Adding Cetuximab to chemoradiation in non-operable oesophageal cancer does not improve survival

Adding cetuximab to concurrent chemoradiotherapy did not improve overall survival in patients with adenocarcinoma or squamous cell carcinoma of the esophagus according to findings of RTOG 0436, a randomized phase III trial presented by David Ilson, MD, of Memorial Sloan Kettering Cancer Center, at the 16th World Congress of Gastrointestinal Cancer on June 26, 2014.


One Response to “Adding Cetuximab to chemoradiation in non-operable oesophageal cancer does not improve survival”

  1. Dr Mel Green

    The primary endpoint used in this RTOG study – overall survival (OS) – and the inclusion of a large proportion of patients with advanced disease mean that unfortunately it does not specifically answer whether cetuximab acted as a radiosensitizer to enhance localized RT response, which could permit dose escalation/improve resectability rates. The design infers the expectation that the combination of cetuximab and RT may improve the control of metastatic disease, but is this what we should realistically expect cetuximab to do? The secondary endpoint of cCR histology at 8 weeks was a more appropriate endpoint for assessment of the primary tumour, but may still not be the best indicator of how well a defined local tumour responds to chemoradiation or whether it is ‘controlled’ – is it realistic to expect it to disappear histologically?

    We set up the first phase II clinical trial (PACER) specifically to address whether cetuximab acted as a radiosensitizer in locally advanced pancreatic cancer (LAPC), using RT and cetuximab only (no chemo) and found evidence indicating the efficacy of cetuximab as an EGFR targeted radiosensitizer consistent with pre-clinical studies
    (Rembielak et al., 2014;
    Local control was excellent on RECIST CT assessment with tumour shrinkage for the majority of patients. This extended survival in some patients (11% alive at 3 yrs) and no patients died of localized disease. Toxicity was low (no gd 3 or 4). However, overall median survival in the group was affected by the poor control of metastatic disease in other patients. This showed the need to effectively and simultaneously control metastasis as well as to identify which patients may respond best to localized treatment with cetuximab. The results were consistent with a localized radiosensitizing effect that accord with other trials that have tested cetuximab in combination with RT in locally advanced disease and shown encouraging results (Crane et al., 2011, Arnoletti et al., 2011, Demols et al., 2008, Munter et al., 2008). But cetuximab tested with chemotherapy agents in advanced pancreatic cancer trials have been negative (Philip et al., 2011, Cascinu et al., 2008, Kullmann et al., 2009), lending weight that it is unrealistic for cetuximab to have independent effects upon disseminated disease outside the radiation field.

    Interestingly, associations were found in this study between longer OS and cCR at 8 weeks and small primary tumour (<5cm), indicating that controlling the primary tumour is important for survival – as with pancreatic cancer. It is also notable that low toxicity of cetuximab is reported (‘treatment related adverse events similar between both arms’), consistent with the PACER trial, and that the study was halted partly due to the negative results of the SCOPE1 trial – which reported high toxicity and again used histology on endoscopic biopsy to determine treatment failure. Incidences of severe adverse reactions to cetuximab appear to be much greater in trials where prophylactic chlorphenamine/NSAID/steroids were not used – as in SCOPE1 ‒ with more severe infusion, skin and other adverse reactions reported (Crane et al 2011; Chakravarthy et al 2012, Crosby et al., 2013). The RT methodology and choice of chemotherapy agent may also be important for the reduction of adverse events. Importantly, in SCOPE1 the higher toxicity (particularly the 22% grade 3 and 4 dermatological) led to lower compliance, significantly less pts in the cetuximab arm receiving their planned RT treatment (making it difficult to fairly compare arms or the potential effect of cetuximab as a radiosensitizer) and contributed to decisions on not recommending the treatment – the negative results that then influenced continued accrual in this study, despite the experience of lower toxicity.

    In summary, cetuximab should not yet be ‘written off’ as a potential adjunct to CRT in oesophageal cancer on the basis of this study and the SCOPE1 trial. This is because potential cetuximab efficacy depends on whether its effects are being appropriately tested and whether toxicity has been minimized. Individualized patient/tumour response to cetuximab also needs to be considered. Several other studies have also shown encouraging results in support of cetuximab as an effective radiosensitizer in localized cancer (including Rembielak, 2014; Crane 2011; Arnoletti 2011; Bonner 2006) and these should not be dismissed entirely. Neither should the basic and translation studies that indicate potential involvement of molecules such as KRAS, ERBB2, SMAD4 and DPC4 in EGFR oncogenesis and radiosensitivity. Instead they should be used as balance indicators to highlight the need to i) more effectively study the potential clinical action of cetuximab and ii) return to translational and basic biological studies to elucidate mechanisms and markers of cetuximab+RT action/efficacy that may occur in selected individual patients.