Predictive Role of Skin Rash in Advanced Pancreatic Cancer Patients Treated With Gemcitabine Plus Erlotinib: A Systematic Review and Meta-analysis

Purpose: The survival benefit from gemcitabine plus erlotinib was on average marginal for advanced pancreatic cancer (APC) patients. Skin rash developed shortly after starting treatment seemed to be associated with better efficacy and might be used to assist clinical decision-making, but the results across studies were inconsistent. Thus, we conducted a systematic review and meta-analysis.
Methods: PubMed, Embase, Cochrane Central Register of Controlled Trials, three Chinese databases, and the abstracts of important conferences were searched for eligible studies. The primary outcome was overall survival (OS), and the secondary outcomes were progression-free survival (PFS) and objective response. The random-effects model was used to pool results across studies if heterogeneity was substantial. Otherwise, the fixed-effect model was used.
Results: A total of 16 studies with 1,776 patients were included. Patients who developed skin rash during treatment had longer OS (8.9 vs 4.9 months, HR=0.57, 95% CI 0.50–0.64) and longer PFS (4.5 vs 2.4 months, HR=0.53, 95% CI 0.40–0.68) than those who did not. A dose–response relationship was also observed for both OS (HR=0.64 for grade-1 rash vs no rash and HR=0.46 for ≥grade-2 rash vs no rash) and PFS (HR=0.72 for grade-1 rash vs no rash and HR=0.43 for ≥grade-2 rash vs no rash).
Conclusion: Skin rash was associated with better OS and PFS in APC patients treated with gemcitabine plus erlotinib. It might be used as a marker for efficacy to guide clinical decision-making toward a more precise and personalized treatment.

Keywords: pancreatic neoplasms, targeted treatment, acne, prognosis 

INTRODUCTION

As a highly malignant disease, pancreatic cancer is the leading cause of cancer-related deaths in the world, including the US, the UK, and Hong Kong.^1–3 More than 80% of patients are diagnosed when the cancer is already at a locally advanced or metastatic stage, which is often referred to as advanced pancreatic cancer (APC).² Without adequate treatment, the median survival of these patients is only ~3–5 months.⁴ For years, gemcitabine has been the standard treatment of APC.⁵ However, its benefit is small, with an increase of only 1.2 months in overall survival (OS).⁶ Recently, three regimens, namely gemcitabine plus erlotinib (6.24 vs 5.91 months, P=0.038),⁷ gemcitabine plus nab-paclitaxel (8.5 vs 6.7 months, P<0.001),⁸ and combination therapy of oxaliplatin, irinotecan, fluorouracil, and leucovorin (FOLFIRINOX; 11.1 vs 6.8 months, P<0.001),⁹ were proved to be able to achieve longer survival as compared with gemcitabine alone.

Although the survival benefit provided by gemcitabine plus erlotinib is less than that by gemcitabine plus nab-paclitaxel or FOLFIRINOX, this regimen is associated with much fewer severe (grade 3 or grade 4) adverse events as compared with the other two regimens.^7–9 In fact, the adverse events induced by gemcitabine plus erlotinib are mostly mild or moderate (grade 1 or grade 2).⁷ Interestingly and more importantly, it seems that those who are thus treated and develop skin rash during the treatment could achieve a significantly longer survival than those without skin rash. For example, Aranda et al¹⁰ found that the median survival with gemcitabine plus erlotinib was 3.3, 6.6, and 10.3 months in patients who developed no rash, grade-1 rash, and ≥grade-2 rash, respectively (P<0.001). Similar results were shown by Moore et al.⁷ In the study of Beveridge et al,¹¹ the benefit was even larger (5.2 vs 12.0 months for rash vs no rash, P=0.025). This relationship was also observed in patients with other cancers treated with erlotinib.^12,13 These findings suggested that skin rash could be a useful marker for predicting the efficacy of gemcitabine plus erlotinib and informing treatment decision.

However, the findings of existing studies regarding the role of skin rash seem inconsistent. For example, in some studies, the median survival time of patients with rash was longer than that of patients without rash,¹¹ while in others, it was the other way round.¹⁴ Some suggested that rash could be a useful marker to inform choice of treatment,^7,10,11 while others disagreed and even stated that “decision for interruption or maintenance of GEM + E, therefore, should not be based on the rash phenomenon”.^14,15 Importantly, many studies had small sample size, and their results were statistically insignificant.^14–20 Could the discrepancy between the studies be explained merely by chance or different sample sizes? Alternatively, could effect modifiers such as some clinical characteristics play important roles, so that skin rash is truly not associated with prolonged survival in some populations? The answer is yet to be explored. Thus, we conducted a systematic review and meta-analysis to synthesize existing evidence on the association of skin rash with clinical outcomes in APC patients treated with gemcitabine plus erlotinib.