Treatment patterns in advanced/metastatic non-small-cell lung cancer in China: results from the CancerMPact^® survey 2021

Abstract

Aim: To report the treatment patterns of advanced/metastatic non-small-cell lung cancer (NSCLC) in China from a physician survey (CancerMPact^®). Materials & methods: A total of 206 Chinese physicians from 27 cities in urban mainland China reported on their treatment of NSCLC in September 2021. Results: Platinum doublets received 70.5% utilization for squamous NSCLC with PD-L1 expression <1% in first-line, whereas nonsquamous NSCLC was treated with platinum doublets (35.2%) or bevacizumab with platinum doublets (35.3%). Checkpoint inhibitors were utilized in >50% of all PD-L1-positive NSCLC cases. Driver-mutated NSCLC was most frequently treated with targeted therapy or platinum-based combinations. Conclusion: NSCLC treatment in China varies by histology, PD-L1 status and driver mutations, illustrating the complexity of decision-making for Chinese physicians as treatment markets expand.

Plain language summary

The most common type of lung cancer is called non-small-cell lung cancer (NSCLC). When lung cancer spreads beyond the lung, it is called advanced. Doctors in China who treat advanced NSCLC were identified. They were surveyed in September 2021 and asked about how they treat their patients. The survey included 206 doctors from 27 cities in China. There are many drugs available for NSCLC. This means that it can be hard for doctors to decide how to treat their patients. The doctors in China often reported using multiple drugs together, instead of using only one drug. One type of drug that can be used to treat NSCLC is called a checkpoint inhibitor (CPI). The doctors reported that they often used CPIs to treat their patients. They also reported that they were more likely to use CPIs made in China rather than CPIs that were made outside of China. Before receiving treatment, most patients were tested for biomarkers. Biomarkers can tell doctors important information about cancers. Doctors can use biomarkers to help decide which treatments to offer their patients. In China, the doctors often did use certain drugs based on patient biomarkers. This choice often depended on the specific biomarker that the patient had. There are many different factors for doctors to consider when treating NSCLC. More and more drugs are becoming available to use in China. While this is good news for patients with cancer, treatment decisions are becoming more complex for doctors.

Summary points

• Physicians in China face complex decisions for their patients with advanced/metastatic stage IV non-small-cell lung cancer (NSCLC), as targeted therapies and immunotherapies have been transforming the treatment landscape.

• This physician survey study found that platinum-based regimens remained the standard of care in first-line for patients with both squamous and nonsquamous PD-L1-negative NSCLC in China in 2021.

• For patients with PD-L1-positive NSCLC, first-line therapy commonly consisted of checkpoint inhibitors (CPIs).

• Chinese physicians reported that combination regimens were used more frequently than CPI monotherapy and that domestically developed CPIs were used more frequently than those developed outside of China.

• Physicians reported that most patients, regardless of NSCLC histology or first-line regimen, received maintenance therapy.

• Most patients were tested for at least one biomarker, typically prior to initiation of first-line therapy, although fewer patients with squamous histology were tested for each biomarker compared with patients with nonsquamous histology.

• Patients most frequently received testing for EGFR mutation and ALK rearrangement; however, physicians reported that over half of their patients (including patients with squamous and nonsquamous histologies) were tested for ROS1 rearrangement, BRAF mutation, RET fusion and MET mutation.

• Targeted therapy was most frequently utilized for patients with EGFR-mutated, ALK-rearranged and ROS1-rearranged NSCLC in first-line, whereas platinum-based regimens were most commonly utilized for patients with BRAF-mutated, RET-fusion-positive and MET-mutated NSCLC.

Keywords: :

• biomarkers
• chemotherapy
• immunotherapy
• lung

Lung cancer is the most common form of cancer in China, with approximately 900,000 new cases diagnosed each year [1]. It is also the deadliest cancer in this country, with approximately 24% of all cancer deaths in China attributed to lung cancer [2]. Whereas in the USA the incidence and mortality rates of lung cancer have been declining along with a decrease in smoking prevalence and an increase in clean air legislative measures, China has observed a gradual increase in lung cancer incidence and mortality rates as smoking prevalence rates also continue to increase and as approximately 70% of urban areas are exposed to air quality that has reached a hazardous level [3,4].

Non-small-cell lung cancer (NSCLC) is the most common histological subtype of lung cancer, accounting for approximately 85% of all lung cancer cases [5]. Treatments currently available for NSCLC are considered noncurative, with the singular exception of complete surgical resection for patients with early-stage disease [^5-8]. In the advanced/metastatic setting, systemic therapy offers only limited improvements in survival, although therapeutic advancements have been extending this benefit for certain subgroups of patients in recent years [8,9]. The identification of biomarkers that have implications for treatment – including EGFR mutations, ALK rearrangements, ROS1 rearrangements, BRAF mutations, RET fusions and MET exon 14 skipping mutations – has spurred the development of targeted therapies which, along with immunotherapies, have been rapidly transforming the advanced/metastatic NSCLC treatment landscape [5,^9-12].

In urban China, patients with metastatic NSCLC (particularly patients with nonsquamous histology) are frequently tested for driver mutations [12]. These driver mutations often have a substantial impact on the course of treatment for Chinese patients. This is illustrated particularly well by the impact of EGFR, with EGFR being considerably more prevalent in China (nearly 40%) than in Western regions (~15% in Europe and 25% in North and South America) [13]. The development and approval of various domestically developed therapeutic agents within China, in addition to the availability of numerous agents developed outside of China, adds further complexity to the decision-making process for Chinese physicians as they make treatment choices for their patients with advanced/metastatic NSCLC [10,12,14]. Data on the treatment patterns of advanced/metastatic NSCLC in urban China are currently limited, and the information that is available becomes rapidly outdated as new treatments are frequently approved, resulting in large shifts in the treatment landscape [15,16].

The goal of this report is to describe the results of a 2021 survey of physicians treating advanced/metastatic stage IV NSCLC in urban China.

Materials & methods

CancerMPact^® (CMP) is a proprietary database from Oracle Life Sciences, which contains oncology epidemiology and cancer treatment data [1]. The treatment data are derived from annual surveys with physicians representing multiple specialties, including oncologists, who treat cancer within four major markets: the USA, Western Europe (WE5) (including France, Germany, Italy, Spain and the UK), Japan and China. In 2021, the CMP surveys were conducted across 7592 physician specialists from within these four markets, including 30 tumor types in the USA and WE5, 27 tumor types in Japan, and 21 tumor types in China. For NSCLC, 640 physicians total were surveyed across the four markets, with Chinese physicians comprising approximately a third of these physician respondents.

The NSCLC CMP survey in China was administered online and recruited 206 Chinese physicians. Participation eligibility included the following: 1) being a board-certified practitioner of medical oncology, surgical oncology, internal medicine, surgery, respiratory internal medicine, respiratory surgery, radiation oncology or oncology department; 2) being a practicing physician for a minimum of 5 years and 3) treating a minimum of ten patients with NSCLC each month. A detailed description of the participating physicians is presented in Table 1.

Table 1. Characteristics of physicians surveyed, China, 2021.

Characteristics of physician respondents	n/%
Number of physicians surveyed	206
Average number of years of practice after residency (range)	17.6 (5–36)
Average number of patients with non-small-cell lung cancer treated by each physician monthly (range)	42.9 (10–500)
Total number of patients with non-small-cell lung cancer treated by these physicians monthly	8834
Board-certified specialty
Medical oncology (%)	46.1
Surgical oncology (%)	1.9
Internal medicine (%)	4.9
Surgery (%)	1.5
Respiratory internal medicine (%)	27.2
Respiratory surgery (%)	1.9
Radiation oncology (%)	0.5
Oncology department (%)	16.0
Hospital levels
Level III (%)	81.6
Level II (%)	9.7
Cancer specialty (%)	8.7
Practice locations
Beijing (%)	20.4
Shanghai (%)	16.0
Guangzhou/Shenzhen (%)	21.4
Other^† (%)	42.2

† ‘Other’ category comprises 23 other practice locations within China, with each of these locations individually accounting for 0.5–3.9% of the total sample.

To support the development of the annual NSCLC treatment survey, Oracle Life Sciences oncology experts conduct a review of treatment patterns that are reported within international guidelines. These sources include reports from the Chinese Society of Clinical Oncology, in addition to new drug approvals by the National Medical Products Administration (NMPA), the national agency within China responsible for the regulation of drugs and medical devices. Literature reviews are also conducted in both English and Chinese language, which include pivotal trial data identified within the peer-reviewed medical literature and from oncology conferences, so that the current treatment practices and anticipated future treatment developments for NSCLC in China can be identified.

For the 2021 NSCLC survey in China, participating physicians were asked questions regarding their professional experience and medical practice characteristics, as well as how they had chosen to treat their patients with NSCLC. Unless specified otherwise, the treatment questions asked physicians to consider how their patients had been treated over the course of the last 6 months. The in-depth questionnaire covered the treatment of NSCLC at all stages of disease and asked questions by histology (squamous and nonsquamous), PD-L1 status and relevant driver mutations in advanced/metastatic stage IV NSCLC (EGFR, ALK, ROS1, BRAF, RET and MET). Respondents reported on the proportion of their patients that were treated with each modality type and combination and the proportion of their patients that they treated with each systemic therapy regimen by clinical situation. For each regimen reported, physicians were also asked to provide the average number of months that their patients actually received each regimen (not the planned number of months). Respondents were asked to limit their responses to only their utilization of Western medicines and treatment modalities and were instructed to not consider the utilization of traditional Chinese medicine when responding to the survey questions.

The online survey was programmed, fielded and hosted by an online-survey company and the anonymized raw data were securely transferred to Oracle Life Sciences, where the data analysis portion of the project was conducted. The survey data are reported here as unweighted averages of all physician responses and no formal statistical tests were conducted. Data presented within this report were obtained from a physician survey fielded in September 2021 with the purpose of describing the treatment of patients with NSCLC in urban China. The data source referenced within this research neither collects nor utilizes patient-level data, or any data that directly involves people, patient medical records or human samples. No patient charts were reviewed by the researchers, who also did not survey patients or otherwise interact with patients in any manner. All data included within has been retrieved from online physician questionnaires regarding their personal treatment practices. Due to the study design outlined above, there was no requirement to seek institutional review board approval.

Results

The CMP treatment surveys were completed by 206 physicians (Table 1) treating patients with NSCLC in urban mainland China. These respondents reported being in practice postresidency for an average of 17.6 years and treating an average of 42.9 patients with NSCLC each month. In aggregate, the physician respondents treated 8834 patients with NSCLC each month. The physicians' reported specialties were varied, with 46.1% specializing in medical oncology and 27.2% specializing in respiratory internal medicine. Most participants (81.6%) reported that they practiced within a level III general hospital, and although physicians were recruited across 27 Chinese cities, over half of the respondents (57.8%) were located in Beijing, Shanghai or Guangzhou/Shenzhen.

Advanced/metastatic stage IV NSCLC with no actionable driver mutations: first-line systemic therapy

NSCLC consists of multiple subtypes that are relevant for treatment and so the respondents were asked questions regarding their utilization of systemic therapy separately by NSCLC type, according to histology, PD-L1 status and common biomarkers. For patients with no actionable driver mutations, the regimens utilized by physicians most often in first-line varied based on both histology and PD-L1 status (Tables 2 & 3).

Table 2. Most commonly used treatments for first-line non-small-cell lung cancer in patients with nonsquamous cell histology and without activating mutations by PD-L1 expression, China, 2021.

Regimen	Nonsquamous NSCLC
	PD-L1 expression <1% or untested		PD-L1 expression 1–49%		PD-L1 expression ≥50%
	Utilization (%)	Average duration in months (range)	Utilization (%)	Average duration in months (range)	Utilization (%)	Average duration in months (range)
Global CPI + chemotherapy	11.6	–	19.3	–	17.1	–
Atezolizumab, cisplatin, pemetrexed	4.7	6.0 (2–15)	6.8	6.3 (2–15)	6.4	6.2 (2–20)
Atezolizumab, carboplatin, pemetrexed	2.7	5.5 (1–12)	4.8	5.6 (2–15)	4.6	5.9 (2–20)
Pembrolizumab, carboplatin, pemetrexed	^†	–	4.1	5.3 (2–12)	3.9	5.4 (1–12)
Pembrolizumab, cisplatin, pemetrexed	2.4	5.6 (1–12)	3.1	6.1 (2–12)	^†	–
Other global CPI + chemotherapy	^†	–	^†	–	^†	–
Domestic CPI + chemotherapy	11.2	–	25.2	–	22.7	–
Camrelizumab, carboplatin, pemetrexed	4.8	5.4 (2–12)	9.6	5.7 (2–18)	8.6	5.5 (2–20)
Camrelizumab, cisplatin, pemetrexed	2.0	5.7 (2–15)	5.4	5.5 (2–18)	5.9	6.0 (1–24)
Sintilimab, carboplatin, pemetrexed	^†	–	3.4	5.1 (2–12)	2.5	4.6 (1–10)
Sintilimab, cisplatin, pemetrexed	^†	–	2.3	5.7 (2–15)	2.4	4.8 (2–10)
Other domestic CPI + chemotherapy	3.0	–	4.5	–	3.3	–
Global CPI monotherapy	–	–	1.2	–	10.1	–
Atezolizumab	–	–	–	–	3.4	7.8 (2–24)
Pembrolizumab	–	–	^†	–	6.7	9.3 (2–24)
Domestic CPI monotherapy	–	–	5.2	–	9.8	–
Camrelizumab	–	–	^†	–	4.5	7.5 (3–20)
Sintilimab	–	–	^†	–	2.9	8.8 (1–24)
Other domestic CPI monotherapy	–	–	^†	–	2.4	–
Bevacizumab + platinum doublet	35.3	–	21.7	–	19.8	–
Bevacizumab, carboplatin, docetaxel	6.5	5.7 (2–12)	3.5	5.2 (1–18)	3.6	5.1 (2–12)
Bevacizumab, carboplatin, paclitaxel	5.7	5.1 (1–12)	5.5	5.4 (2–15)	4.2	5.7 (2–12)
Bevacizumab, carboplatin, pemetrexed	13.8	5.6 (1–24)	7.4	5.2 (1–12)	6.6	5.3 (1–12)
Bevacizumab, cisplatin, pemetrexed	9.2	5.3 (2–14)	5.3	5.3 (2–12)	5.4	5.5 (2–18)
Platinum doublet	35.2	–	22.7	–	16.1	–
Carboplatin, docetaxel	4.2	4.9 (2–10)	2.1	5.1 (2–12)	2.3	5.3 (2–18)
Carboplatin, gemcitabine	3.6	4.8 (1–12)	2.4	5.2 (2–18)	^†	–
Carboplatin, nab-paclitaxel	4.1	5.2 (2–24)	2.3	5.1 (2–18)	^†	–
Carboplatin, paclitaxel	2.5	4.3 (2–12)	2.2	5.0 (2–12)	^†	–
Carboplatin, pemetrexed	4.9	4.8 (1–12)	2.6	5.0 (2–12)	^†	–
Cisplatin, gemcitabine	2.1	5.0 (2–12)	^†	–	^†	–
Cisplatin, paclitaxel	2.4	4.7 (1–12)	^†	–	^†	–
Cisplatin, pemetrexed	6.8	4.8 (2–12)	3.6	5.2 (2–12)	2.4	5.4 (3–18)
Other platinum doublet	4.5	–	4.5	–	3.0	–
Other	6.7	–	4.7	–	4.4	–

Total numbers of responders for each question, from the 206 physicians surveyed: 111 physicians completed data for patients with nonsquamous PD-L1 expression <1% or untested NSCLC, 111 for patients with nonsquamous PD-L1 expression 1–49% NSCLC and 111 for patients with nonsquamous PD-L1 expression ≥50% NSCLC. Data are inclusive of patients with nonsquamous histology with NSCLC that was either wild-type for EGFR and ALK mutations or that was not tested for those biomarkers.

† Less than 2%.

CPI: Checkpoint inhibitor; NSCLC: Non-small-cell lung cancer.

Table 3. Most commonly used treatments for first-line non-small-cell lung cancer in patients with squamous cell histology and without activating mutations by PD-L1 expression, China, 2021.

Regimen	Squamous NSCLC
	PD-L1 expression <1% or untested		PD-L1 expression 1–49%		PD-L1 expression ≥50%
	Utilization (%)	Average duration in months (range)	Utilization (%)	Average duration in months (range)	Utilization (%)	Average duration in months (range)
Global CPI + chemotherapy	4.6	–	10.1	–	7.3	–
Pembrolizumab, carboplatin, nab-paclitaxel	2.1	4.9 (1–12)	5.3	6.6 (2–15)	5.1	6.3 (1–15)
Pembrolizumab, carboplatin, paclitaxel	2.5	4.8 (1–12)	4.8	6.2 (2–15)	2.2	6.5 (2–14)
Domestic CPI + chemotherapy	17.6	–	36.9	–	33.9	–
Camrelizumab, carboplatin, nab-paclitaxel	4.4	5.8 (1–15)	8.4	5.9 (1–15)	4.9	6.1 (2–20)
Camrelizumab, carboplatin, paclitaxel	2.2	4.9 (1–12)	6.5	6.1 (2–15)	6.6	6.1 (2–20)
Sintilimab, carboplatin, nab-paclitaxel	^†	–	6.1	6.2 (2–15)	7.0	5.7 (1–15)
Sintilimab, carboplatin, paclitaxel	2.1	4.3 (1–12)	3.4	5.6 (2–15)	2.6	4.7 (1–15)
Tislelizumab, carboplatin, nab-paclitaxel	3.1	6.0 (1–15)	5.3	6.5 (2–15)	4.1	6.2 (1–15)
Tislelizumab, carboplatin, paclitaxel	^†	–	2.7	5.9 (2–15)	3.8	5.7 (2–15)
Toripalimab, carboplatin, nab-paclitaxel	^†	–	2.5	7.0 (2–15)	2.4	7.4 (2–15)
Toripalimab, carboplatin, paclitaxel	^†	–	2.0	5.7 (2–15)	2.4	5.4 (1–15)
Global CPI monotherapy	–	–	1.3	–	11.6	–
Atezolizumab	–	–	–	–	4.4	8.0 (1–24)
Pembrolizumab	–	–	^†	–	7.2	8.0 (2–24)
Domestic CPI monotherapy	–	–	5.4	–	8.9	–
Camrelizumab	–	–	^†	–	3.1	8.0 (1–20)
Tislelizumab	–	–	^†	–	2.3	7.1 (3–13)
Sintilimab	–	–	^†	–	2.8	6.7 (1–20)
Other domestic CPI monotherapy	–	–	^†	–	^†	–
Platinum doublet	69.8	–	41.7	–	34.0	–
Carboplatin, docetaxel	10.4	6.0 (1–33)	5.4	5.5 (1–12)	5.9	5.4 (1–12)
Carboplatin, gemcitabine	7.4	4.7 (1–12)	4.4	5.0 (2–12)	2.8	5.1 (2–12)
Carboplatin, nab-paclitaxel	12.7	5.0 (1–15)	8.1	5.3 (1–12)	5.4	5.3 (1–12)
Carboplatin, paclitaxel	6.6	4.8 (2–12)	3.1	5.1 (2–18)	4.0	5.1 (2–12)
Cisplatin, docetaxel	7.6	4.9 (1–12)	5.2	5.1 (2–12)	3.9	5.4 (2–12)
Cisplatin, gemcitabine	10.6	4.5 (1–12)	5.6	4.9 (2–18)	3.9	4.9 (2–12)
Cisplatin, paclitaxel	6.4	4.9 (2–12)	4.9	5.4 (2–12)	3.0	4.9 (1–12)
Cisplatin, liposomal paclitaxel	3.4	4.8 (1–12)	^†	–	^†	–
Cisplatin, vinorelbine	2.8	4.3 (2–6)	2.6	4.9 (2–12)	2.6	5.2 (2–12)
Other platinum doublet	^†	–	^†	–	^†	–
Chemotherapy combinations	3.9	–	2.6	–	2.1	–
Gemcitabine, docetaxel	2.0	4.4 (2–6)	^†	–	^†	–
Other chemotherapy combinations	^†	–	^†	–	^†	–
Other	4.1	–	2.0	–	2.2	–

Total numbers of responders for each question, from the 206 physicians surveyed: 111 physicians completed data for patients with squamous PD-L1 expression <1% or untested NSCLC, 110 for patients with squamous PD-L1 expression 1–49% NSCLC and 110 for patients with squamous PD-L1 expression ≥50% NSCLC. Data are inclusive of patients with squamous histology with NSCLC that was either wild-type for EGFR and ALK mutations or that was not tested for those biomarkers.

† Less than 2%.

CPI: Checkpoint inhibitor; NSCLC: Non-small-cell lung cancer.

For nonsquamous NSCLC with PD-L1 expression <1% or untested, platinum-based regimens were reported to be the most frequently utilized regimens for first-line therapy (70.5% total utilization), with these regimens often including bevacizumab (35.3% utilization of bevacizumab with platinum doublets, compared with 35.2% utilization of platinum doublets only). Checkpoint inhibitor (CPI)-based regimens were reported more frequently in patients with nonsquamous histology and PD-L1 positivity. For patients with nonsquamous histology with a PD-L1 expression of 1–49%, CPI-based regimen utilization was 50.9%. This utilization jumped to 59.7% for patients with nonsquamous histology and PD-L1 expression ≥50%. For patients with nonsquamous histology, combination CPI-based regimens were utilized more frequently than CPI monotherapy across all PD-L1 expression levels (PD-L1 <1% or untested: 22.8% combination CPI, 0% CPI monotherapy; PD-L1 1–49%: 44.5% combination CPI, 6.4% CPI monotherapy; PD-L1 ≥50%: 39.8% combination CPI, 19.9% CPI monotherapy). Further, domestically developed CPIs were utilized slightly more frequently than CPIs developed outside of China for patients with PD-L1 positivity (PD-L1 1–49%: 30.4% domestic CPI, 20.5% global CPI; PD-L1 ≥50%: 32.5% domestic CPI, 27.2% global CPI). Although there are several available options for domestically developed CPIs, camrelizumab stood out as the option most frequently utilized (PD-L1 <1% or untested: 6.8%; PD-L1 1–49%: 16.5%; PD-L1 ≥50%: 19.0%). The regimen that was reported to be most frequently utilized for patients with nonsquamous histology and PD-L1 positivity was camrelizumab plus carboplatin/cisplatin and pemetrexed (PD-L1 1–49%: 15.0%; PD-L1 ≥50%: 14.5%).

Similar to patients with nonsquamous histology, the physicians reported that platinum-based regimens were their most common choice for first-line therapy for patients with squamous histology and PD-L1 expression <1% or untested (69.8% utilization of platinum doublets). CPI-based regimens were also more frequently reported in patients with squamous histology who were PD-L1-positive (PD-L1 1–49%: 53.7%; PD-L1 ≥50%: 61.7%) and combination regimens were more frequently utilized than CPI monotherapy across all PD-L1 expression levels (PD-L1 <1% or untested: 22.2% combination CPI, 0% CPI monotherapy; PD-L1 1–49%: 47.0% combination CPI, 6.7% CPI monotherapy; PD-L1 ≥50%: 41.2% combination CPI, 20.5% CPI monotherapy). For patients with squamous histology and PD-L1 positivity, use of domestically developed CPIs was even more common than for patients with nonsquamous histology, with utilization at over 40% (PD-L1 1–49%: 42.3%; PD-L1 ≥50%: 42.8%). Although camrelizumab was the most frequently utilized domestically developed CPI for patients with squamous histology (PD-L1 <1% or untested: 6.6%; PD-L1 1–49%: 16.4%; PD-L1 ≥50%: 14.6%), sintilimab and tislelizumab also received high utilization (PD-L1 <1% or untested: 3.9% sintilimab, 4.9% tislelizumab; PD-L1 1–49%: 10.8% sintilimab, 9.8% tislelizumab; PD-L1 ≥50%: 12.4% sintilimab, 10.2% tislelizumab). The regimen that physicians reported using most frequently for patients with squamous histology and PD-L1 expression 1–49% was camrelizumab plus carboplatin and nab-paclitaxel (8.4%), while the regimen most frequently reported as utilized in patients with squamous histology and PD-L1 expression ≥50% was pembrolizumab monotherapy (7.2%). Total pembrolizumab utilization, including pembrolizumab monotherapy and combination regimens, was 14.5% for these patients with PD-L1 expression ≥50%, compared with an overall utilization of 14.6% for camrelizumab.

Maintenance therapy following first-line therapy

Tables 4 & 5 outline the regimens that physicians reported utilizing in maintenance following first-line therapy, presented both by histology and by the treatments that were initially used in first-line. The physician respondents indicated that most patients, regardless of their histology or the first-line regimen that was received, did continue their treatment with maintenance therapy. Although the use of maintenance therapy did vary within each histology according to the first-line regimen used, patients with squamous histology were overall less likely to receive maintenance therapy than patients with nonsquamous histology. The range in proportion of patients receiving no maintenance following common first-line regimens was 8.9–12.1% for patients with nonsquamous histology compared with 14.5–23.9% for patients with squamous histology.

Table 4. Maintenance regimens for non-small-cell lung cancer in patients with nonsquamous cell histology according to the first-line systemic therapy regimen used, China, 2021.

Maintenance therapy used in nonsquamous non-small-cell lung cancer	Utilization (%)	Average duration of maintenance therapy in months (range)
1: Maintenance after bevacizumab plus platinum doublet
Maintenance bevacizumab monotherapy	26.0	7.6 (1–24)
Maintenance bevacizumab plus pemetrexed	33.8	7.4 (2–24)
Maintenance or consolidation pemetrexed	30.2	7.5 (2–24)
Other	0.3	–
No maintenance or consolidation	9.8	–
2: Maintenance after domestic PD-1 inhibitor (with or without a platinum doublet)
Maintenance or consolidation pemetrexed	21.3	7.0 (2–20)
Maintenance domestic PD-1 inhibitor monotherapy	37.4	8.2 (2–24)
Maintenance domestic PD-1 inhibitor plus pemetrexed	32.4	7.5 (1–24)
No maintenance or consolidation	8.9	–
3: Maintenance after pembrolizumab (with or without a platinum doublet)
Maintenance pembrolizumab	32.9	10.5 (2–66)
Maintenance pembrolizumab plus pemetrexed	29.4	7.4 (2–20)
Maintenance or consolidation pemetrexed	25.5	7.7 (2–20)
No maintenance or consolidation	12.1	–

Total numbers of responders for each question, from the 206 physicians surveyed: question 1: 104 responders; question 2: 86 responders; question 3: 50 responders.

Table 5. Maintenance regimens for non-small-cell lung cancer in patients with squamous cell histology according to the first-line systemic therapy regimen used, China, 2021.

Maintenance therapy used in squamous non-small-cell lung cancer	Utilization (%)	Average duration of maintenance therapy in months (range)
1: Maintenance after platinum doublet
Maintenance or consolidation S-1	18.5	6.1 (2–20)
Maintenance or consolidation gemcitabine plus docetaxel	24.4	6.5 (1–26)
Maintenance or consolidation nab-paclitaxel	0.7	6.0 (6)
Maintenance gemcitabine	31.8	6.3 (1–20)
Maintenance domestic PD-1 inhibitor monotherapy	0.7	10.0 (8–12)
No maintenance or consolidation	23.9	–
2: Maintenance after platinum doublet and pembrolizumab
Maintenance or consolidation S-1	16.2	6.5 (2–15)
Maintenance or consolidation gemcitabine plus docetaxel	15.8	6.3 (2–12)
Maintenance gemcitabine	21.7	6.5 (1–15)
Maintenance pembrolizumab monotherapy	31.8	8.8 (2–24)
No maintenance or consolidation	14.5	–
3: Maintenance after domestic PD-1 inhibitor (with or without a platinum doublet)
Maintenance gemcitabine	30.7	6.3 (2–15)
Maintenance domestic PD-1 inhibitor monotherapy	52.6	7.8 (2–24)
Other	0.2	–
No maintenance or consolidation	16.6	–

Total numbers of responders for each question, from the 206 physicians surveyed: question 1: 103 responders; question 2: 76 responders; question 3: 90 responders.

For patients with nonsquamous histology treated in first-line with bevacizumab-based regimens, continuation maintenance (with all or part of the first-line systemic therapy regimen continued for an extended time period) was utilized most frequently (59.8%). Still, 30.2% of these patients received pemetrexed alone in maintenance, and 9.8% received no maintenance therapy at all. Similarly, patients with nonsquamous histology who received pembrolizumab or domestically developed PD-1 inhibitors with or without a platinum doublet in first-line most commonly continued utilizing a PD-1 inhibitor during maintenance (62.3% utilization for patients who received pembrolizumab in first-line and 69.8% utilization for patients who received a domestically developed PD-1 inhibitor in first-line). While the proportion of these patients receiving no maintenance therapy was also approximately 10% (12.1 and 8.9%, respectively), use of pemetrexed alone as maintenance was somewhat less common (25.5 and 21.3% utilization, respectively).

For patients with squamous histology who received only a platinum doublet in first-line, S-1 (18.5%) and gemcitabine with docetaxel (24.4%) or without (31.8%) were the most commonly utilized regimens. For patients with squamous histology treated with pembrolizumab plus a platinum doublet in first-line, 31.8% continued pembrolizumab as monotherapy in maintenance, while 52.6% of the patients with squamous histology who received a domestically developed PD-1 inhibitor with or without a platinum doublet in first-line continued with the same PD-1 agent as monotherapy in maintenance. While nearly one-quarter (23.9%) of the patients with squamous histology who received only a platinum doublet in first-line did not receive maintenance, the proportion of patients not receiving maintenance was lower for patients with squamous histology who had received pembrolizumab plus a platinum doublet (14.5%) or a domestically developed PD-1 inhibitor with or without a platinum doublet (16.6%) in first-line.

Advanced/metastatic stage IV NSCLC with driver mutations: first-line systemic therapy

The surveyed physicians reported on the proportion of their patients with metastatic NSCLC that received biomarker testing prior to receiving first-line therapy or after initiating first-line therapy, or were never tested (Table 6). Most patients, regardless of histology, were tested for at least one biomarker, although fewer patients with squamous histology were tested for each biomarker than patients with nonsquamous histology. The biomarkers that were tested for most frequently included EGFR mutation (84.7% nonsquamous, 68.6% squamous) and ALK rearrangement (79.1% nonsquamous, 62.7% squamous), although over half of all patients with nonsquamous and squamous histology were tested for the other biomarkers of interest, including ROS1 rearrangement (73.5% nonsquamous, 62.1% squamous), BRAF mutation (74.4% nonsquamous, 60.6% squamous), RET fusion (71.0% nonsquamous, 60.8% squamous) and MET mutation (73.0% nonsquamous, 60.2% squamous). Testing of all biomarkers occurred more frequently prior to first-line therapy rather than following initiation of first-line therapy. In patients with nonsquamous histology, this testing timeframe breakdown was the following for each driver mutation: EGFR mutation (68.7% prior to first-line, 16.0% after first-line), ALK rearrangement (61.5% prior to first-line, 17.6% after first-line), ROS1 rearrangement (54.0% prior to first-line, 19.5% after first-line), BRAF mutation (55.0% prior to first-line, 19.4% after first-line), RET fusion (50.8% prior to first-line, 20.2% after first-line) and MET mutation (54.7% prior to first-line, 18.3% after first-line). In patients with squamous histology, the testing of patients for driver mutations following first-line therapy was comparable to that of patients with nonsquamous histology, while the testing of patients prior to first-line therapy was notably lower for patients with squamous histology: EGFR mutation (49.7% prior to first-line, 18.9% after first-line), ALK rearrangement (43.5% prior to first-line, 19.2% after first-line), ROS1 rearrangement (42.3% prior to first-line, 19.8% after first-line), BRAF mutation (41.8% prior to first-line, 18.8% after first-line), RET fusion (41.0% prior to first-line, 19.8% after first-line) and MET mutation (41.1% prior to first-line, 19.1% after first-line).

Table 6. Biomarker testing rates and the most commonly used treatments for first-line non-small-cell lung cancer in patients with driver mutations, China, 2021.

Biomarkers and regimens	Nonsquamous histology (% of patients with metastatic non-small-cell lung cancer tested)		Squamous histology (% of patients with metastatic non-small-cell lung cancer tested)		Utilization (%)	Average duration in months (range)
	Tested prior to 1L	Tested after 1L	Tested prior to 1L	Tested after 1L
EGFR mutation	68.7	16.0	49.7	18.9
Gefitinib-based regimens					18.2	9.1 (2–48)
Icotinib					11.0	9.6 (2–24)
Osimertinib					14.1	12.7 (2–36)
Bevacizumab-based regimens					28.6	5.9 (2–24)
Platinum doublet regimens^†					20.2	5.6 (2–16)
Other^‡					7.8	–
ALK rearrangement	61.5	17.6	43.5	19.2
Alectinib					17.4	18.2 (3–60)
Ceritinib					5.9	17.2 (3–55)
Crizotinib					29.1	14.2 (4–50)
Bevacizumab-based regimens					21.7	5.7 (2–15)
Platinum doublet regimens^†					21.9	6.7 (2–66)
Other^‡					3.9	–
ROS1 rearrangement	54.0	19.5	42.3	19.8
Ceritinib					7.2	9.9 (3–16)
Crizotinib					52.2	11.4 (3–36)
Bevacizumab-based regimens					20.4	5.8 (2–16)
Platinum doublet regimens^†					17.7	6.6 (2–16)
Other^‡					2.4	–
BRAF mutation	55.0	19.4	41.8	18.8
Dabrafenib, trametinib					25.3	9.5 (3–24)
Bevacizumab-based regimens					41.8	6.8 (2–16)
Platinum doublet regimens^†					28.1	7.4 (3–16)
Other^‡					4.8	–
RET fusion	50.8	20.2	41.0	19.8
Pralsetinib					19.0	8.7 (3–18)
Investigational drug (clinical trial)					6.8	–
Bevacizumab-based regimens					40.8	6.0 (2–18)
Platinum doublet regimens^†					29.8	6.0 (2–18)
Other^‡					3.7	–
MET mutation	54.7	18.3	41.1	19.1
Savolitinib					26.2	10.0 (2–24)
Investigational drug (clinical trial)					5.7	–
Bevacizumab-based regimens					40.2	5.1 (2–12)
Platinum doublet regimens^†					22.6	5.4 (2–12)
Other^‡					5.3	–

Total number of responders for each question, from the 206 physicians surveyed: 201 physicians completed the data for nonsquamous biomarker testing, 206 physicians completed the data for squamous biomarker testing, 107 physicians completed the data for EGFR mutation regimens, 81 physicians completed the data for ALK rearrangement regimens, 47 physicians completed the data for ROS1 rearrangement regimens, 49 physicians completed the data for BRAF mutation regimens, 38 physicians completed the data for RET fusion regimens and 53 physicians completed the data for MET mutation regimens.

† Platinum doublet regimens here include carboplatin/cisplatin in combination with either a taxane or gemcitabine or vinorelbine or S-1 or pemetrexed.

‡ Regimens not individually reported if less than 5% utilization.

1L: First-line therapy.

Patients with advanced/metastatic stage IV NSCLC that harbored an EGFR mutation were most likely to be treated with an EGFR tyrosine kinase inhibitor (TKI)-based regimen. Most of these patients received treatment with TKIs that were developed outside of China (including 18.2% receiving gefitinib-based regimens and 14.1% receiving osimertinib monotherapy), but 11.0% of patients with EGFR mutations were treated with the domestically developed icotinib as monotherapy. Beyond TKI-based regimens, bevacizumab-based combinations were also common in patients with EGFR mutations with first-line utilization at 28.6%. The average duration of therapy for first-line therapy in patients with EGFR mutations varied from 5.6 months (range: 2–16 months) for platinum doublet regimens to 12.7 months (range: 2–36 months) for osimertinib monotherapy.

Among patients with advanced/metastatic stage IV NSCLC determined to have an ALK rearrangement, targeted agents were used most frequently in first-line, with utilization of crizotinib, alectinib and ceritinib at 29.1, 17.4 and 5.9%, respectively. Beyond ALK inhibitors, patients with an ALK rearrangement were also commonly treated with bevacizumab-based regimens (21.7%) and platinum doublet regimens (21.9%). The range of average therapy duration was broader for these patients compared with patients with EGFR mutations, extending from 5.7 months (range: 2–15 months) for bevacizumab-based regimens to 18.2 months (range: 3–60 months) for alectinib monotherapy.

Patients with advanced/metastatic stage IV NSCLC who had a ROS1 rearrangement were also most frequently treated with targeted therapy in first-line. Crizotinib was the clear standard of care in first-line for these patients, with utilization at 52.2% (average duration of use: 11.4 months; range: 3–36 months). Ceritinib was utilized in more limited cases (7.2%; average duration of use: 9.9 months; range: 3–16 months) and a notable proportion of patients were still treated with bevacizumab-based regimens (20.4%; average duration of use: 5.8 months; range: 2–16 months) or platinum doublet regimens (17.7%; average duration of use: 6.6 months; range: 2–16 months).

Among patients with advanced/metastatic stage IV NSCLC harboring BRAF mutations, 69.9% were treated with platinum-based combinations with or without bevacizumab in first-line. Although not specifically approved for BRAF mutation in NSCLC in China, dabrafenib and trametinib combination therapy utilization was 25.3%. The average durations of therapy were similar across regimens, ranging from 6.8 months (range: 2–16 months) for bevacizumab-based regimens to 9.5 months (range: 3–24 months) for dabrafenib and trametinib combination therapy.

Similar to patients with BRAF mutations, 70.6% of patients with advanced/metastatic stage IV NSCLC who had RET fusions were treated with platinum-based combinations with or without bevacizumab in first-line. Recently approved for RET fusion-positive NSCLC in China, pralsetinib utilization was 19.0% for this population. A notable minority of patients with RET fusions were treated with an investigational drug within a clinical trial (6.8%). The average durations of therapy ranged from 6.0 months (range: 2–18 months) for bevacizumab-based regimens and platinum doublet regimens to 8.7 months (range: 3–18 months) for pralsetinib monotherapy.

Patients with advanced/metastatic stage IV NSCLC harboring MET mutations were also most frequently treated with platinum-based combinations with or without bevacizumab (62.8%) in first-line. Recently approved for treating NSCLC with MET exon 14 skipping mutations in China, savolitinib utilization was 26.2%. Beyond these regimens, 5.7% of patients with MET mutations were treated with investigational drug within a clinical trial. The average durations of therapy varied across regimens, ranging from 5.1 months (range: 2–12 months) for bevacizumab-based regimens to 10.0 months (range: 2–24 months) for savolitinib monotherapy.

Discussion

The data presented here outline the recent treatment landscape for advanced/metastatic stage IV NSCLC within urban mainland China, as reported by Chinese physicians treating NSCLC, setting the stage for newly developed NSCLC therapies as they are approved and begin to be adopted within the increasingly competitive Chinese market.

For both patients with nonsquamous histology and squamous histology with PD-L1 expression <1% or untested, platinum-based regimens have remained the standard of care in first-line. In contrast, physicians reported that over 50% of their patients with PD-L1-positive NSCLC (regardless of histology) were treated with CPI-based regimens in first-line. This utilization represents a notable change in physician practice in recent years, with physicians reporting a marked decrease in utilization of platinum-based therapies in favor of immunotherapies, coinciding with the approval and availability of several new domestically developed CPIs in China [12,14]. In this PD-L1-positive setting, physicians reported (across histologies and PD-L1 expression levels) using domestically developed CPIs more frequently than CPIs developed outside of China, with an overall preference for camrelizumab observed. However, sintilimab and tislelizumab were also utilized for patients with PD-L1-positive NSCLC, particularly within the squamous setting. Physicians also reported a higher utilization of combination CPI regimens compared with CPI monotherapy.

The fact that domestically developed CPIs were utilized more often than CPIs developed outside of China can be at least partially attributed to the fact that the domestically developed CPIs had prices that were substantially lower than their international counterparts. Camrelizumab, which enjoyed the highest utilization as reported by the surveyed physicians, was the first domestically developed PD-1 inhibitor to obtain an approved indication in NSCLC in China [17]. Notably, at the time that the CMP survey was fielded, camrelizumab was the only domestically developed CPI that was covered by the National Reimbursement Drug List (NRDL) in China (for nonsquamous NSCLC only) [17]. In contrast, sintilimab and tislelizumab received their approved indication in NSCLC in China only months before the CMP survey was fielded in September 2021, and both had not yet received coverage by the NRDL.

The majority of patients, regardless of their tumor's histology or the regimen that they received in first-line, received maintenance therapy, with physicians most commonly utilizing continuation maintenance. This practice is consistent with international guidelines, which recommend continuation maintenance therapy for patients who have achieved at least stable disease following first-line systemic therapy [18,19]. Approximately 20–30% of patients with nonsquamous histology received pemetrexed alone as maintenance, with this proportion varying by the regimen received in first-line. Pemetrexed is the only systemic therapy recommended as switch maintenance, yet the authors note that it is not possible from the physician survey to determine the proportion of patients that were receiving switch versus continuation maintenance with pemetrexed [18,19].

The physicians reported that the majority of their patients with NSCLC received testing for at least one biomarker, but that fewer patients with squamous histology were tested than patients with nonsquamous histology, likely due to the fact that most driver mutations in NSCLC are less common in patients with squamous histology [20]. Physicians reported that their patients were most frequently tested for EGFR mutations and ALK rearrangements, with over 60% of patients with nonsquamous histology tested for EGFR and ALK prior to receiving first-line therapy, compared with 50% or fewer of patients with squamous histology. Pan-Asian guidelines assert that testing for these biomarkers in patients with squamous histology is optional, yet still recommended in patients with squamous histology who were not heavy smokers [18].

Patients with advanced/metastatic stage IV NSCLC that was EGFR-mutated, ALK-rearranged or ROS1-rearranged were most commonly treated with targeted therapy in first-line. This is a marked change from 1 year prior, when Chinese physicians had reported that targeted therapies were utilized in a third or fewer of their patients with actionable mutations [12]. However, physicians did still report that their patients who had NSCLC that was BRAF mutated, RET-fusion positive and MET mutated were most frequently treated with platinum-based combinations with or without bevacizumab. Although not approved by the NMPA for use in BRAF-mutated NSCLC until March 2022, physicians surveyed in 2021 reported that approximately 25% of their patients with BRAF mutations were already receiving dabrafenib and trametinib combination therapy in first-line, as the regimen was previously approved for melanoma and thus available for use. It remains to be seen if adoption levels will further increase following this specific approval for NSCLC. Similarly, the NMPA approved pralsetinib in March 2021 and savolitinib in June 2021, but these approvals were notably limited. Unlike in the USA and WE5 where pralsetinib was initially approved in first-line, in China pralsetinib received initial approval in second-line for patients with RET-fusion-positive NSCLC who had already received platinum-based chemotherapy. The approval of savolitinib for patients with MET exon 14 skipping mutations was limited to patients with relapsed and refractory NSCLC and to patients who were ineligible for chemotherapy. Although a substantial adoption of these agents was already observed in first-line in patients with RET-fusion-positive NSCLC (nearly 20%) and patients with MET-mutated NSCLC (over 25%), respectively, only time will tell how much additional traction these agents will be able to gain within their respective spaces, particularly as they obtain coverage by the NRDL.

This survey study included inherent limitations. The authors acknowledge that, as a survey study that did not include the review of individual patient charts, the possibility that the results may have been influenced by participant recall bias cannot be eliminated. However, this risk of bias is mitigated by the fact that physicians were asked to limit their responses to a consideration of only the last 6 months. Further, the respondents were limited to reporting only on their own patients and practices, so it is possible that the obtained responses may not be representative of advanced/metastatic stage IV NSCLC treatment patterns across the broader Chinese NSCLC population. However, efforts were made to recruit physicians representatively, placing a focus on including participants from major urban areas in addition to a broad range of regions throughout mainland China. Additionally, it is unknown if certain norms and practices unique to the individual physicians' institutions may have influenced the treatment practices that they reported. This may include access to certain treatments and availability of clinical trials, but as such details were not collected directly within the survey it is difficult to speculate on this potential influence. Similarly, physicians were not asked to justify why they had chosen specific treatments for their patients, so this study is only able to report on the treatment choices that were made and not the thought processes underlying these choices.

The authors note that many patient characteristics may influence physician treatment decisions, including age, performance status and patient comorbidities. Physicians were not asked to specifically consider such factors when reporting on how they treated their patients, and so it is not possible to draw conclusions from the physician survey responses on more nuanced aspects of NSCLC treatment decision-making. It should also be noted that the study relies on the physicians' own clinical interpretations and experiences and that definitions of key oncology terminology were generally not provided within the survey. Further, due to the quantitative nature of the physician survey, the authors are unable to report here on more qualitative aspects, including physician beliefs, knowledge, attitudes and values, which may influence physician prescribing behaviors. The study did not include patient-level data for the purposes of validating the physician responses or explaining the rationale behind the reported treatment decisions. The results from the physician survey were reported here as unweighted means, providing an overall picture of treatment. These results cannot be evaluated with formal statistical analyses.

The authors further acknowledge that this survey study has a cross-sectional design, is descriptive in nature, and can only provide a snapshot view of physician treatment patterns for NSCLC in 2021 in urban China and cannot predict future patterns of utilization. New agents continue to be approved in NSCLC in China and national guidelines continue to be updated regularly, with NSCLC treatment patterns rapidly evolving in China. As one example, at the time that this survey was fielded, NTRK gene fusions, EGFR exon 20 insertions, HER2 mutations and KRAS G12C mutations were not yet actionable biomarkers. Oracle Life Sciences continues to conduct the CMP physician surveys in China on an annual basis in order to closely monitor all new treatment developments. The results of these surveys contribute to the knowledge base for how patients with cancer are currently being treated in China and may be useful to draw upon for identifying opportunities for future research on this topic.

Conclusion

The results of this physician survey study revealed how physicians across mainland China reported treating their patients with advanced/metastatic stage IV NSCLC in 2021. Differences in treatment across histology, PD-L1 status and clinically relevant biomarkers were observed, as well as a rapid uptake of immunotherapies and targeted therapies, particularly domestically developed agents, as they have become available within the country. Physicians in China already face often difficult decisions for how to best treat their patients with NSCLC, and as the global and domestic markets for NSCLC treatment become increasingly competitive, these decisions are likely to only become even more complex.

Financial disclosure

All authors are employed by Oracle Life Sciences, a global consultancy company that acts in the healthcare market, and have as clients pharmaceutical companies, health insurance companies and hospitals. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Writing disclosure

No writing assistance was utilized in the production of this manuscript.

Acknowledgments

The authors would very much like to thank the entire CancerMPact^® team for their tireless efforts in completing the research reflected in this report and the entirety of CancerMPact: K Aguinaga, M Barry, A Boudoures, K Clapp, J Cohn, V Daliparthi, M Epperson, C Fish, M Hasanagic, L Howell, A Hackett, G Kanas, V Kaul, J Lankin, J Little, M Machado, S McClerklin, F Mendez, K Nersesyan, M Ramirez, F Saadeh, G Schuman, P Shah, C Strader, C Su, S Trueman, H Vikis, J Voss, C Wahlman, P Wagner, H Zannit and L Zhao.

Competing interests disclosure

The authors have no competing interests or relevant affiliations with any organization or entity with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.