Impact of docetaxel plus ramucirumab on metastatic site in previously treated patients with non-small cell lung cancer: a multicenter retrospective study.

BACKGROUND: Docetaxel (DOC) plus ramucirumab (RAM) has been recommended as an optimal therapy for previously treated patients with non-small cell lung cancer (NSCLC). In a clinical setting, there are few reports about DOC plus RAM, therefore its effect on factors such as Eastern Cooperative Oncology Group (ECOG) performance status (PS) and metastatic sites is still unknown.
METHODS: We recruited NSCLC patients who received DOC plus RAM in four medical facilities in Japan from June 2016 to March 2020. We retrospectively investigated the overall response rate (ORR), disease control rate (DCR), and progression-free survival (PFS) of DOC plus RAM and conducted univariate and multivariate analyses using PFS as a dependent factor. Patients were followed up until June 30, 2020.
RESULTS: A total of 237 patients were consecutively enrolled. For all patients, the ORR, DCR, and median PFS were 25.2%, 63.9%, and 4.5 months, respectively. The ORR and DCR for malignant pleural effusion (MPE), lung metastasis, and liver metastasis were 7.7% and 53.8%, 30.3% and 77.5%, and 48.6% and 71.4%, respectively. In the multivariate analysis, MPE, lung metastasis, and liver metastasis were not prognostic factors for poor PFS. However, ECOG-PS 2 or more [hazard ratio (HR): 1.66, 95% confidence interval (CI): 1.14-2.40, P=0.008] and brain metastasis (HR: 1.71, 95% CI: 1.23-2.37, P=0.001) were significant and independent factors associated with shorter PFS.
CONCLUSIONS: DOC plus RAM could be an optimal therapy for previous treated NSCLC patients with lung and liver metastasis, and furthermore, should be used carefully for patients with poor ECOG-PS or brain metastasis. KEYWORDS: Docetaxel and ramucirumab; non-small cell lung cancer (NSCLC); metastatic site; poor performance status. 2021 Translational Lung Cancer Research. All rights reserved.

Introduction

Non-small cell lung cancer (NSCLC) is a common cause of cancer mortality worldwide (1). A majority of patients with NSCLC have a metastatic disease at diagnosis, for which no curative therapy exists. Although treatments for driver mutations have yielded impressive improvements in many regions, most patients relapse despite using breakthrough tyrosine kinase inhibitors such as osimertinib (2). Actually, a majority of the patients do not have sensitive driver mutations associated with approved targeted drugs, and platinum-based chemotherapies plus immune checkpoint inhibitors (ICIs) are the standard first-line treatment for NSCLC patients with good performance status in recent years (3-5). Although some patients have long-term survival after using platinum-based chemotherapy combined with ICIs, most patients experience disease progression after administration.

Some clinically approved second-line therapies for NSCLC patients include docetaxel (DOC), pemetrexed, and vinorelbine (6-8). Particularly, DOC plus ramucirumab (RAM) significantly increased the progression-free survival (PFS), overall survival (OS), and complete overall response rate (ORR) when directly compared to DOC monotherapy in the REVEL trial (9). Therefore, DOC plus RAM has been recommended as an optimal therapy for previously treated patients in many countries.

Although a few reports have been published concerning the retrospective analysis of DOC plus RAM as a therapeutic regimen, such as evaluating the effect of age and previous ICIs usage, analyses of the effect of clinical factors, such as Eastern Cooperative Oncology Group (ECOG) performance status (PS) on survival are still limited (10-13).

Currently, two anti-vascular endothelial growth factor (VEGF) antibodies have been clinically used for NSCLC patients: bevacizumab (BEV) and RAM. BEV is a humanized monoclonal antibody (mAb) that targets all isoforms of VEGF-A, which prevents the activation of the membrane receptors such as vascular endothelial growth factor receptor (VEGFR) 1 and VEGFR2. RAM is a fully human antagonistic mAb which targets VEGFR2. By binding to VEGFR1 and VEGFR2, VEGF-A induces inflammatory responses and disrupts cell-to-cell connections to increase vascular permeability (14). BEV has been effective in patients with malignant pleural effusion (MPE) by blocking the binding of VEGF-A to its receptors (15-17). Liver metastases also benefit from BEV because of the overexpression of VEGF-A in liver lesions (4,18-21). However, the clinical efficacy of RAM for these metastatic sites is still unknown.

Therefore, we performed a multicenter retrospective observational cohort study for previously treated patients with NSCLC who were administered DOC plus RAM to identify the clinical factors which affect its efficacy and to elucidate the relationship between metastatic sites and efficacy.

We present the following article in accordance with the STROBE reporting checklist (available at http://dx.doi.org/10.21037/tlcr-20-1263).

Methods

Study population

This study was a multicenter, observational, retrospective study. Participants consisted of previously treated NSCLC patients who received DOC plus RAM between June 2016 and March 2020 in four medical facilities in Japan, including the Osaka International Cancer Institute, the Osaka Habikino Medical Center, the Osaka General Medical Center, and the National Hospital Organization Kinki-Chuo Chest Medical Center. This study was conducted in accordance with the Declaration of Helsinki (as revised in 2013) and the World Health Organization’s Guidelines for Good Clinical Practice, and the protocol was reviewed and approved by the Institutional Review Board at the Osaka International Cancer Institute (IRB No.20098), the Osaka Habikino Medical Center (IRB No.1046), the Osaka General Medical Center (IRB No.2020-060), and the National Hospital Organization Kinki-Chuo Chest Medical Center (IRB No. 2020-034). The patient’s informed consent was waived for the retrospective nature of study, and we used an opt-out method so that patients and families could refuse to participate in the study.

Data collection

We collected data on age, sex, smoking status, ECOG-PS, histology, EGFR mutation, and metastatic site (MPE, lung, liver, or brain) from electronic medical records and pharmacy databases. Data on age, ECOG-PS, and metastatic site were evaluated just before the start of DOC plus RAM. Clinical responses were defined according to the Response Evaluation Criteria in Solid Tumors, version 1.1 (22).

The ORR included all patients with complete response (CR) and partial response (PR). The disease control rate (DCR) included the patients with stable disease (SD) or PR. We defined the anti-tumor effect of MPE as follows: PR was defined as the status with an obvious decrease of MPE without thoracentesis and SD was defined as the status without an unequivocal increase of MPE at the evaluation after more than 6 weeks from initiation of DOC plus RAM compared to baseline. The PFS was determined from the date of commencing DOC plus RAM therapy to the date of disease progression or death from any cause. The OS was determined from the date of commencement of the relevant therapy to the date of death. Patients were followed until June 30, 2020.

Statistical analysis

The survival outcomes were analyzed using the Kaplan-Meier method. The differences between patient groups according to each factor were compared using the log-rank test. P values <0.05 were regarded as statistically significant. Cox regression analysis was used to calculate the hazard ratio (HR) of each factor with 95% confidence interval (CI). To identify prognostic factors (such as age, sex, smoking status, ECOG-PS, and metastatic site), univariate and multivariate analyses were conducted. χ2 test was used for categorical variables. Statistical analyses were performed using SPSS version 26.0 (IBM, Armonk, NY).

Results

Patient characteristics

A total of 237 previously treated patients with NSCLC were included in this study. Patient characteristics are summarized in . The median age was 66.0 years (range, 33–82 years), and a few elderly (≥75 years) patients (12.7%) were included. A majority of the patients were men (63.3%), had adenocarcinoma (75.9%), had a history of smoking (75.5%), and had an ECOG-PS of 0–1 (79.7%). Sixty-six patients (29.5%) had sensitive EGFR mutations. Approximately 60% of patients received DOC plus RAM on the early line such as 2nd or 3rd line. In addition, 40% of patients received immunotherapy as the front-line treatment (Table S1). RAM was not used with any regimens except for DOC. Of the investigated metastatic sites, lung metastasis (100 cases) was the most detected site, followed by MPE (71 cases), brain metastasis (60 cases), and liver metastasis (38 cases).

Table 1

Clinical characteristics of 237 patients

Characteristics	N (%)
Median age [range], years	66 [33–82]
Age group, years
<75	207 (87.3)
≥75	30 (12.7)
Sex
Male	150 (63.3)
Female	87 (36.7)
Histology
Adenocarcinoma	180 (75.9)
Squamous cell carcinoma	38 (16.0)
Other	19 (8.1)
Smoking status
No	58 (24.5)
Yes	179 (75.5)
ECOG-PS
0	28 (11.8)
1	161 (67.9)
2	40 (16.9)
3	6 (2.5)
4	2 (0.9)
EGFR mutation
Positive	66 (29.5)
Negative	158 (70.5)
Median number of prior treatments (range)	2.0 (1.0–11.0)
Malignant pleural effusion	71 (30.1)
Lung metastasis	100 (42.2)
Liver metastasis	38 (16.0)
Brain metastasis	60 (25.4)

ECOG-PS, Eastern Cooperative Oncology Group-performance status; EGFR, epidermal growth factor receptor.

Treatment efficacy in all patients

In all patients, the ORR, DCR, and progressive disease rate were 25.2%, 63.9%, and 36.1%, respectively. At the time of analysis, the PFS was based on 195 events (82.3% maturity), whereas the OS was based on 148 events (62.4% maturity). The median follow-up time in censored patients was 12.5 months (range, 3.1–42.6 months). The median PFS was 4.5 (95% CI, 3.36–5.62) months () and the median OS was 13.4 (95% CI, 11.7–15.1) months ().

Figure 1

Kaplan-Meier survival curves of PFS and OS in all patients treated with DOC plus RAM. (A) The median PFS was 4.5 (95% CI, 3.36–5.62) months and (B) the median OS was 13.4 (95% CI, 11.7–15.1) months. PFS, progression-free survival; OS, overall survival; DOC, docetaxel; RAM, ramucirumab; CI, confidence interval.

The log-rank test revealed no differences in the median PFS of patient treated with DOC plus RAM in terms of age (<75 vs. ≥75) (), sex (male vs. female) (), and smoking status (non-smoker vs. smoker) (). Conversely, the median PFS in ECOG-PS groups (0–1 vs. 2–4) were significantly different [5.46 (95% CI: 4.42–6.51) vs. 2.79 (95% CI: 2.18–3.39) months; P<0.001] ().

Figure 2

Kaplan-Meier survival curves of PFS. There were no significant differences in the median PFS of patient treated with DOC plus RAM in terms of (A) age, (B) sex, and (C) smoking status; (D) Poor PS group had shorter PFS compared to good PS group significantly [2.79 (95% CI: 2.18–3.39) vs. 5.46 (95% CI: 4.42–6.51) months; P<0.001]. PFS, progression-free survival; DOC, docetaxel; RAM, ramucirumab; PS, performance status; CI, confidence interval.

Treatment efficacy by metastatic site

The ORR and DCR are listed by metastatic site in . The ORR of DOC plus RAM for MPE was relatively low, and DOC plus RAM seemed to be unable to decrease pleural effusion. However, about 50% of patients were able to prevent an increase in pleural effusion. The ORR and DCR for lung and liver metastasis were higher than those in all patients, showing 30.3% and 77.5%, and 48.6% and 71.4%, respectively. The ORR and DCR for brain metastasis were almost the same as those of all patients. Of 60 patients with brain metastases, 52 patients (86.7%) received palliative treatment in addition to DOC plus RAM. Among these, five patients underwent surgery and 48 received radiation therapy, with one patient receiving both therapies.

Figure 3

The ORR and DCR on each metastatic site. The ORR for MPE was 7.7%, relatively low. The ORR and DCR for lung and liver metastasis were higher than those in all patients, showing 30.3% and 77.5%, and 48.6% and 71.4%, respectively. The ORR and DCR for brain metastasis were almost the same as those of all patients. ORR, overall response rate; DCR, disease control rate; MPE, malignant pleural effusion.

The log-rank test revealed no differences in the median PFS of patients who received DOC plus RAM concerning MPE (positive vs. negative) (), lung metastasis (positive vs. negative) () and liver metastasis (positive vs. negative) (). The median PFS of patients with brain metastasis (positive vs. negative) was significantly different [3.14 (95% CI: 2.38–3.90) vs. 5.14 (95% CI: 4.25–6.03) months; P=0.002] ().

Figure 4

Kaplan-Meier survival curves of PFS. There were no significant differences in the median PFS of patient treated with DOC plus RAM in terms of (A) MPE, (B) lung metastasis, and (C) liver metastasis; (D) patients with brain metastasis had shorter PFS compared to patients without brain metastasis significantly [3.14 (95% CI: 2.38–3.90) vs. 5.14 (95% CI: 4.25–6.03) months; P=0.002]. PFS, progression-free survival; DOC, docetaxel; RAM, ramucirumab; MPE, malignant pleural effusion; CI, confidence interval.

Association between patient characteristics and progression free survival

The results of the Cox proportional hazards model for predicting the PFS in patients treated with DOC plus RAM are shown in . In the univariate analysis, ECOG-PS (0–1 vs. 2–4, HR: 1.76, 95% CI: 1.24–2.50, P=0.002) and brain metastasis (negative vs. positive, HR: 1.63, 95% CI: 1.19–2.25, P=0.003) were significantly associated with poorer PFS. Similarly, in the multivariate analysis, ECOG-PS (0–1 vs. 2–4, HR: 1.66, 95% CI: 1.14–2.40, P=0.008) and brain metastasis (negative vs. positive, HR: 1.71, 95% CI: 1.23–2.37, P=0.001) were identified as independent factors significantly associated with poorer PFS.

Table 2

Univariate and multivariate analysis of PFS in patients treated with DOC plus RAM

Factor	Univariate analysis			Multivariate analysis
	HR (95% CI)	P		HR (95% CI)	P
Age, years
<75	1			1
≥75	0.98 (0.63–1.51)	0.914		1.00 (0.64–1.57)	0.992
Sex
Male	1			NA
Female	0.83 (0.62–1.10)	0.197		NA	NA
ECOG-PS
0–1	1			1
2–4	1.76 (1.24–2.50)	0.002		1.66 (1.14–2.40)	0.008
Smoking status
Non-smoker	1			NA
Smoker	1.31 (0.95–1.83)	0.103		NA	NA
MPE
(−)	1			1
(+)	1.22 (0.89–1.67)	0.215		1.13 (0.82–1.55)	0.458
Lung metastasis
(−)	1			1
(+)	1.01 (0.76–1.35)	0.922		0.99 (0.74–1.32)	0.928
Liver metastasis
(−)	1			1
(+)	1.40 (0.97–2.03)	0.073		1.34 (0.90–1.98)	0.145
Brain metastasis
(−)	1			1
(+)	1.63 (1.19–2.25)	0.003		1.71 (1.23–2.37)	0.001

PFS, progression-free survival; DOC, docetaxel; RAM, ramucirumab; ECOG-PS, Eastern Cooperative Oncology Group-performance status; MPE, malignant pleural effusion; HR, hazard ratio; CI, confidence interval.

In the univariate and multivariate analyses for predicting the OS, ECOG-PS (0–1 vs. 2–4, HR: 2.02, 95% CI: 1.37–2.97, P=0.001) and MPE (negative vs. positive, HR: 1.63, 95% CI: 1.15–2.31, P=0.007) were significant poor prognostic factors, whereas brain metastasis was not consistently identified as a poor prognostic factor of OS (Table S2).

Discussion

Our study provides the most comprehensive set of data regarding the efficacy of DOC plus RAM in previously treated patients with NSCLC in a clinical setting. To the best of our knowledge, this is the first study that elucidates the relationship between metastatic site and therapeutic efficacy. Our findings demonstrate that poor ECOG-PS (≥2) and brain metastasis were independent factors significantly associated with a shorter PFS in patients who received DOC plus RAM. Furthermore, no differences in the PFS between patients with and without MPE and lung metastasis were found. There are several possible explanations as to why these differences in the PFS were observed in our study.

Although there are no reports on the efficacy of DOC plus RAM for patients with poor ECOG-PS, there are various reports about the efficacy and safety of DOC alone in ECOG-PS 2 patients. These studies have made its use controversial because of the significant increase in febrile neutropenia (23-25). Although the 2018 National Comprehensive Cancer Network Guidelines recommend DOC plus RAM treatment in previously treated patients with ECOG-PS 0–2 with NSCLC, ECOG-PS 2 patients were not included in the REVEL trial and there is no evidence that demonstrates the efficacy and safety in ECOG-PS 2 patients (26). We first showed that poor ECOG-PS was a significant and independent factor associated with a shorter PFS and OS. Although the number of treatment cycles was not collected, there was no significant difference in the discontinuation rate due to adverse events between good and poor PS patients. Nevertheless, the shorter survival of poor PS patients may result from the tumor burden at the start of DOC plus RAM. Therefore, further investigations on poor PS populations are needed.

The most common sites of metastatic disease in patients with NSCLC have been reported to be the bone, lungs, liver, brain, and adrenal glands (27,28). The MPE, occurring in approximately 15.0% of patients with NSCLC, also affects patient management and quality of life and is a prognostic factor for poor survival (29,30). NSCLC patients with MPE or liver metastasis have a poorer prognosis than patients without MPE or liver metastasis (14,30,31). Brain metastasis is also considered a poor prognostic factor in NSCLC patients; therefore, patients with brain metastasis have often been an excluded from clinical trials (32). BEV has been reported to be effective for metastatic sites, such as the MPE, lung, liver, and brain by binding to serum VEGF-A and inhibiting the binding of VEGF-A and VEGFR (14-18,20,33,34). Although RAM also inhibits VEGF-A/VEGFR signaling using a method similar to BEV (i.e., by binding to VEGFR2), the relationship between these metastatic sites and RAM efficacy is unknown.

Our study revealed that brain metastasis was a factor associated with poor PFS. This is similar to previously reported data that patients with brain metastasis have a poorer prognosis than those without brain metastasis (34). Subsequently, the blocking of VEGFR2 alone may be a weak preventive strategy against angiogenesis in the brain. According to existing research, in tissues with high vascular density, such as the brain, tumors grow not only by angiogenesis, but also by vessel co-option, which is a non-angiogenic process through which tumor cells utilize pre-existing tissue blood vessels. However, the same study also shows that blocking VEGF-A resulted in the undetectability of tumors (35). These observations may contribute to the difference of efficacy between BEV and RAM in treating brain metastasis in mouse models. Although the reason brain metastasis was not a poor prognostic factor of OS remains unclear, the number of prior treatments and the treatment regimen before and after administration may be involved.

Although there were no differences in the PFS and OS between patients with and without lung and liver metastasis in our study, the ORR and DCR for lung and liver metastasis were higher than those for all patients. There are several reasons as follows: VEGFR2 is constitutively phosphorylated, especially in the liver, lungs, and kidneys in vivo, and highly expressed in surgically resected lung specimen of NSCLC (36,37). It has been reported that VEGFR2 antibody inhibited the growth of lung metastasis of renal cell carcinoma by 26% (38). VEGFR2 is also up-regulated in patients with hepatocellular carcinoma and induces angiogenesis with consecutive endothelial growth in the hepatic sinusoids (39). Clinically, RAM was approved for patients with hepatocellular carcinoma who were previously treated with sorafenib on the basis of the REACH-2 trial results in 2019 (40). Although further studies are required to determine whether DOC plus RAM improves the PFS and OS in patients with lung or liver metastasis, according to our data on anti-tumor effects, the therapy may be effective for patients with rapid progression of lung or liver metastasis. Recently, some studies have showed the efficacy of combination therapies of DOC and anti-angiogenic agents after ICIs. Brueckl et al. demonstrated the high response rate and long-term survival of DOC plus RAM followed immediately after ICIs (41). Grohé et al. also revealed the combination therapy of DOC plus nintedanib, triple angiokinase inhibitor after ICIs showed the clinical benefit in the prospective study (42). Our study included 40% of patients who received immunotherapy as the previous therapy, which may have affected the ORR and DCR for lung and liver metastasis.

MPE was associated with the low response and significantly shortened the OS. Generally, VEGF-A/VEGFR2 signaling increases vascular permeability and promotes pleural effusion retention. VEGF-A/VEGFR1 signaling may also contribute to tumor-associated angiogenesis (43). Unlike BEV, RAM does not block the VEGF-A/VEGFR1 pathway. As a result, MPE may hardly have been decreased, which may have affected poor OS. However, about 50% of patients were able to prevent an increase in pleural effusion, and therefore MPE was likely not a significant poor prognostic factor of PFS. Furthermore, regarding the clinical efficacy of MPE, results of prospective studies such as the PLEURAM study are expected in the future (44).

This study has several limitations. First, biases are inevitable due to the retrospective study design, including selection bias. However, confounding effects were adjusted for using multivariate models. Second, data on first-line and further lines of treatments before DOC plus RAM were not analyzed. Therefore, the OS may not have been evaluated accurately. Third, the efficacy of DOC plus RAM on metastatic sites was not directly compared to that of DOC monotherapy in this study. Therefore, it is difficult to elucidate the efficacy of RAM alone for each metastatic site. Further comparative studies are needed to validate our findings.

In conclusion, a poor ECOG-PS and brain metastasis were independently associated with poor PFS in patients with NSCLC treated with DOC plus RAM in a clinical setting. Careful monitoring is required for patients with these factors. Although DOC plus RAM did not prolong the PFS of patients with lung metastasis, and liver metastasis, the ORR and DCR for lung and liver metastasis were higher than those of all patients. DOC plus RAM could be an optimal therapy for patients with previously treated NSCLC who have lung and liver metastasis. Further studies are warranted to confirm the results of our study.

The article’s supplementary files as