Local Immunomodulation After Radiofrequency of Unresectable Colorectal Liver Metastases

Early clinical trials of immunotherapy in colorectal cancer (CRC) have provided first evidence of activity of adoptive cell vaccines combining inactivated autologous tumor cells and BCG, a Toll Like Receptor (TLR) agonist. RFA induces tumor cell necrosis and apoptosis, leading to the activation of anti-tumoral immunity through the release, exposure or denaturation of tumor antigens, which are captured by dendritic cells. Associated pro-inflammatory effects increase the anti-tumoral immune response but remain insufficient to avoid recurrence. In a preclinical study in an aggressive rectal cancer model, complete and long-lasting tumor responses were achieved after RFA combined with in situ injection of an immunomodulatory hydrogel combining a TLR agonist and GM-CSF. These results were obtained without further adjuvant therapy after complete clearance of macroscopic liver lesions by RFA(1,2). BCG, bacteria Protein Derivatives can be used as TLR or NOD2 agonists. The hypothesis of the study is that local immunotherapy may reduce recurrence rates after RFA in patients with unresectable liver metastases from CRC. Thermal ablation by RFA may be used as an inducer of anti-tumor immune response and the combination with immunomodulators such as a TLR or NOD2 agonists and GM-CSF may improve its efficacy. In the recent preclinical study, the efficacy has been demonstrated on distant lesions (so-called abscopal effect) of two immunomodulators combined in a muco-adherent hydrogel, injected in a RFA-treated tumor. The treatment was safe. Complete macroscopic ablation by RFA combined with in situ immunomodulators may be able to prime an effective immune response capable of eradicating the microscopic residual disease. This study is an open label, quasi-experimental, non-randomized, single-arm trial. Will be included patients with unresectable colorectal hepatic metastases, without extra hepatic localization. A complete clearance of lesions must be obtained by the use of radiofrequency alone or combined with surgery. These patients will have complete destruction, by laparotomy, by radiofrequency alone or associated with surgery of all the liver metastases detectable at the time of treatment. 12 patients meeting inclusion criteria and not exclusion ones will be included and expected to be valuable patients. 3 subgroups of 4 patients each will receive an immunogel injection in 1, 2 and 3 metastases respectively as dose escalation. The primary endpoints will be collected in post-operative course until 3 months however, the patients will be followed in the trial until 12 months after RFA. The radiofrequency of one, two or three lesions (dose escalation) selected on preoperative imaging, will be followed by injection, under ultrasound control via a coaxial device, of 5 ml (lesion \< 2 cm) or 10 ml (lesion ≥ 2 cm) of thermoreversible gel vectorising 2 immunomodulators, mifamurtide (8 µg / ml) and GMCSF (25 µg / ml). The choice of mifamurtide, the liposomal form of the smallest synthetic immunogenic derivative of the wall of mycobacteria, is justified by the need to replace BCG, used in preclinical studies, by an inert immunomodulator. This compound has demonstrated its clinical efficacy in anti-tumor immunotherapy. GMCSF was largely used in antitumor immunotherapies studies and platforms.