A large analysis of data comparing selective internal radiotherapy (SIRT) plus chemotherapy to chemotherapy did not show expected improvements in survival, but an unanticipated benefit was observed with SIRT in patients with metastatic colorectal cancer.
Harpreet Wasan, MD
A large analysis of data comparing selective internal radiotherapy (SIRT) plus chemotherapy to chemotherapy did not show expected improvements in survival, but an unanticipated benefit was observed with SIRT in patients with metastatic colorectal cancer (mCRC), according to findings from the SIRFLOX trial presented at the 2017 World Congress on Gastrointestinal Cancer.1
The benefit was observed in patients with primary tumors originating on the right side of the colon. In addition, significant control of liver involvement was achieved with SIRT across the entire intent-to-treat population.
The premise for SIRT therapy, which involves injection of yttrium-90 labeled resin microspheres into the blood supply of liver tumors to deliver 1 large dose, is that it would provide control of liver metastases that would lead to improved overall survival (OS) in patients with mCRC.
SIRT was approved by the FDA in 2002 and is supported by National Comprehensive Cancer Network and European Society for Medical Oncology guidelines. SIRT has been authorized in several countries for patients with mCRC who are refractory to chemotherapy.
“Of more than a million patients diagnosed with metastatic colorectal cancer yearly, nearly half will go on to develop liver metastases,” said Harpreet Wasan, MD, head of the gastrointestinal clinical research program at Hammersmith Hospital, in London, England. “SIRT delivered with FOLFOX as first-line therapy improves local control of liver metastases.
Wasan and colleagues performed an analysis of 3 prospective randomized studies of first-line treatment with SIRT plus FOLFOX in 1103 patients with mCRC. In each study, the enrolled patients had adenocarcinoma of the colon or rectum, plus nonresectable liver metastases. All patients had ECOG 0-1 performance status and were eligible for systemic chemotherapy as first-line treatment. Patients with ascites, cirrhosis, or portal hypertension were excluded.
The patients were randomized to receive oxaliplatin-based chemotherapy (mFOLFOX6/OxMdG) with or without bevacizumab or, in the second arm, the same regimen plus a preparation for SIRT during the first cycle of FOLFOX and SIRT delivery during the second cycle. The primary endpoint of all studies was OS, and secondary endpoints included progression-free survival (PFS) at any site, liver specific PFS, objective tumor response, hepatic resection rate, toxicity and safety, and health-related quality of life.
Patient characteristics among the 549 patients in the control arm and 554 patients in the SIRT arm were balanced; the median age was 63 years (range, 23 to 90); 65.6% were male; and 63.5% of patients were ECOG 0. More patients in the SIRT arm had extra-hepatic metastases (35.9% versus 34.8% in the control group) and more had liver involvement (32.3% versus 30.6%). Most patients in both arms presented with synchronous liver involvement, and approximately half had primary tumor in situ.
OS curves were identical for both control and SIRT patients. OS was 23.3 months with FOLFOX versus 22.6 months with SIRT (HR, 1.04; 95% CI, 0.90-1.19;P=.609). Only the FOXFIRE global trial showed a trend toward improved OS, favoring SIRT (HR, 1.04;P= 789). PFS showed a trend also favoring SIRT. The PFS was 10.3 months for control versus 11 months with SIRT (HR, 0.90; 95% CI, 0.79-1.02;P= 0.108).
“Patients in the chemotherapy-plus-SIRT arm had a lower risk of progression in the liver as a first event,” said Wasan.
Liver-specific PFS was statistically superior with SIRT; patients receiving SIRT were 25% less likely to show first progression to the liver over controls (HR, 0.51; 95% CI, 0.43-0.62;P<.0001). The best objective response in the pooled studies was 63% versus 72.2% (P= .001), and the best hepatic response was 63.8% versus 75.8% (P<.001) in the FOLFOX versus SIRT arms, respectively. The liver resection rates were 16% with FOLFOX compared to 17% with SIRT (P= .669).
Toxicity was higher in the SIRT arm, particularly regarding hematological events. Adverse events that were elevated over controls included neutropenia, febrile neutropenia, thrombocytopenia, and leukopenia.
“Addition of SIRT to FOLFOX first-line chemotherapy in patients with liver only or liver dominant mCRC did not improve OS or PFS,” said Wasan.
However, a subgroup analysis of the data revealed patients with primary tumor located on the right side derived significant benefit from SIRT over control (HR, 0.67; 95% CI, 0.48-0.92), suggesting that SIRT treatment may improve response in these patients.2
This is different from the usual pattern for patients with mCRC that has metastasized to the liver, wherein having a primary tumor on the left, as opposed to the right side of the colon, confers an advantage in treatment response. The significant benefit with SIRT observed in liver metastases from right-sided primary tumors merits further evaluation, Wasan concluded.
This advantage uncovered by Wasan’s analysis was elaborated upon by Guy van Hazel, MD, of the University of Western Australia, Perth. He cited the CRYSTAL study, which showed OS was improved in patients with primary right side (PRS) versus primary left side (PLS) origin.3 A 36% better OS rate (HR, 1.06;P= .002) was observed after treatment with a combination of first-line cetuximab/FOLFOX and SIRT in right-sided versus left-sided location.1The SIRT chemotherapy combination was no better than chemotherapy in patients with PLS.
“The location of the primary tumor in mCRC is emerging as a major prognostic factor of response to treatment,” explained van Hazel. “These findings are good news for patients with right-sided primary tumors, who have a much worse prognosis and fewer treatment options than patients with left-sided tumors,” said van Hazel.
Van Hazel and colleagues performed a retrospective analysis that revisited the 3 studies of SIRT; however, data regarding sidedness of the primary tumor were only available in the SIRFLOX and FOXFIRE global studies, comprising a total of 719 patients. Using these studies, van Hazel reported in more detail on the impact of sidedness on SIRT efficacy.
PRS was defined as any primary tumor near to the splenic flexure and PLS included any primary tumor at the splenic flexure, the more distal colon, or the rectum. The patient population was similar to that in the prior analysis, except that 81 patients with PRS tumors received FOLFOX and 98 received SIRT, while 276 patients with PLS received FOLFOX and 264 received SIRT. The mean liver burden across these cohorts was higher in the SIRT-treated arms: 15.4%, 19.4%, 15.8%, and 16.5%, respectively.
Although, survival outcomes were similar across treatment arms, significantly improved OS was seen in patients with liver metastases from PRS with SIRT of 22 months compared with 17.1 months in controls (HR, 0.64; 95% CI, 0.46-0.89;P= .007). This result was not seen in patients with PLS, who demonstrated OS of 24.6 versus 25.6 months with SIRT and FOLFOX respectively (HR 1.12; 95% CI, 0.92-1.36;P= .279).
“That means that PRS patients treated with chemotherapy plus SIRT have a 36% reduced risk of dying at any time,” said van Hazel.
A non-statistical trend toward improved PFS of about 27% was also observed (HR, 1.12;P= 0.279). This analysis confirmed the benefit to the liver observed with SIRT. “There was a higher hepatic response in the entire population in favor of the SIRT/ chemotherapy combination,” said van Hazel.
“This is the first time that location of primary tumour has been linked to radiation therapy,” said van Hazel, “SIRT in patients with mCRC and primary tumors originating on the right side of the colon results in statistically significant and clinically meaningful improvements. This opens a new treatment option for these patients.”
Session co-chair, Eric Van Cutsem, University Hospitals Leuven, Belgium, said, “The well-known worse prognosis of right-sided tumors increases the relative importance of a nonsystemic treatment option; however, it remains to be confirmed whether these results mean that right-sided tumors are more sensitive to this kind of radiotherapy, or whether this is related to the fact that the molecular characteristics of right-sided tumors allow fewer treatment options, because they have more mutations,”
In response, van Hazel said: “The significant treatment interaction by location provides further evidence that the observed benefit was not a chance finding.” However, he agreed: “The drivers of the observed side-based differences in treatment impact remain to be elucidated.”