Sequencing strategies with ramucirumab and docetaxel following prior treatments for advanced non-small cell lung cancer: a multicenter retrospective cohort study

Ramucirumab (RAM) and docetaxel (DOC) are commonly used after first-line therapy for advanced non-small cell lung cancer (NSCLC). Therefore, we aimed to elucidate sequencing strategies of RAM and DOC following prior treatments, including immune checkpoint inhibitor (ICI), cytotoxic agent (CTx) alone, bevacizumab (BEV), and tyrosine kinase inhibitor (TKI). We recruited patients with NSCLC who received RAM and DOC and compared the groups with and without prior ICI, CTx alone, BEV, and TKI, respectively. By tumor response to such treatments, the patients were further classified into “complete response (CR) + partial response (PR),” “stable disease,” and “progressive disease” groups, respectively. We compared RAM and DOC efficacy among these groups. In total, 237 patients were registered. In the group with prior ICI, the objective response rate and disease control rate were significantly higher than those without prior ICI (p = 0.012 and 0.028, respectively), and the median progression-free survival (PFS) was also significantly longer (p = 0.027). There were no significant differences in PFS between the groups with and without CTx alone, BEV, and TKI. Multivariate analysis revealed that prior ICI was an independent factor associated with better PFS. Furthermore, the prior ICI group with CR + PR significantly prolonged PFS compared to the group without prior ICI (p = 0.013). RAM and DOC may be preferably administered after ICI, rather than after CTx alone, BEV, or TKI, and, furthermore, enhanced if the prior ICI has a favorable tumor response.


Introduction
Lung cancer is the leading cause of cancer-related death worldwide [1]. In recent years, immune checkpoint inhibitors (ICIs) have been developed for the treatment of non-small cell lung cancer (NSCLC) regardless of the clinical stage [2]. Such a breakthrough therapy, especially for advanced NSCLC, allows long-term survival for years [3][4][5]. Some patients who have a beneficial treatment response can continue to survive for long periods; however, unfortunately, half of the other patients experience relapse within approximately 1 year. Hence, it is essential to develop second-line treatments.
Satoshi Tanizaki and Kinnosuke Matsumoto contributed equally to this manuscript. Some chemotherapies are currently approved as secondline treatments for advanced NSCLC, including docetaxel (DOC), pemetrexed, and vinorelbine [6][7][8]. Ramucirumab (RAM) is a molecularly targeted monoclonal antibody that directly inhibits vascular endothelial growth factor receptor (VEGFR) 2 [9]. RAM and DOC combination therapy, used as a second-line treatment, significantly extended progressionfree survival (PFS) and overall survival (OS) compared to DOC monotherapy in the REVEL trial [10]. RAM plus DOC is a more promising second-line treatment than chemotherapy alone; however, its efficacy remains limited.
Several researchers have retrospectively analyzed RAM and DOC combination therapy [11][12][13][14][15]. Harada et al. demonstrated that the PFS of patients who received RAM and DOC after ICI was more than twice that of those who did not, showing a significant difference (5.7 months vs. 2.3 months, p = 0.020) [13]. In recent years, some researchers demonstrated the interaction between immunotherapy and anti-angiogenic agents [16]. Therefore, we performed a meta-analysis on the treatment timing and sequence of ICIs and anti-angiogenic agents. As a result, PFS and OS were favorable when anti-angiogenic agents were administered concomitantly with ICI or immediately after ICI therapy [17]. Treatment with anti-angiogenic agents following ICIs may be an even more promising strategy than anti-angiogenic agents alone. Conversely, the prior epidermal growth factor receptor (EGFR) tyrosine kinase inhibitor (TKI) did not affect the efficacy of subsequent cytotoxic agents (CTx) [18]. Thus, treatment sequencing is an important strategy in influencing the subsequent treatment's effectiveness. However, the previously reported studies have limitations; they were conducted only at a few centers and included a small population.
Moreover, a subgroup analysis of the REVEL trial demonstrated that RAM and DOC are appropriate options for the difficult-to-treat population of patients with NSCLC who are refractory to CTx in the first-line setting [19]. Regardless of the effectiveness of prior CTx, RAM and DOC are equally effective. However, the impact of the effectiveness of prior treatments, including ICI, bevacizumab (BEV), and TKI, on RAM and DOC remains unknown.
We conducted a larger multicenter observational cohort study to elucidate the efficacy of RAM and DOC following several treatments, including ICI, CTx alone, BEV, and TKI, and the relationship between tumor responses to the prior treatments and RAM and DOC efficacy in patients with advanced NSCLC.

Study population
This multicenter, observational, retrospective study enrolled previously treated patients with NSCLC who received RAM and DOC between June 2016 and March 2020 at four medical institutions in Japan. This study was conducted in accordance with the Declaration of Helsinki and the World Health Organization's Guidelines for Good Clinical Practice and used a protocol that was reviewed and approved by the ethical review boards and institutional review boards of all participating institutions. This study was reviewed and approved by the Institutional Review Board (IRB) (IRB No. 2020-034). Due to this study's retrospective nature, the requirement for informed consent was waived, and an optout approach was used to allow patients and their families to refuse participation in this study.

Clinical outcomes
Data on age, sex, smoking status, Eastern Cooperative Oncology Group (ECOG) performance status (PS), clinical stage, histology, EGFR mutation, treatment lines, and history of receiving ICIs, CTx alone, BEV, and TKI were obtained from electronic medical records and pharmacy databases. We defined patients who received ICIs immediately before RAM and DOC as the "prior ICI+ " group and those who received treatments except for ICIs immediately before RAM and DOC as the "prior ICI− " group. Moreover, The prior ICI-group was classified into three groups: the "prior CTx+ " group, "prior BEV+ " group, and "prior TKI+ " group, who received CTx alone, BEV, and TKI, respectively, immediately before RAM and DOC. In addition, we defined patients who received ICIs before RAM and DOC in the past as the "past ICI" group and those who had no ICIs before RAM and DOC in the past as the "no ICI" group ( Fig. 1). Data on age, ECOG-PS, and clinical stage were evaluated immediately before the start of RAM and DOC. Clinical responses were defined according to the Response Evaluation Criteria in Solid Tumors version 1.1 [20].
The objective response rate (ORR) included all patients who demonstrated a complete response (CR) and a partial response (PR). Disease control rate (DCR) included patients who demonstrated stable disease (SD) and PR. PFS was defined as the time from the date of RAM and DOC initiation to the date of disease progression or death from any cause. OS was defined as the time from the date of RAM and DOC initiation to the date of death from any cause. According to tumor response to prior treatments, the patients with prior treatment were classified into three groups showing "complete response (CR) + partial response (PR)," "stable disease (SD)," and "progressive disease (PD)," respectively. The group evaluated the efficacy of RAM and DOC. The patients were followed-up until June 30, 2020.

Statistical analyses
Survival analysis was performed using the Kaplan-Meier method and the log-rank test. Cox regression analysis was used to perform univariate and multivariate analyses and to calculate the hazard ratio (HR) with a 95% confidence interval (CI). The Mann-Whitney U test was used for continuous variables, whereas the χ 2 test was used for categorical variables. Statistical analyses were performed using SPSS version 28.0 (IBM, Armonk, NY, USA). Statistical significance was defined as a two-sided p-value of < 0.05.

Patient characteristics
In total, 237 patients who received RAM and DOC were retrospectively assessed. Patient characteristics are summarized in Table 1. The median age was 66.0 years (range, 33-82 years), and some older (≥ 70 years) patients were included as well (32.1%). Most patients were men (63.3%), smokers (75.5%), with ECOG-PS 0-1 (79.7%). The majority of patients had adenocarcinoma (75.5%), and approximately 30% of patients had sensitive EGFR mutations. About 60% of patients received RAM and DOC as second-or third-line treatment.
With regard to prior treatments, 94 (39.7%) patients received ICI immediately before RAM and DOC (prior ICI+ group), whereas 143 (60.3%) patients received treatments except for ICI immediately before RAM and DOC (prior ICI− group) (Supplementary Table 1). According to tumor responses to prior ICI, PD (44 patients) was the most common, followed by SD (32 patients) and CR + PR (15 patients). In the prior ICI-group, 69 (29.1%) patients received CTx alone, 46 (19.4%) patients received BEV, and 28 (11.8%) patients received TKI immediately before RAM and DOC. Finally, RAM was not used with any regimen except for DOC.

Treatment efficacy by tumor response to prior ICI
In the prior ICI+ group showing CR+PR, SD, and PD, the ORRs and DCRs were 42.9% and 85.7%, 21.9% and 68.8%, and 38.6% and 70.5%, respectively. The ORR and DCRs in these three groups in the prior ICI+ were all higher than those in the prior ICI− group, although the number of patients differed in each (Table 4).
Moreover, in the prior ICI+ groups showing CR + PR, SD, and PD, and the prior ICI− group, the median PFS was 11.6 (95% CI, 6.2-17.0), 6.6 (95% CI, 4.5-8.6), 4.4 (95% CI, 2.2-6.6), and 3.9 (95% CI, 2.9-4.9) months, respectively (Fig. 3). Compared with the prior ICI− group, who received treatments except for immunotherapies immediately before RAM and DOC, there was a significant difference in the median PFS in the CR + PR group by the log-rank test (p = 0.013). In contrast, there were no significant differences in those of the SD and PD groups.

Discussion
In this study, we provided the largest data set concerning the efficacy of RAM and DOC in patients with advanced NSCLC in clinical practice. To the best of our knowledge,  28 (11.8) this is the first study to elucidate the impact of prior treatment and its effectiveness on RAM and DOC efficacy. Our findings demonstrated that only prior ICI+ was an independent factor significantly associated with better PFS in patients with NSCLC receiving RAM and DOC. Furthermore, we demonstrated that tumor response to prior ICI We demonstrated that prior ICI+ significantly increased anti-tumor responses and extended PFS in patients who received RAM and DOC compared to prior ICI− , whereas past ICI did not. This may be attributed to the duration from the final administration of ICIs to RAM and DOC initiation.
In recent years, some clinical studies have shown that combination therapy with ICIs and anti-angiogenic agents improves tumor immune response. Several studies have investigated this mechanism. The tumor microenvironment (TME) is composed of the tumor, immune cells, and several signaling molecules and contains several immunosuppressive cells, such as regulatory T cells [21][22][23][24][25]. VEGF signaling can promote the immune suppression of the antitumor immune response in the TME and induces resistance to immunotherapy by promoting the differentiation and proliferation of these suppressor cells [26,27]. VEGF also suppresses T cell infiltration into tumors by decreasing the expression of adhesion molecules, such as integrin ligand, which is termed tumor endothelial cell allergy [28,29]. Therefore, blocking VEGF signaling decreases the number of immunosuppressor cells in the TME and increases the number of effector T cells that can directly eliminate tumor cells. Thus, anti-angiogenic agents restructure the immunosuppressive TME and enhance immunotherapy efficacy.
In addition, Osa et al. evaluated the duration of antiprogrammed death-1 (PD-1) antibody binding to T cells and revealed that the bindings continued for more than 20 weeks after the last ICI administration and then the bindings gradually uncoupled [30]. Other clinical reports have also demonstrated that ICI efficacy sustains for years after discontinuation [31].
Based on these reports, in the prior ICI+ group, ICIbound T cells can remain in the blood for a while, reactivated by RAM's administration, and infiltrate tumors again. The failure to show superiority in the past ICI group compared to the no ICI group may be due to the long duration from the final administration of ICIs to RAM and DOC initiation.

3
In recent years, clinical trials similar to our concept, using anti-angiogenic agents immediately after ICI tolerance, have been reported successively [32,33]. In the future, sequential therapy with ICI and anti-angiogenic agents will become an optimal treatment option, and further comparative studies are required to validate our findings.
We also demonstrated that tumor response to prior ICI might predict the survival benefits of RAM and DOC. Tumeh et al. demonstrated that in patients with melanoma, following anti-PD-1 therapy, the number of tumor-infiltrating CD8 + T cells in ICI responders increased but did not in non-responders [34]. Hence, in conjunction with the report of Osa et al., anti-PD-1 antibodies bind to existing T cells for a while after immunotherapy discontinuation, and the more effective the ICIs are, the more such T cells may be present in the blood. Such anti-PD-1 antibody-bound T cells, different from finally differentiated exhausted T cells that have completely lost their proliferative potential and are refractory to immunotherapy, are memory-like exhausted T cells that retain their proliferative potential and can differentiate into T cells with anti-tumor effector activity [35]. In short, in the prior ICI+ group compared to the prior ICI− group, pre-existing ICI-bound T cells with anti-tumor activity may infiltrate into and eliminate tumors again with enhancement by RAM.
This study has some limitations. First, some biases were inevitable because of the retrospective design of the present study. In fact, as selection biases, the patients for whom ICI should be avoided as a prior treatment are more likely to have rapid tumor progression and complications such as interstitial pneumonia and collagen disease. However, confounding factors were adjusted using multivariate analyses as much as possible. Second, the therapeutic efficacy by clinical response to prior ICI therapy was evaluated in a relatively small number of patients. Third, data on the treatment lines after RAM and DOC were not analyzed. Therefore, the OS may not have been accurately evaluated. Fourth, the efficacy of RAM and DOC immediately after each treatment was not directly compared to that of the placebo in this study. Therefore, it is difficult to elucidate whether the prior treatment actually affects RAM and DOC more than the other treatments. Finally, this study evaluated efficacy alone; safety was not assessed. In clinical practice, treatment decisions should be based on efficacy and safety. Further comparative studies are warranted to validate our findings.
In conclusion, prior ICI+ significantly increased antitumor responses and prolonged PFS in patients who received RAM and DOC compared to prior ICI− , whereas past ICI did not. Therefore, RAM and DOC should be administered immediately after ICI tolerance. In the group with prior ICI, ORR, DCR, and PFS were significantly higher than those without prior ICI, whereas there were no significant differences in PFS between the groups with and without CTx alone, BEV, and TKI. Prior ICI+ showing CR + PR significantly prolonged PFS in patients with RAM and DOC compared with prior ICI− . Thus, RAM and DOC may be preferably administered after ICI, rather than after CTx alone, BEV, or TKI, and, furthermore, enhanced if the prior ICI has Fig. 3 Kaplan − Meier curves of PFS compared among the prior ICI + showing CR + PR, SD, and PD, and prior ICI − groups in patients with NSCLC who received RAM and DOC. PFS, progression-free survival; ICI, immune checkpoint inhibitor; CR, complete response; PR, partial response; SD, stable response; PD, progressive disease; NSCLC, non-small cell lung cancer; RAM, ramucirumab; DOC, docetaxel; CI, confidence interval 1 3 a favorable tumor response. Further studies are warranted to confirm the results of this study.