This was the first study indicating that a blood-based protein signature (VS) might be a novel and valid method to predict the efficacy of pemetrexed/platinum or bevacizumab in first-line treatment for Chinese advanced lung adenocarcinoma patients.
The concept of combination therapy refers to using the best combination of existing treatment methods according to the specific conditions of patients, such as body condition, pathological type, invasion range (pathological stage) and development trend, to greatly improve the cure rate, prolong the survival period and improve the quality of life of patients. Surgery, radiotherapy, and systemic therapy are the 3 modalities of combination therapy for cancer patients. Systemic therapy, including chemotherapy, targeted therapy and immunotherapy, is suitable for the treatment of advanced NSCLC patients. Recently, although significant progress has been made in immunotherapy and targeted therapy, chemotherapy remains the cornerstone for advanced NSCLC patients[13-20]. Pemetrexed is a third-generation cytotoxic agent. It can inhibit cell replication and tumour growth by disrupting folate-dependent normal cellular metabolism. Several clinical trials demonstrated that first-line pemetrexed-based therapy was associated with significantly better survival outcomes than other chemotherapy regimens for advanced lung adenocarcinoma patients. In addition, the combination of bevacizumab could further improve survival for these patients. Therefore, pemetrexed/platinum combined with bevacizumab has been the standard first-line chemotherapy regimen for advanced non-squamous NSCLC patients. However, some patients still showed poor response to pemetrexed-based regimens[21, 22]. Currently, due to the lack of clinical evidence, no specific biomarkers have been applied in clinical practice. Therefore, promising biomarkers are urgently needed to predict the efficacy of cytotoxic agents.
Protein polypeptide distributions varied significantly in advanced lung adenocarcinoma patients' serum, and these protein polypeptides could be detected in different molecular weight ranges by using MALDI-TOF-MS. Eight characteristic peaks in the serum were found to be related to the efficacy of chemotherapy. The height and area of characteristic peaks in the mass spectra of patients with good efficacy were lower, while those with poor efficacy were significantly higher. According to the location, height and area of characteristic peaks, data models were established to predict the efficacy. After comparing the sensitivity and specificity of different models, the VeriStrat test was the model with the best sensitivity and specificity. As a blood-based test, VS could effectively divide patients into either good efficacy (VS-G) or poor efficacy (VS-P) groups. In 2007, David Carbone et al initially used the VeriStrat method to predict the first-line efficacy of EGFR tyrosine kinase inhibitors (TKIs) for advanced NSCLC patients who did not receive EGFR mutation tests before treatment. In this study, the median survival was 306 days in the VS-G group of 69 patients, far more than the 107 days in the VS-P group of 27 patients[4]. A series of follow-up studies demonstrated that VS is a predictor of therapeutic benefit from EGFR TKI therapy[23]. In the PROSE study, the VS method was utilized to predict second-line single-drug chemotherapy for advanced lung cancer (pemetrexet/docetaxel). The results show that among the 129 patients receiving single-drug chemotherapy, the OS and PFS were significantly lower in the VS-P group than those in the VS-G group [24]. Another study examined the performance of VeriStrat in three independent clinical trials from 481 patients treated with platinum-based chemotherapy as a first-line treatment. Patients classified as VS-G had significantly longer PFS and OS than VS-P patients. These results demonstrated that VS is a strong predictive test in NSCLC patients treated with platinum-based regimens in the first line[10]. However, to date, there has been no study using this VS method for the prediction of first-line pemetrexed plus platinum-based chemotherapy or combined with bevacizumab for Chinese advanced lung adenocarcinoma patients.
To address these issues, we conducted a retrospective analysis of plasma samples from lung adenocarcinoma patients with stage IIIB or IV disease who received first-line pemetrexed-based chemotherapy. Our study showed that median PFS was unreached in the VS good group and was significantly superior to that in VS poor group, the PFS of which was 4.2 months. For the 35 patients who received chemotherapy, an improved PFS was still observed for patients in the VS-G vs. VS-P group (median PFS: Unreached vs. 4.0 months). A recent study included 76 non-squamous patients treated with a combination of carboplatin or cisplatin with pemetrexed. Patients classified as VS-G had longer PFS and OS than VS-P: 6.5 vs 1.6 months and 10.8 vs 3.4 months, respectively.[11] The PFS in our study was longer than that found in previously reported data, likely because pemetrexed maintenance therapy was administered in our study. The PARAMOUNT study demonstrated that continuation maintenance therapy with pemetrexed was an effective and well-tolerated treatment option for patients with advanced non-squamous NSCLC with good performance status who did not progress after induction therapy with pemetrexed plus cisplatin[25]. In the PARAMOUNT study, the median PFS was 4.1 months for pemetrexed and 2.8 months for placebo[26]. The result in our study was consistent with that in the PARAMOUNT study.
In the POINTBREAK study, PFS was significantly improved with pemetrexed/carboplatin plus bevacizumab and pemetrexed/bevacizumab maintenance therapy (median PFS, 6.0 vs 5.6 months; P = 0.012)[27]. In the AVAPEAL study, bevacizumab plus pemetrexed maintenance was also associated with a significant PFS benefit compared with bevacizumab alone (median, 3.7 v 7.4 months; P < 0.001)[28]. Updated survival analysis of the AVAPERL study showed that maintenance with bevacizumab-pemetrexed was associated with a nonsignificant increase in OS over bevacizumab alone[29]. A recent study (WJOG 5610L) reported in 2019 that the American Society of Clinical Oncology annual meeting demonstrated that bevacizumab and pemetrexed maintenance therapy could not prolong OS compared with bevacizumab maintenance therapy alone (median, 23.3 vs 19.6 months; P=0.069). In our study, for 37 patients treated with chemotherapy and bevacizumab, PFS was also significantly longer in the VS-G group than in the VS-P group (median PFS: Unreached vs. 4.8 months). Our study indicated that VS is also predictive for chemotherapy and bevacizumab combined therapy. Although this study clearly identified patients who might have worse outcomes on pemetrexed-based therapy, these data were not compelling enough to deny pemetrexed therapy to VS-P patients. However, perhaps in these VS-P patients, alternative treatment approaches could be considered.
Several limitations of our study are worthy of note. First, 72 patients were eligible for inclusion and analysis, and only a small subset of participants (12/72, 16.7%) tested as VS-P. This limited the power of the analysis we performed. Furthermore, OS data were not available. Because OS was typically not calculated until more than 50% of patients experienced events, median OS was not reached in either group. Third, the sample number was small, so it was difficult to avoid selective bias. However, we took a series of measures to reduce bias. Patients were consecutively included in this study to avoid selective bias. Patients' baseline characteristics were compared using the T test for age and the X2 test for all other characteristics. We found that patient characteristics were well balanced between the VS-G and VS-P groups. In addition, we used the Cox proportional hazards model to conduct multivariate analysis. After adjusting for gender, stage, smoking status and treatment by multivariate analysis, the interaction between PFS and VS classification was also statistically significant.