Targeting the Epidermal Growth Factor Receptor in Addition to Chemotherapy in Patients with Advanced Pancreatic Cancer: A Systematic Review and Meta-Analysis

Abstract

Overexpression of epidermal growth factor receptors (EGFR) occurs in >90% of pancreatic ductal adenocarcinomas (PDACs) and is associated with a poorer prognosis. A systematic review of electronic databases identified studies exploring the addition of EGFR-targeted treatment to chemotherapy in patients with locally advanced (LA)/metastatic PDAC. Efficacy, safety and tolerability of EGFR-targeted therapy were explored using meta-analysis of randomised controlled trials (RCTs). Meta-regression was utilised to explore factors associated with improved prognosis (all studies) and benefit from EGFR-targeted therapy (RCTs). Twenty-eight studies (7 RCTs and 21 cohort studies) comprising 3718 patients were included. The addition of EGFR-targeted treatment to chemotherapy did not improve progression-free (pooled hazard ratio (HR): 0.90, p = 0.15) or overall survival (HR: 0.94, p = 0.18). EGFR-targeted therapy was associated with increased treatment-related deaths (pooled odds ratio (OR): 5.18, p = 0.007), and grade (G)3/4 rash (OR: 4.82, p = 0.03). There was a borderline significant increase in G3/4 diarrhoea (OR: 1.75, p = 0.06), but no effect on treatment discontinuation without progression (OR: 0.87, p = 0.25). Neither G3/4 rash nor diarrhoea were associated with increased survival benefit from EGFR-targeted therapy. The effect of EGFR-targeted therapy on overall survival (OS) appeared greater in studies with a greater proportion of LA rather than metastatic patients (R = -0.69, p < 0.001). Further studies in unselected patients with advanced PDAC are not warranted. The benefit from EGFR inhibitors may be limited to patient subgroups not yet clearly defined.

Keywords: KRAS; advanced pancreatic cancer; chemotherapy; epidermal growth factor receptors (EGFR); rash.

Figure 1

Flow chart outlining search strategy and details on final included and excluded studies in the meta-analysis. n: number; ASCO: American Society of Clinical Oncology; ESMO: European Society of Medical Oncology; EGFR: epidermal growth factor receptor.

Figure 2

Forest plot showing hazard ratio for progression-free survival for addition of EGFR-targeted treatment to chemotherapy versus control. Experimental arm: EGFR-targeted therapy + chemotherapy; Control: chemotherapy; CI: confidence interval.

Figure 3

Forest plot showing hazard ratio for overall survival for addition of EGFR-targeted treatment to chemotherapy versus control. Experimental arm: EGFR-targeted therapy + chemotherapy; Control: chemotherapy; CI: confidence interval.

Figure 4

Individual study association of proportion of patients with KRAS mutations and median overall survival (in months). Each study is represented by a circle, and the area of the circle is proportional to the number of patients enrolled in each study. The gradient of the line represents the results of the meta-regression (R = −0.88).

Figure 5

Forest plots showing odds ratio for toxic death (A); treatment discontinuation (B); risk of grade 3–4 skin toxicity (C); risk of grade 3–4 diarrhoea (D); fatigue (E) and stomatitis (F) for addition of EGFR-targeted treatment to chemotherapy versus control. Experimental arm: EGFR-targeted therapy + chemotherapy; Control: chemotherapy; CI: confidence interval.

Figure 5

Forest plots showing odds ratio for toxic death (A); treatment discontinuation (B); risk of grade 3–4 skin toxicity (C); risk of grade 3–4 diarrhoea (D); fatigue (E) and stomatitis (F) for addition of EGFR-targeted treatment to chemotherapy versus control. Experimental arm: EGFR-targeted therapy + chemotherapy; Control: chemotherapy; CI: confidence interval.