 patients were either diagnosed by their referring physicians or during
their first visit to our Liver Center. The number and size of lesions, as reported by CT scan or MRI, were
recorded. The diagnostic criteria for cirrhosis were by imaging findings of a nodular surface, platlet count <
3
140,000 mm , presence of esophageal varices or ascites, or by liver biopsy.
The TGR was determined for all 164 patients. The diagnosis of HCC by MRI or CT scan were according
to AASLD criteria from their 2005 and 2011 recommendations [12,13] . Prior to that time period, imaging
criterion for HCC diagnosis relied on findings of a hypervascular lesion, elevated levels of AFP, tumor
growth on subsequent imagin, and biopsy of the lesion if the above criteria were not clear. The dates and
corresponding tumors sizes from the first and second imaging studies (CT or MRI) were recorded. The time
interval between the first and second images were ≥ 30 days (median time 92 days), and all were prior to any
[5]
treatments for HCC .

Baseline laboratory tests
Baseline laboratory tests were obtained from all patients. These included platelet counts, serum albumin,
total bilirubin, alkaline phosphatase, aspartate aminotransferase, alanine aminotransferase, and AFP. For
HBV patients, virus genotype, HBV DNA levels, precore mutation, basal core promoter mutation, and
HBeAg were recorded. For HCV patients, virus genotype and HCV RNA levels were recorded. Sera from
patients whose HCC was diagnosed prior to 1991 were retrospectively tested for anti‐HCV antibodies and
HCV RNA.

HCC treatments
Of the 164 patients followed in this study, 113 received definitive treatments, 7 received chemotherapy,
and 44 were offered supportive care. HCC patients were referred to academic centers for surgical and/or
locoregional therapies. Treatment options included orthotopic liver transplantation (OLT), surgical resection,
RFA, transcatheter arterial chemoembolization (TACE), or percutaneous ethanol injection (PEI). If a
patient had multiple treatments, they were assigned to the most definitive treatment category. OLT, surgical
resection, and RFA were considered to be the most definitive treatments. Patients who did not receive one of
the above treatments were given chemotherapy or supportive care.

Post-treatment outcomes
Patients who returned for regular follow-up care were continuously screened with imaging studies and
laboratory tests. In order to calculate dates of recurrence-free survival, dates of diagnosis, initial treatment,
recurrence, and latest follow-up or death were recorded.

Statistical analyses
Tumor growth rate calculation
The TGR was calculated using Schwartz’s equation: TGR = log(V/V )/(T - T ) where T - T indicates the
0
0
0
2
time interval between the two measurements and V and V represent the tumor volumes (V = 4/3pR ) at the
0
[14]
two measurement points . The Schwartz equation assumes early, exponential stage growth with the TGR
reported in % per month. In the analyses below, log TGR is used since log TGR has a distribution closer to
10
the normal distribution.