ESMO 2017-NO BENEFIT OF SIRT FOR COLORECTAL CANCER

Although the expected improvements in survival did not emerge from a large analysis of data comparing selective internal radiotherapy (SIRT) plus standard chemotherapy to chemotherapy in patients with metastatic colorectal cancer (mCRC) and inoperable liver metastasis, an unanticipated potential benefit was observed with SIRT in patients having primary tumours originating on the right side of the colon, researchers reported at ESMO 2017, the Annual Congress of the European Society for Medical Oncology in Madrid, Spain.

The premise behind SIRT therapy, which involves injection of yttrium-90 labeled resin microspheres into the blood supply of liver tumours to deliver large but safe doses of radiation, is that it would provide control of liver metastases leading to improved overall survival (OS) in patients with mCRC.

SIRT was approved by the US Food and Drug Administration (FDA) in 2002 and has been authorised in several countries for patients with mCRC who are refractory to chemotherapy.

Harpreet Wasan, head of the gastrointestinal clinical research programme at Hammersmith Hospital, Imperial College London, UK and Australian colleagues performed an analysis of 3 similarly designed, prospective randomised studies (FOXFIRE, ISRCTN83867919, SIRFLOX NCT00724503, and FOXFIRE-Global NCT01721954).

The studies all evaluated first-line treatment with SIRT, using yttrium-90 resin microspheres, plus chemotherapy with FOLFOX (+/- bevacizumab) in 1,103 chemotherapy-naive patients with mCRC and non-resectable liver metastases.

The patients were randomised to receive standard oxaliplatin-based chemotherapy with or without a biologically targeted agent (control arm, n=549) or to the same chemotherapy regimen plus SIRT (test arm, n=554). Patients with ascites, cirrhosis or portal hypertension were excluded.

The primary endpoint of all studies was OS, and in this updated analysis the influence of KRAS mutation and origin of tumour by side of colon were analyzed.

OS was shorter in patients with mutated KRAS but no difference between treatments was observed

At a median follow-up of 43.3 months, OS for the whole trial population was identical for control and SIRT patients, hazard ratio [HR] 1.04; 95% confidence interval [CI] 0.90, 1.19 (p = 0.609). No difference in progression-free survival (PFS) between the treatment arms was noted, HR 0.90; 95% CI, 0.79 – 1.02 (p = 0.108). However, the objective response rate was higher in the SIRT arm than control (72.2% and 63.0%, respectively, p=0.001)

Analysis of the KRAS status of the patients showed that 182 (16.5%) patients had a KRAS mutation, 279 (25.3%) had KRAS wild-type status, and 642 (58.2%) patients currently had an unknown status.

The OS did not differ between treatment arms in patients harbouring a KRAS mutation, HR 1.03; 95% CI 0.74, 1.43 (p = 0.878) or in patients with wild-type KRAS, HR 1.10; 95% CI 0.83, 1.46 (p = 0.499). Patients with currently unknown KRAS status also showed no difference in OS between treatment arms, HR 1.01; 95% CI 0.85–1.21 (p = 0.878).

However, it was determined that OS was lower overall in patients having mutated KRAS in both arms, than in patients with either wild-type of unknown or untested KRAS status. Patients with mutated KRAS demonstrated median OS in the control versus SIRT arms of 19.1 months versus 18.7 months, whereas OS in patients with wild-type KRAS was 28.3 months versus 24.2 months, and OS in patients with unknown or currently untested KRAS status was 23.1 versus 22.6 months, respectively.

Subgroup of patients with right-sided primary tumours demonstrated an OS benefit with SIRT

However, subgroup analysis of the data suggested that patients with the primary tumour located on the right side had significantly prolonged OS with SIRT of 22.0 months versus 17.1 months with control (p = 0.007), while no significant difference in OS between treatment arms was observed in patients with primary tumours located on the left side, where OS was 24.6 versus 26.6 months in the respective arms (p = 0.279).

These findings are consistent with emerging data from many other studies showing that in mCRC, right-sided tumours behave differently in that they are prognostically worse and also appear to derive less benefit from current cytotoxic and biological strategies. This study raises the intriguing question whether SIRT confers survival efficacy in the subset of right sided tumours.   

Conclusions

The investigators concluded that KRAS mutation in the cohorts currently tested, does not appear to be a predictive factor for survival outcome following SIRT in mCRC.

However, OS was significantly better with the addition of SIRT in patients having right-side primary tumours.

These data suggest that primary tumour site and not KRAS status may predict for potential treatment interaction with SIRT.

Thomas Gruenberger of the Department of Surgery I, Rudolfstiftung Hospital, Vienna, Austria, who discussed the study results said that in the SIRFLOX and FOXFIRE Global study cohorts, the addition of SIRT to first-line chemotherapy was retrospectively associated with a statistically and clinically significant gain in OS for patients with right-sided primary tumours, but not for left-sided primaries. This effect is not seen for KRAS mutation status. These data suggest that the primary tumour site but not KRAS status may predict for potential treatment interaction with SIRT.

This analysis adds to the increasing literature for primary tumour location-based differences in mCRC outcomes with treatments, and may support a side-based approach to patient selection for SIRT. However, SIRT in the first-line setting cannot be recommended today.