Prospective longitudinal quality of life and survival outcomes in patients with advanced infiltrative hepatocellular carcinoma and portal vein thrombosis treated with Yttrium-90 radioembolization

Background To determine the effect of Yttrium-90 (Y90) radioembolization on health-related quality of life (HRQOL) and its effect on overall survival advanced, unresectable infiltrative hepatocellular carcinoma (HCC) patients with concurrent portal vein thrombosis (PVT). Methods Consecutive patients with unresectable infiltrative HCC and PVT were recruited. The Short-Form 36 (SF-36) questionnaire was used to assess HRQOL for consecutive patients treated with glass-based Y90 based on a prospective phase II trial. MR imaging was used to determine tumor progression every 3 months post-treatment. Overall survival (OS) from treatment and time to progression (TTP) was analyzed using Kaplan-Meier estimation and log-rank test. Results Thirty patients were treated and followed for 17.4 months; physical and mental component summary scores (PCS & MCS) remained unchanged at one, three, and six months. While no difference was observed in baseline SF-36 scores for patients with prolonged TTP (≥4 months) and OS (≥ 6 months), corresponding 1-month PCS were significantly higher than those with TTP < 4 months and OS < 6 months. At 1 month, patients with normalized Physical Function (PF), Role Physical (RP) and PCS within 2 standard deviations (SD) of US normalized baseline scores had a significantly prolonged median OS (15.7 vs. 3.7 months; p < 0.001) and TTP (12.4 vs. 1.8 mo; p < 0.001) compared those with physical component scores greater than 2SD below normalized US population values. Conclusion Y90 radioembolization for HCC demonstrated long-term preservation of HRQOL. Lower baseline HRQOL scores were predictive of poorer OS. Early (1 month post-treatment) significant decreases in PCS were independent predictors of poorer OS and TTP. Trial registration ClinicalTrials.gov identifier NCT01556282, registered March 16, 2012.


Background
Worldwide, hepatocellular carcinoma (HCC) ranks as the sixth most prevalent cancer and causes the third most cancer-related deaths [1][2][3]. At diagnosis, approximately 10 to 40% of HCC patients have portal vein thrombosis concurrently [4,5]. These patients often have very poor prognosis, and are not eligible for curative surgery via resection or transplantation. Without treatment, the reported median overall survival (OS) for Barcelona Clinic Liver Cancer (BCLC) Stage C patients range from 2 months to 4.8 months, as compared to 10 to 24 months in HCC patients without PVT [5][6][7][8][9].
Due to the risk of hepatic artery infarction, embolic intra-arterial therapies such as transarterial chemoembolization (TACE) have previously been deemed as contraindications in HCC with PVT [10,11]. However, more recent studies establishing the safety and efficacy of conventional TACE (cTACE) in prolonging survival for HCC patients with PVT have been conducted [12,13]. Nevertheless, increased survival benefits have not been specifically established in infiltrative HCC patients [14]. In long-term studies, selective intra-arterial yttrium-90 ( 90 Y) radioembolization has demonstrated equal or better overall survival (OS) and time to progression (TTP) vs. cTACE [15], with comparable safety and efficacy in patients with infiltrative HCC with PVT [16]. Given the lowered risk of hepatic ischemia and decreased embolic effect with 90 Y radioembolization compared to cTACE, some authors have noted that it may be a preferred therapeutic option for patients with concurrent PVT [10,12,13].
In palliative oncology, health-related quality of life (HRQOL) has become an increasingly relevant measure of patient prognosis. In recent years, a progressively higher number of investigators have suggested using HRQOL as an independent prognostic factor for response to treatment and progression of disease in patients with advanced HCC [17,18]. As a widely-used tool in the evaluation of HRQOL, the Short-Form 36 (SF-36) contains 36 items/8 domains, including physical and social well-being, vitality, and pain [19]. In HCC patients, significant improvements in mental health at four months post-therapy [20] have been reported using the SF-36 for patients treated with cTACE, while significant increases in HRQOL with 90 Y radioembolization have been observed at one month following therapy [21]. However, to date no studies have demonstrated both survival outcomes of 90 Y therapy and its effect on HRQOL in unresectable, infiltrative HCC with PVT in particular. Our purpose was to investigate the effect of 90 Y radioembolization on HRQOL in patients with infiltrative, unresectable HCC with PVT, and its effect on overall survival (OS) and time to progression (TTP). We hypothesized that patients with higher HRQOL scores prior to therapy would demonstrate more favorable survival outcomes, with maintenance of HRQOL posttherapy.

Study design and patient selection
In an Institutional Review Board (IRB) approved correlative study based on a prospective phase II trial (trial registration at ClinicalTrials.gov; identifier NCT01556282, registered March 16, 2012), consecutive patients with infiltrative HCC with PVT were enrolled for possible therapy with glass-based 90 Y radioembolization. Diagnosis of HCC was made according to American Association for Study of the Liver Diseases (AASLD) using dynamic contrast-enhanced magnetic resonance imaging (MRI). All patients in this study had PVT due to tumor invasion. As previously described, infiltrative HCC was defined as a geographic region with high T2 signal, arterial enhancement and early washout on T1 gadolinium-enhanced images [22]. A multidisciplinary liver cancer tumor board evaluated every patient for his/her suitably to undergo 90 Y radioembolization.
Pre-treatment evaluation, planning, inclusion/exclusion criteria, and 90 Y therapy As previously described by our group, patients who were ≥18 years with Eastern Cooperative Oncology Group (ECOG) performance score (PS) ≤2 and life expectancy of at least 3 months were included. Additionally, those with tumor burden <75% of entire liver, and adequate liver reserve based on liver function tests (serum bilirubin <2 mg/dl, serum albumin >2.5 g/dl, and/or AST or ALT <5 times the upper limit of normal) were included. Following lung shunt study, those with estimated lung dose >30Gy, lung shunt fraction (LSF) >20%, observable uncorrectable GI flow on diagnostic angiogram were excluded from this study as previously described by the authors [23]. In general, 120 Gy dosimetry to the treated lobe of the liver was planned based on Therasphere® (Biocampatibles, London, UK) published guidelines [12,23].

Clinical follow-up and health-related quality of life assessment
Patients were followed up in the Interventional Oncology Clinic at one week post-therapy, and subsequently at one month, three months, and every three months thereafter until death. The SF-36 Health Survey Form (version 1) assessment tool was used to measure HRQOL scores at the initial pre-treatment visit and at each subsequent posttherapy 1-, 3-and 6-month follow-up visit, with a conversion of raw scores to a 0-100 scale [19]. In line with the SF-36 Health Survey Manual and Interpretation Guide [17], the survey was self-administered by patients and scored independently.

Statistical analysis
Overall mean scores were compared at each time point. HRQOL scores of each domain pre-and post-therapy were compared using paired t-tests. The chi-squared (χ 2 ) test was used to compare categorical variables. Using the Kaplan-Meier method overall survival (OS) times were calculated from initial 90 Y procedure. SAS version 9.3 (SAS Institute Inc., Cary, NC, USA) software was used for all computations.

Overall baseline characteristics
Between 2011 and 2013, thirty patients with advanced, infiltrative HCC with PVT received glass-based 90 Y radioembolization. All patients in this study had PVT due to tumor invasion. Detailed patient demographics, tumor morphology, and characteristics of portal vein thrombosis at initial presentation are shown in Table 1. Median age at baseline was 62 years (range 35-92), with the majority being male (n = 23; 77%) and of Caucasian ethnicity (n = 24; 80%). At baseline, 20 patients (67%) had documented Child Pugh A disease, and the majority had an Eastern Co-operative Oncology Group (ECOG) performance score of 1 (n = 17; 57%). All 30 patients demonstrated PVT on MR imaging and were thus classified as BCLC Stage C, with most patients demonstrating occlusive PVT (n = 23; 77%) occurring in a branch of the portal vein (n = 24, 80%). Twenty-one patients (70%) had evidence of hepatic cirrhosis and/or portal hypertension on baseline MR imaging. Twenty-two patients (73%) had a non-liver-related comorbidity at the baseline, and the majority of these patients suffered from hypertension and/ or diabetes. Overall at the time of receipt of 90 Y radioembolization, there were 6 patients who were receiving Sorafenib concurrently, and an additional 11 patients had received Sorafenib prior to 90 Y therapy. With the exception of Sorafenib, no other concurrent locoregional or systemic HCC-directed therapy was received.

SF-36 survey results
Of the 30 patients, all patients completed 1-month follow-up surveys, with 26 (87%) and 20 (67%) having completed 3-month and 6-month follow-up surveys respectively at time of data collection. The decline in response rate at was due to 4 patient deaths at 3 months and 10 patients deaths by 6 months. Twenty-three patients (77%) were deceased at the time of data analysis. The median time interval after initial 90 Y radioembolization therapy to the first follow-up survey was 4.5 weeks. Median follow-up time for all patients was 17.4 months following initial 90 Y therapy. Lower baseline raw SF-36 scores in each of the 8 domains were observed on comparison to the raw scores of a healthy, age-adjusted healthy population in the US (Fig. 1a).
Additionally, the mean SF-36 HRQOL Normalized Baseline Scores (NBS) in all domains were significantly lower when compared to mean scores of the healthy, age-adjusted population in the US (p < 0.001) (Fig. 1b). Baseline physical component scores, including Physical Functioning (PF) and Role-Physical (RP), as well as the overall Physical Component Summary Score (PCS) were more substantially decreased vs. age-adjusted norms compared to mental component scores (Fig. 1b).

HRQOL analysis
There were no significant changes in any of the 8 domains at 1-month, 3-month and 6-months post-90 Y therapy compared to pre-90 Y baseline scores on longitudinal analysis (   (Table 1). Eighteen patients (60%) had 1-month PCS within 2 standard deviations (SD) of US age-adjusted healthy normalized baseline scores, while 12 patients (40%) had 1-month PCS greater than 2 SD below that of US age-adjusted healthy NBS. No significant differences in baseline characteristics were found between patients who had 1month post-90 Y PCS within 2SD vs. >2SD below that of mean NBS. However, it was noted that baseline mean largest tumor size was significantly greater in patients with 1-month PCS within 2SD of NBS (9.7 cm vs. 9.1 cm, p = 0.04).

Discussion
This study examined the efficacy of 90 Y radioembolization therapy on HRQOL in patients with advanced, infiltrative HCC with PVT. Currently, 90 Y radioembolization is considered a viable therapeutic option for palliative treatment of advanced stage HCC patients who are not candidates for surgery. As our results demonstrated, such patients tended to have baseline HRQOL scores significantly below that of the normal age-adjusted healthy population. Given the overwhelmingly poor prognosis of infiltrative HCC with PVT and the tendency to have exhausted other The survival benefits of 90 Y radioembolization therapy on HRQOL in infiltrative, advanced HCC with PVT have not been extensively characterized in the literature for adequate comparison with the current study. However, in a cohort of patients with majority intermediateadvanced (BCLC B/C) disease, 90 Y radioembolization therapy gave rise to significant increases in social and functional well-being at 1 month, without significant changes in overall HRQOL scores [21]. In addition, for HCC patients without PVT, 90 Y radioembolization therapy improved HRQOL scores compared to cisplatin administered through hepatic arterial infusion (HAI) [24], with significantly higher functional well being at 3-months and 6-months post-90 Y therapy vs. cisplatin HAI. Previous studies investigating HRQOL patients receiving other forms of intra-arterial therapy such as transarterial embolization or cTACE have reported decreased HRQOL was at 3 month follow-up compared to baseline [25,26], and improved HRQOL at 6 months vs. 3 months [25]. A study by Wible et al. [20] indicated that HCC patients who received ≥3 cTACE procedures had significantly improved mental health scores following the initial procedure (13 point change, n = 21; p = 0.008) as well as the second procedure (12.2 point change, n = 23; p = 0.002). In addition, significant improvements in bodily pain (9.9 point change, n = 21; p = 0.047) after the first procedure were observed.
HRQOL is increasingly being used as an important prognostic tool in palliative oncology, and some have      reported that it may be just as useful as objective outcomes such as survival or tumor response [2,27,28]. When examining patients with chronic liver disease in the presence or absence of HCC, Kondo et al. found that even though HCC patients had worse physical and overall HRQOL, it was liver function and not the presence of HCC that served as main determinant of HRQOL [24,29,30]. Our study demonstrated positive trends in HRQOL, particularly in the Role-Emotional, Role-Physical, Vitality and Social Functioning domains. Concurrently, there were no significant decreases in any of the remaining domains at 1-3-, or 6-months following treatment. This suggests that it is possible to treat with advanced, infiltrative HCC with PVT with 90 Y radioembolization and increase survival without significant concern about further decreases in HRQOL in these patients.
Our study also demonstrated a median OS of 13.0 months from initial 90 Y radioembolization and a median time to progression of 9.0 months. The survival outcomes of 90 Y radioembolization for infiltrative HCC with PVT have not been widely reported. However, given that all patients in this study were BCLC Stage C, such survival times are comparable with other published studies assessing the efficacy of DEB-TACE in HCC, with median OS of 13.3 months [31] to 13.5 months [32] in BCLC Stage C patients alone. Prior studies of noninfiltrative HCC patients with PVT reported median OS of 10 to 13 months [33,34]. In our study, patients with PCS within 2SD of NBS demonstrated significantly increased median OS (15.7 months vs. 3.7 months, p < 0.001) and significantly prolonged time to progression (12.4 months vs. 1.8 months, p < 0.001) compared those with PCS >2SD below a healthy, age-adjusted US population. This was the case despite the fact that patients with PCS within 2SD of NBS had significantly higher mean index tumor size (9.7 cm vs. 9.1 cm, p = 0.04) compared to patients with PCS >2SD below normalized US population. Furthermore, PCS at 1-month post-90 Y radioembolization was a significant independent prognostic indicator for prolonged overall survival (HR 1.73; 95% CI 1.02-3.18; p = 0.041) and time to progression (HR 3.28; 95% CI 1.48-5.13; p = 0.044), and significant decreases in PCS were independent predictors of poorer OS and TTP. In addition to the highlighting the value of physical component assessments in predicting patient survival, these results indicate the potentially integral role HRQOL analysis may play in comparative effectiveness analysis of therapeutic options.
The strengths of this study included its prospective nature as a correlative study with a phase II randomized controlled trial, with extensive short-and medium-term follow-up of patient outcomes in advanced, infiltrative HCC with PVT. Some limitations included the study's small sample size, the fact that not all patients were able to complete surveys at all follow-up intervals, due either to worsening clinical status or death and subsequent loss to follow-up. In addition, as all HRQOL survey measures are dependent on multifactorial internal and external influences, possible bias (including reporter and observer bias) and confounding factors cannot be completely ruled out.

Conclusions
In summary, our study demonstrated that health-related quality of life was preserved in the short-and mediumterm for patients who received 90 Y radioembolization for infiltrative, advanced HCC with PVT. Furthermore, ECOG performance status, Child-Pugh score, lung shunt fraction and 1-month post-90 Y therapy physical component summary scores were independent prognostic factors for overall survival from therapy.

Availability of data and materials
The data that support the findings of this study were generated at Emory University School of Medicine and may be available upon reasonable request with permission from that institution. As all authors of this study are no longer affiliated with Emory University School of Medicine, the authors are unable to make these data publicly available.
Authors' contributions MX conducted analysis and interpretation of the data, statistical analysis, drafting of the manuscript, and critical revision of the manuscript for important intellectual content. NK was involved in acquisition of data; analysis and interpretation of data; drafting of the manuscript; critical revision of the manuscript for important intellectual content; and statistical analysis. JCC was involved in acquisition of data; analysis and interpretation of data; and critical revision of the manuscript for important intellectual content. HSK was involved in study concept and design; acquisition of data; analysis and interpretation of data; drafting of the manuscript; critical revision of the manuscript for important intellectual content; statistical analysis; and study overall supervision. All authors were involved in critically revising the manuscript prior to final submission. All authors read and approved the final manuscript.
Ethics approval and consent to participate Ethics approval to conduct this study was provided by the Emory University Institutional Review Board. Informed consent to participate in this study was obtained from each participant prior to study enrollment. The study was registered at ClinicalTrials.gov, identifier NCT01556282, on March 16, 2012.

Consent for publication
This manuscript contains only aggregate results and does not present any individual person's data. This manuscript does not contain details, images, or videos relating to any individual person and therefore does not require written informed consent from each individual for publication.