This article is mentioned in:

Introduction

Lung cancer is the leading cause of cancer-related mortality worldwide (1). Non-small cell lung cancer (NSCLC) accounts for ~85% of all histological types (2). The treatment plan for patients with advanced NSCLC depends on their genotype, and treatments include targeted therapy, chemotherapy and immunotherapy (3). Immune checkpoint inhibitors (ICIs), such as programmed cell death-1 (PD-1)/programmed cell death-ligand 1 (PD-L1) or cytotoxic T lymphocyte-associated antigen-4 (CTLA-4) antibodies, have been approved for the treatment of patients with advanced NSCLC based on their encouraging efficacy in a series of clinical trials (4,5). Although combination with chemotherapy or antiangiogenic agents improves the clinical outcome of ICIs, a considerable fraction of patients does not benefit from ICIs (6–8). Therefore, it is important to identify biomarkers that can be used to predict the response to ICIs in patients with NSCLC.

At present, PD-L1 expression, tumor mutational burden (TMB) and microsatellite instability (MSI) status have been recognized as reliable indicators to predict the response to ICIs in NSCLC, hepatocellular carcinoma, urothelial carcinoma and other tumors (9,10). As the most important target of immunotherapy, the infiltration, T cell receptor repertoire diversity and gene signatures of CD8+ T cells can also predict the clinical benefit of ICIs treatment in advanced NSCLC (11–13). Additionally, several routine blood parameters, including absolute eosinophil count (14), neutrophil-lymphocyte ratio (15) and lactate dehydrogenase levels (16), have been found to be valuable in the prediction of the clinical outcome of ICIs treatment. However, the role of serum alkaline phosphatase (ALP) in the prediction of ICIs treatment efficacy remains unclear.

ALP is a glycoprotein that catalyzes hydrolytic and phospho-transfer reactions. Elevated serum ALP has been reported in bone and liver-related diseases (17,18). Additionally, serum ALP has been found to be an independent prognostic factor in NSCLC, gastric cancer, breast cancer and other cancer types (19–21). Furthermore, elevated ALP levels have been associated with bone or liver metastasis in patients with lung cancer, prostate cancer, breast cancer and other cancer types (22). In recent studies, bone and liver metastases have been reported to restrain immunotherapy efficacy via regulation of the tumor immune microenvironment (23,24). Therefore, an investigation was conducted to determine whether ALP could be a predictive indicator of ICIs treatment efficacy in patients with NSCLC.

In the present study it was found that bone or liver metastasis was associated with poor prognosis in patients with NSCLC receiving ICIs. Furthermore, elevated serum ALP levels were associated with bone and liver metastases, indicating that ALP levels might be an independent prognostic indicator of ICIs treatment efficacy in patients with NSCLC.

Patients and methods

Patients and data collection

Between January 2018 and December 2020, patients with NSCLC receiving ICIs treatment were investigated at the Institute of Cancer, Xinqiao Hospital, Third Military Medical University (Chongqing, China). Patients who met the following criteria were included in the present study: i) Pathologically diagnosed lung squamous cancer or adenocarcinoma without receiving an operation; ii) not harboring driver gene mutations, including EGFR, ALK, ROS1 and MET; iii) receiving ICIs treatment (including PD-1, PD-L1 or CTLA-4 antibody) for at least 2–3 cycles (6 weeks); iv) complete image examination of evaluable foci pretreatment and post-treatment; and v) complete record of pretreatment ALP levels and complete record of follow-up. Patients with mutations of driver genes, or without records of pretreatment ALP levels, or receiving ICIs treatment <6 weeks were excluded. Overall, 143 patients were included and data regarding their clinical characteristics [age, sex, histology type, TNM stage (according to the 8th AJCC staging system) (25), metastatic sites, treatment lines (first-line or ≥2nd line) and treatment methods] were collected from medical records. The treatment methods included ICIs monotherapy or combined therapy with other treatments (chemotherapy, anti-angiogenic agents or both). The present study was approved by the Ethics Committee of Xinqiao Hospital, Third Military Medical University (approval no. AMUWEC20210386; Chongqing, China). All patients or their legal representatives gave written informed consent for the treatment and inquiries related to the present study.

ALP detection and cut-off value selection

ALP was detected using the endpoint spectrophotometric method in the peripheral blood of patients <1 week before treatment with ICIs, and the ALP level data were collected from medical records. The normal level of ALP is between 0 and 110 U/l (26). In the present study, 110 U/l, which is the normal upper limit of ALP used at Xinqiao Hospital, was selected as the cut-off value to divide patients into high and low ALP level groups.

Follow-up evaluation and definition of progression-free survival (PFS)

Patients were evaluated during follow-up by CT of the lungs and abdomen or MRI of the brain. The response to ICIs treatment was classified according to iRECIST criteria (27) as complete response (iCR), partial response (iPR), stable disease or progressive disease. Treatment response was evaluated every 6 weeks.

The ORR was defined as the proportion of patients with response of iPR and iCR. PFS was defined as the time from the beginning of immunotherapy until recurrence of disease or death due to any cause. Telephone follow-up was performed every 6 months, and the last follow-up date was August 31, 2021.

Statistical analysis

The ALP level data are presented as the median and interquartile range. The mPFS data are presented as Kaplan-Meier curves. The composition of patients or response rate data are presented as n (%), n or the proportion. The results of univariate and multivariate Cox regression analyses are presented as the hazard ratio and 95% CI. The difference in ALP levels was analyzed using an unpaired Student's t-test or one-way ANOVA with a Bonferroni post hoc test. The associations between ALP levels and clinical characteristics were analyzed using a χ2 test and Fisher's exact test if the expected count was <5. The percentages of clinical response of patients to ICIs treatment in two groups were compared using a χ2 test. Kaplan-Meier survival curves and the log-rank test were used to estimate mPFS, and univariate and multivariate Cox regression analyses were used to determine the associations between clinicopathological factors and survival. A forest plot was used to show the results of multivariate Cox regression analysis using Prism GraphPad 9.0 software (GraphPad Software, Inc.). All statistical analyses were conducted using SPSS 20.0 software (IBM Corp.). P<0.05 was considered to indicate a statistically significant difference.

Results

Patient characteristics

A total of 143 patients with NSCLC receiving ICIs were included, and their baseline characteristics are presented in Table I. There were 61 (42.7%) older patients (age ≥60 years), and the mean age was 57.26±9.86 years (range, 35–76 years). Most patients were male (79.0%) and had stage IV disease (87.4%). In addition, 51.7% of patients had squamous cancer, 59.4% of patients received first-line ICIs treatment, and 45.5% of patients were treated with combination therapy of ICIs and other treatments. The median follow-up time was 30.2 months (range, 2.7-52.4 months).

Table I. Baseline characteristics of 143 patients with lung cancer.

Table I.

Baseline characteristics of 143 patients with lung cancer.

Characteristics	No. (%)
Age, years
≥60	61 (42.7)
<60	82 (57.3)
Sex
Male	113 (79.0)
Female	30 (21.0)
Histology type
SC	74 (51.7)
AD	69 (48.3)
TNM stage
III	18 (12.6)
IV	125 (87.4)
Metastatic sites
Bone only	45 (31.4)
Liver only	10 (7.0)
Both bone and liver	17 (11.9)
Neither bone or liver	71 (49.7)
Treatment lines
1st	85 (59.4)
≥2nd	58 (40.6)
Combined therapy
Yes	65 (45.5)
No	78 (54.5)

[i] AD, adenocarcinoma; SC, squamous cancer.

Bone and liver metastases predict poor prognosis of patients with NSCLC receiving ICIs

Previous preclinical and clinical evidence has demonstrated that bone or liver metastases can restrain immunotherapy efficacy (23,24). Therefore, the predictive value of bone or liver metastasis in the prognosis of patients with NSCLC receiving ICIs was investigated. There was a significant association between the response and bone metastasis (P<0.05; Fig. 1A). Furthermore, bone metastasis was associated with a shorter mPFS time in patients with NSCLC receiving ICIs treatment (10.2 vs. 17.3 months; P=0.010; Fig. 1B). Similarly, there was also a significant association between the response and liver metastasis (P<0.05; Fig. 1C). However, liver metastasis was not associated with mPFS time (P=0.119; Fig. 1D).

Figure 1. Bone and liver metastasis restrain efficacy of ICIs treatment. (A) Response to ICIs treatment of patients with NSCLC with or without bone metastasis. (B) mPFS of patients with NSCLC with or without bone metastasis after ICIs treatment. (C) Response to ICIs treatment of patients with NSCLC with or without liver metastasis. (D) mPFS of patients with NSCLC with or without liver metastasis after ICIs treatment. (E) ALP levels in patients with NSCLC with or without bone metastasis and liver metastasis. *P<0.05, **P<0.01. ALP, alkaline phosphatase; ICIs, immune checkpoint inhibitors; iCPD, progressive disease; iPR, partial response; iSD, stable disease; NSCLC, non-small cell lung cancer; mPFS, median progression-free survival.

Elevated ALP levels are associated with bone and liver metastases in patients with NSCLC

ALP has been recognized as a valuable marker for skeletal and hepatobiliary disorders, including bone and liver metastases in patients with cancer (22). Therefore, the relationship between ALP levels and bone or liver metastasis in patients with NSCLC was investigated. ALP levels were higher in patients with bone or liver metastasis than in those without (119.6 or 103.6 vs. 83.3 U/l, respectively; P<0.05; Fig. 1E). Furthermore, the ALP levels of patients with both bone and liver metastases were higher compared with the ‘None’ group (135.7 vs. 83.3 U/l; P<0.01; Fig. 1E). In addition, higher levels of ALP were also associated with bone or liver metastasis, but not with age, sex, stage and other clinical characteristics (Table II).

Table II. Association between clinical variables and serum ALP levels.

Table II.

Association between clinical variables and serum ALP levels.

	Pretreatment ALP levels
Characteristics	≥110 U/l, n	<110 U/l, n	P-value
Age, years			0.809
≥60	16	45
<60	23	59
Sex			0.402
Male	29	84
Female	10	20
Histology type			0.931
SC	21	53
AD	18	51
Stage			0.100
III	2	16
IV	37	88
Bone metastasis			<0.001
Yes	34	28
No	5	76
Liver metastasis			0.036
Yes	10	17
No	29	87
Bone and liver metastasis			<0.001
Both	10	7
Neither	4	67
Treatment lines			0.945
1st	23	62
≥2nd	16	42
Combined therapy			0.304
Yes	15	50
No	24	54

[i] AD, adenocarcinoma; ALP, alkaline phosphatase; SC, squamous cancer.

Elevated ALP levels are an independent prognostic indicator for patients with NSCLC receiving ICIs

To determine the predictive value of ALP, the ALP levels in responders and non-responders to ICIs treatment were first compared. It was found that the levels of ALP were higher in non-responders than in responders (110.1 vs. 88.0 U/l; P<0.01; Fig. 2A). There was a significant association between the response and ALP levels (P<0.05; Fig. 2B). Furthermore, high ALP levels were associated with shorter mPFS (8.5 vs. 15.4 months; P=0.009; Fig. 2C). Additionally, multivariate analysis revealed that ALP level, stage and histology type were independent prognostic indicators for mPFS in the present study (Table III). Patients with high ALP levels or stage IV had a shorter mPFS than those with low ALP levels or stage III, while patients with squamous cancer had a longer mPFS than those with adenocarcinoma (Fig. 3). However, associations between bone metastasis and mPFS were not observed in the multivariate analysis (Table III).

Figure 2. ALP levels predict efficacy of ICIs treatment in patients with NSCLC. (A) ALP levels in responders and non-responders to ICIs treatment. (B) Response to ICIs treatment of patients with NSCLC with or without high ALP levels. (C) Progression-free survival of patients with NSCLC with or without high ALP levels after ICIs treatment. *P<0.05, **P<0.01. ALP, alkaline phosphatase; ICIs, immune checkpoint inhibitors; iCPD, progressive disease; iSD, stable disease; NSCLC, non-small cell lung cancer; iPR, partial response.

Figure 3. Forest plot of univariate analyses of different parameters for progression-free survival. AD, adenocarcinoma; ALP, alkaline phosphatase; HR, hazard ratio; SC, squamous cancer.

Table III. Univariate and multivariate analyses of different parameters for progression-free survival.

Table III.

Univariate and multivariate analyses of different parameters for progression-free survival.

	Univariate			Multivariate
Variables	HR	95% CI	P-value	HR	95% CI	P-value
Sex (male vs. female)	1.274	0.731-2.219	0.417
Age (≥60 vs. <60 years)	1.028	0.653-1.617	0.903
Histology (SC vs. AD)	1.402	0.889-2.209	0.134	0.511	0.302-0.863	0.012
Stage (III vs. IV)	0.343	0.195-0.603	0.006	3.104	1.213-7.944	0.018
Bone metastasis (no vs. yes)	0.563	0.349-0.907	0.010	1.174	0.856-2.641	0.103
Liver metastasis (no vs. yes)	0.658	0.356-1.217	0.119
Treatment lines (1st vs. ≥2nd)	0.898	0.548-1.472	0.662
Combined therapy (yes vs. no)	1.453	0.804-2.262	0.199
ALP (<110 vs. ≥110 U/l)	1.838	1.609-3.191	0.009	1.856	1.030-3.343	0.040

[i] AD, adenocarcinoma; ALP, alkaline phosphatase; HR, hazard ratio; SC, squamous cancer.

Discussion

Immunotherapy, especially ICIs treatment, has been one of the most promising treatment approaches in NSCLC (4,5). ICIs monotherapy or combined therapy has been approved as a first-line treatment and has improved the prognosis of patients with advanced lung cancer (28,29). Although the ORR has been increased by combined therapy independent of the PD-L1 level measured, not all patients benefit from the ICIs treatment, and this is termed a so-called primary resistance (30).

Numerous studies have demonstrated the mechanisms of primary resistance in ICIs treatment (31–33), which include the absence of tumor-specific antigens, downregulation of antigen presentation molecules and suppression of T cell infiltration. Emerging evidence has indicated that bone and liver metastases might induce primary resistance to immunotherapy by regulating the immune microenvironment (23,24). Subgroup analysis of clinical trials revealed that patients with prostate cancer and bone metastasis exhibited primary resistance to ICIs treatment (34). Mechanistically, bone metastatic loci could release TGF-β to restrain Th1 lineage development to further inhibit antitumor immunity (34). Furthermore, a recent study reported that the presence of liver metastasis is associated with poor response to immunotherapy in patients with melanoma or NSCLC, which is consistent with the findings of the present study (23). An animal experiment revealed that liver metastasis could recruit immunosuppressive macrophages to promote antigen-specific T cell apoptosis, resulting in a systemic loss of T cells and diminished immunotherapy efficacy (23). However, a recent meta-analysis reported that patients with lung cancer with and without liver metastasis obtained comparable efficacy after ICIs treatment (35). Therefore, prospective studies might be required to further identify the relationship between liver metastasis and the efficacy of ICIs treatment.

Primary resistance cannot be completely explained by currently recognized biomarkers, such as PD-L1 expression, MSI status or TMB level. Thus, studies are increasingly focusing on other biomarkers to predict the response to ICIs treatment (14,36). As aforementioned, elevated serum ALP levels have been demonstrated to be associated with bone and liver-related diseases. In the present study, it was found that high levels of serum ALP were associated with bone or liver metastasis in patients with NSCLC, which is consistent with previous studies (22,37). Although numerous studies have demonstrated that a high ALP level predicted poor prognosis in various types of cancer, such as NSCLC, gastric cancer and breast cancer (19–21), there is little research on the relationship between serum ALP levels and efficacy of ICIs treatment. In the present study, serum ALP levels were identified as an independent prognostic indicator of ICIs treatment in patients with NSCLC. Recently, it was also reported that high levels of serum ALP predicted poor prognosis in patients with HER-2-negative gastric cancer receiving immunotherapy (38), which might support the findings of the present study. However, the cut-off value of ALP was not the same in the previous study and the present study (225 U/l in the previous study and 110U/l in the present study). The cut-off value in the present study was based on the normal upper limit of ALP, and was consistent with another study (26). Therefore, 110 U/l might be a more suitable cut-off value. There are four isoenzymes of ALP in human serum, including tissue non-specific, germ cell, placental and intestinal ALP, but the standard test could not detect the tissue non-specific isoenzymes, which account for 26–34% of total ALP (22). Thus, whether the undetectable isoenzymes of ALP could also predict the efficacy of ICIs treatment remains to be further studied.

As aforementioned, bone or liver metastasis, and high ALP levels were associated with poor prognosis in patients with NSCLC receiving ICIs treatment in the present study. However, in the present study, it was found that only high ALP levels, rather than bone or liver metastasis, were an independent prognostic factor using multivariate analysis, rather than bone or liver metastasis. This difference indicated that other factors affecting ALP levels, such as benign liver and bone diseases, might also restrain the efficacy of ICIs treatment. Furthermore, it was found that higher ALP levels were associated with a shorter PFS time in patients without bone metastasis (data not shown). However, this hypothesis should be verified in further studies.

There were a number of limitations in the present study. Firstly, this was a retrospective study in a single center with a small number of cases. Secondly, the lack of overall survival data might affect the reliability of the conclusions made. Thirdly, the PD-L1 expression, TMB and MSI status in patients might also affect the outcomes of ICIs treatment (9,10), which should be explored further in the future.

In conclusion, serum ALP levels might be an independent predictor of the response to ICIs treatment in patients with NSCLC. Due to a number of limitations of the present study, prospective studies are required to determine the actual prognostic significance of ALP levels in patients with NSCLC.

Acknowledgements

Not applicable.

Funding

The study was supported by the National Nature Science Foundation of China (nos. 82003006 and 82102878) and Nature Science Foundation of Hainan Province (nos. 821QN384 and 818QN322).

Availability of data and materials

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Authors' contributions

FL and JY conceived and designed the study. KY and TS developed the methodology. TY, JC and SF acquired the data. TY and JC confirmed the authenticity of all the raw data. TY and JC analyzed and interpreted the data. TY and JC wrote the original draft. JY and FL reviewed and revised the manuscript. FL and JY supervised the study. TY, JC and TS were involved in funding acquisition. All authors read and approved the final manuscript.

Ethics approval and consent to participate

The study was approved by the Ethics Committee of Xinqiao Hospital, Third Military Medical University (Chongqing, China). All patients or their legal representatives gave written informed consent for the treatment and inquiries related to this study.

Patient consent for publication

The patients provided written informed consent for publication of any associated data.

Competing interests

The authors declare that they have no competing interests.

Glossary

Abbreviations

Abbreviations:

References