Value of Growth/Differentiation Factor 15 in Diagnosis and the Evaluation of Chemotherapeutic Response in Lung Cancer

Purpose: There is a need for efficient, convenient, and inexpensive methods to accurately diagnose the clinical stage of lung cancer and evaluate the efficacy of chemotherapy in patients with lung cancer. Although growth/differentiation factor 15 (GDF)-15 has great potential as a tumor marker, supporting clinical evidence is still lacking. In this study, we aimed to analyze the relationship between serum GDF15 concentration and the clinical characteristics of patients with lung cancer, and to assess the value of GDF15 in the diagnosis and curative effect of chemotherapy.

Methods: The study comprised 160 participants in total, of whom 88 had lung cancer, 31 had pneumonia, and 41 were control subjects. Among the 88 patients with lung cancer, 64 were willing to participate in follow-up chemotherapy-related studies and meet the inclusion criteria. The serum GDF15 concentration in 288 samples (31 cases, pneumonia group samples; 41 cases, control samples; 88 cases, lung cancer group samples; 64 cases, after 1 chemotherapy cycle; and 64 cases, after 2 chemotherapy cycles) with advanced lung cancer were detected by ELISA. The possible correlations between serum GDF15 level and sex, age, height, weight, body mass index, smoking history, diabetes status, and laboratory findings (hemoglobin, prealbumin, and lactate dehydrogenase) were analyzed using parametric and nonparametric tests. Thereafter, the sensitivity of GDF15 in diagnosing lung cancer was calculated. The serum levels of GDF15, carcinoembryonic antigen (CEA), neuron-specific enolase (NSE), and cytokeratin 19 fragment (CYFRA) 21-1 were determined in 64 patients with lung cancer, before and after chemotherapy reception. For the evaluation of the efficacy of chemotherapy, receiver operating characteristic curves were plotted.

Findings: Serum GDF15 concentration at baseline was significantly higher in the lung cancer group than were those in the pneumonia and control groups (both, P < 0.001). An increased expression of serum GDF15 was significantly correlated with diabetes, anemia, and clinical stage (tumor size, nodal involvement, and presence/absence of metastasis). After 2 cycles of chemotherapy among the 64 patients who received it, serum GDF15 concentrations were significantly different from baseline in those who had progressive disease (P = 0.003), stable disease (P < 0.001), or partial response (P = 0.039). The AUC of GDF15 was greater than those of CEA, NSE, and CYFRA 21-1 (0.851 vs 0.630, 0.720, and 0.654, respectively).

Implications: GDF15 is complementary to CEA, NSE, and CYFRA 21-1 in diagnosing lung cancer and, when used in combination, it could be of great diagnostic value and may facilitate correct predictions of the efficacy of chemotherapy. Therefore, serum GDF15 concentration is valuable in lung cancer diagnosis and in the evaluation of the efficacy of chemotherapy.