The jury is still out on whether stereotactic body radiation therapy (SBRT) should truly be considered a preferred frontline treatment modality for patients with hepatocellular carcinoma (HCC) who are not candidates for surgery, despite an ever-growing body of evidence supporting this approach.
In a recent retrospective study of patients with inoperable, nonmetastatic HCC, 1- and 2-year freedom from local progression (FFLP) rates were greater with SBRT versus radiofrequency ablation (RFA). The results of the study were reported in theJournal of Clinical Oncologyby Daniel Wahl, MD, PhD, and colleagues with the University of Michigan Medical Center.1Prior to this publication, data comparing SBRT and RFA were lacking.
While surgical resection is the preferred first-line therapeutic intervention for patients with a single HCC lesion and preserved liver function, many patients with multifocal disease, poor liver function, or major vascular invasion need to be managed with other locoregional therapy. These alternative therapeutic strategies may include RFA, SBRT, microwave ablation, percutaneous ethanol injection, or various embolization techniques.2,3
Previous reports have shown similar local recurrence rates between SBRT and RFA, ranging from 10% to 30% for small tumors.4,5However, specific outcomes and treatment failure in various patient and tumor populations, especially those with large tumors, was unknown. With retrospective results in hand, the next step will be a prospective comparison of the two approaches, if it is deemed feasible.
Support of SBRT as a Frontline Therapy in HCC
In the study by Wahl et al, 249 lesions in 161 patients were treated with RFA, and 83 lesions in 63 patients were treated with SBRT. Compared with patients who had received SBRT, those treated with RFA had significantly lower alpha-fetoprotein levels, higher rates of cirrhosis, fewer previous liver-directed treatments, and longer follow-up. Median maximum tumor diameter was similar between treatment groups.1
Freedom from local progression rates were greater for tumors treated with SBRT compared with those treated with RFA. The 1-year and 2-year FFLP rates were 97.4% versus 83.6% and 83.8% versus 80.2% for SBRT and RFA, respectively.
To account for any imbalances in treatment assignment, authors used an inverse probability of treatment weighting (IPTW) to the Kaplan-Meier and Cox models for FFLP. Using the IPTW univariate analysis, patients who received RFA had a significant association with local progression compared with SBRT (HR, 2.63;P= .016). Among all of the variables analyzed, only tumor size was predictive of local progression (HR, 1.36;P= .029). Larger tumor size was predictive of treatment failure with RFA but not with SBRT.1
Further stratification of data was performed to identify additional differences between treatments. Although FFLP was similar in tumors smaller than 2 cm, significantly worse FFLP was associated with RFA (HR, 3.35; 95% Cl, 1.17-9.62,P =.025). One- and 2-year overall survival (OS) rates were similar between treatment groups.
The rates of grade ≥3 adverse events (AEs) were 11% for the RFA cohort and 5% for the SBRT cohort. Radiofrequency ablation complications included bleeding (n = 3), duodenal and colonic perforation (n = 2), sepsis (n = 2), and pneumothorax (n = 1). Stereotactic body radiation therapy complications included radiation-induced liver disease (RILD; n = 1), gastrointestinal bleeding (n = 1), and worsening ascites (n = 1). Radiofrequency ablation-associated complications resulted in 2 deaths, 1 from hemothorax and 1 from gastrointestinal bleeding, while no deaths resulted from SBRT.
“These results suggest that both SBRT and RFA are excellent choices for smaller tumors but that SBRT may be preferred for larger tumors,” the authors concluded. “Prospective, randomized clinical trials are needed to compare these two modalities, especially for larger tumors, although we are unaware of any such ongoing trials.”1
Opposition to SBRT
Not everyone is in agreement regarding the frontline use of SBRT in HCC, despite the growing evidence of its clinical utility. In response to the retrospective study by Wahl et al, international clinicians have voiced their concerns over the endorsement of SBRT as the “preferred treatment for larger HCC.”
In one commentary, Berardino De Bari, MD, and colleagues, with Centre Hospitalier Universitaire Vaudois, Lausanne, Switzerland, have argued that although the authors have shown that SBRT is safe and efficient, the European Society for Medical Oncology and European Society of Digestive Oncology (EMSO-ESDO) and European Association for the Study of the Liver and European Organization for Research and Treatment of Cancer (EASL-EORTC) clinical practice guidelines largely support the use of radiotherapy (RT) as a palliative treatment option. Furthermore, De Bari and coauthors questioned the potential of some methodologies from the Wahl study that may have influenced the results.6
De Bari et al pointed out that efficacy assessment within the retrospective study used RECIST criteria to determine the FFLP rate. They argued that others have shown that RECIST and modified RECIST (mRECIST) are not optimal for examining HCC response to SBRT, and that the evaluation of response or recurrence itself is challenging because of limitations in imaging technologies.
Additionally, De Bari and colleagues highlighted patient heterogeneity, particularly in follow-up duration, as one contributing influence on the results. Follow-up was 20 versus 13 months for RFA and SBRT, respectively, which could have influenced efficacy in terms of late relapses and/or late complete responses. The commentators suggested that more homogeneous patient cohorts and comparisons with radioembolization or transarterial chemoembolization are needed in future randomized trials.
In another commentary, Jen-Fu Yang, MD, and coauthors, with the Tri-Service General Hospital, Taipei, Taiwan, voiced 2 main concerns with the Wahl et al study.7First, Yang and colleagues pointed to the assessment of treatment response. Many ablative techniques, including RT and RFA, result in tumor necrosis rather than tumor shrinkage. As such, evaluation criteria that rely only on measurements of tumor size may not accurately reflect tumor response. While EASL and mRECIST criteria incorporate tumor necrosis into their evaluation guidelines, the original RECIST, as used by Wahl et al did not, and could have contributed to distorted data.
Second, Yang et al discussed the ambiguity surrounding the timing of liver transplantation in each treatment cohort. When performed before treatment, transplantation has the potential to significantly impact OS rates. The commentators suggested censorship of patients with subsequent liver transplantation from the time of the procedure onward for survival estimates. Similarly to De Bari et al, Yang and colleagues advocated for the need for future prospective randomized trial for SBRT.
Wahl and colleagues have issued a reply in response to the commentaries. They clarify that because their multidisciplinary board evaluates tumor growth as a part of disease progression, the original RECIST criteria most closely aligned with their clinical practice. Further, they note that RECIST is the standard assessment of the Radiation Therapy Oncology Group.8
With regard to liver transplantation, Wahl et al reiterate that few patients underwent the procedure in the original study, and that pathologic complete response (pCR) rates were similar (SBRT 44% vs RFA 33%) and consistent with other published studies.
To address the need for additional studies, Wahl et al outline plans for a prospective clinical trial. “We are in the process of opening a clinical trial comparing RFA with SBRT to test the results of this study in both larger and smaller HCCs,” Wahl and colleagues stated. “Given the difficulties of imaging-based treatment response for HCC, one of the goals of this trial will be to assess the pathologic treatment response of both modalities at the time of transplantation. This trial will also address the differences in follow-up between the two modalities seen in our current study and allow for the prospective determination of acute toxicity rates.”
Additional Studies on the Use of SBRT in the Frontline Setting
Although met with some criticism, additional studies have supported the use of SBRT in inoperable or advanced HCC. Alexander Kirichenko, MD, from Allegheny General Hospital, Pittsburgh, PA, and colleagues found that SBRT alone or in combination with surgery was safe and effective for inoperable liver tumors. The investigators presented their findings at the 2016 ASCO Gastrointestinal Cancers Symposium.9
In the retrospective study, patients with primary (n = 43) or isolated metastatic tumors (n = 48) received from 4 to 6 SBRT treatments. Local control was the primary endpoint of the study, and toxicity and survival were the secondary endpoints.
After a median follow-up of 20.3 months, the 2-year local control rates were 96% and 94% for primary and metastatic tumors, respectively. For tumors ˂4 cm, local control rate was 100%. Of the 14 patients that underwent liver transplant and who received SBRT as a bridging therapy, 10 patients had pCR. In 8 patients, liver resection was performed in conjunction with SBRT for unresectable tumors. After 2 years, the OS rates were 82% and 64% for primary and metastatic disease, respectively.
The toxicity with SBRT was minimal. No incidence of accelerated Child-Pugh class migration, grade ≥3 AEs, or operative or perioperative complications were reported with SBRT.
“Surgery remains the gold standard for treating liver cancer, but unfortunately more than 80% of patients with liver cancer are not eligible for surgery,” said Kirichenko, lead author of the study. “SBRT offers these patients a safe option that can stand on its own as a treatment or serve as an effective bridge to surgical removal of the tumor or a liver transplant.”10
Not only has SBRT been shown to be a viable option for those with inoperable tumors, but it also has shown promise for those with portal vein tumor thrombosis (PVTT) or inferior vena cava tumor thrombosis (IVCTT). In theJournal of Radiation Research, Yoshiro Matsuo and colleagues with the Kobe University Graduate School of Medicine, Japan, reported that SBRT had favorable outcomes over 3-dimensional conformal RT (3DCRT) in patients with HCC with PVTT/IVCTT.11
Portal vein tumor thrombosis and IVCTT are conditions that plague up to 40% of patients with HCC at the time of diagnosis. If left untreated, patients with PVTT/IVCTT have a median survival of just 2 to 4 months. Portal vein tumor thrombosis and IVCTT can lead to aggressive complications that worsen prognosis, including extensive intrahepatic tumor spreading, portal vein hypertension, and ischemic liver damage. Treatment of patients with PVTT/IVCTT has remained challenging, as most patients are unsuitable for surgery.12
In the retrospective analysis, 43 patients were treated with SBRT (either with CyberKnife or TrueBeam), and 54 patients were treated with 3DCRT. In comparison with 3DCRT, rates of tumor response and OS were greater for total SBRT, CyberKnife, and TrueBeam while 1-year local progression rates were lower. Additionally, higher median biologically effective dose 10 (BED10) was achieved with SBRT compared with 3DCRT.11
Studies have also explored outcomes for transarterial chemoembolization (TACE) plus RT, using various delivery methods, for patients with unresectable HCC. In one such analysis, TACE plus RT showed better clinical outcomes compared with TACE alone. Ya Huo, MD, and Guy Eslick, MD, with the University of Sydney, Australia, analyzed combination therapy versus TACE alone, and published their findings inJAMA Oncology.13
The meta-analysis of over 2500 patients from 25 clinical trials (11 randomized, controlled trials) looked for differences in survival, tumor response, and AEs using a random effects model. Radiotherapy techniques included in the analysis were 3DCRT, SBRT, and the moving strip technique for whole-liver irradiation.
Compared with TACE alone, patients who were treated with TACE plus RT had significantly better 1-year survival (odds ratio [OR], 1.36; 95% CI, 1.19-1.54) and complete response (OR, 2.73; 95% CI, 1.95-3.81). There was a corresponding increase in benefits in the 2-, 3-, 4-, and 5-year survival rates (OR, 1.55, 1.91, 3.01, and 3.98, respectively). Gastroduodenal ulcers and elevated alanine aminotransferase and bilirubin levels were observed with a greater incidence in the combination group than in the TACE-alone group.
Importantly, the authors stress, “RT for HCC has traditionally not been used because older techniques have not been able to adequately localize the radiation to the tumor. However, recent improvements in RT have allowed increased intratumor radiation and decreased radiation to the adjacent normal liver and organs, thereby reducing the rate of adverse events such as RILD.”
Direct Comparisons of SBRT With Other Approaches
Additional studies have directly evaluated SBRT and TACE, the standard locoregional therapy for those with intermediate HCC. In one study, SBRT was found to be a safe alternative to TACE, demonstrating better local control. Eli Sapir, MD, with the Hadassah Hebrew University Medical Center, Jerusalem, Israel, and colleagues presented findings at the 2016 ASCO Annual Meeting for local control, OS, and toxicity between these approaches.14
Patients with portal venous involvement, extrahepatic spread, or Child-Pugh class C liver function were excluded from the analysis because they were ineligible for TACE treatment. Outcomes were measured in an 8-year period in 84 patients with 114 lesions treated with TACE and 125 patients with 173 lesions treated with SBRT. The median follow-up was 28 months.
Although OS was not significantly different between treatment groups, 1- and 2-year local control rates both favored SBRT over TACE (97% vs 41% and 91% vs 18%, respectively; HR, 18.8; 95% CI, 6.7-52.7;P<.001). In TACE-treated patients but not SBRT-treated patients, factors that were predictive of worse local control included increasing tumor size and partial PVTT. Grade ≥3 AEs occurred at a higher rate with TACE than with SBRT (14% vs 7%, respectively;P= .05).
In the past, the use of RT was limited by potential detrimental AEs, including the risk of RILD and low tolerance in general. However, technological advances, such as with SBRT, allow for higher and more targeted radiation doses to the tumor site, while sparing adjacent tissue and organs for substantial radiation doses. Although met with some criticism, ongoing clinical trials will address the potential of SBRT as the preferred treatment for large HCC lesions.