Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Open Access

Peer-reviewed

Research Article

Depression in association with neutrophil-to-lymphocyte, platelet-to-lymphocyte, and advanced lung cancer inflammation index biomarkers predicting lung cancer survival

  • Barbara L. Andersen ,

    Contributed equally to this work with: Barbara L. Andersen, John Myers, Peter G. Shields, William E. Carson

    Roles Conceptualization, Investigation, Methodology, Project administration, Supervision, Writing – original draft, Writing – review & editing

    Affiliation Department of Psychology, Ohio State University Comprehensive Cancer Center, The Ohio State University, Columbus, Ohio, United States of America

    ⨯
  • John Myers ,

    Contributed equally to this work with: Barbara L. Andersen, John Myers, Peter G. Shields, William E. Carson

    Roles Data curation, Formal analysis, Methodology, Visualization, Writing – original draft, Writing – review & editing

    Affiliation Department of Biomedical Informatics and Center for Biostatistics, College of Medicine, The Ohio State University, Columbus, Ohio, United States of America

    ⨯
  • Tessa Blevins ,

    Roles Data curation, Visualization, Writing – review & editing

    ‡ These authors also contributed equally to this work

    Affiliation Department of Psychology, Ohio State University Comprehensive Cancer Center, The Ohio State University, Columbus, Ohio, United States of America

    ⨯
  • Kylie R. Park ,

    Roles Investigation, Resources, Validation, Visualization, Writing – review & editing

    ‡ These authors also contributed equally to this work

    Affiliation Department of Psychology, Ohio State University Comprehensive Cancer Center, The Ohio State University, Columbus, Ohio, United States of America

    ⨯
  • Rachel M. Smith ,

    Roles Data curation, Resources, Validation, Visualization, Writing – review & editing

    ‡ These authors also contributed equally to this work

    Affiliation Department of Biomedical Informatics and Center for Biostatistics, College of Medicine, The Ohio State University, Columbus, Ohio, United States of America

    ⨯
  • Sarah Reisinger ,

    Roles Data curation, Investigation, Project administration, Supervision, Writing – review & editing

    ‡ These authors also contributed equally to this work

    Affiliation Department of Internal Medicine, Division of Medical Oncology, College of Medicine, Ohio State University Comprehensive Cancer Center, The Ohio State University, Columbus, Ohio, United States of America

    ⨯
  • David P. Carbone ,

    Roles Funding acquisition, Writing – review & editing

    ‡ These authors also contributed equally to this work

    Affiliation Department of Internal Medicine, Division of Medical Oncology, College of Medicine, Ohio State University Comprehensive Cancer Center, The Ohio State University, Columbus, Ohio, United States of America

    ⨯
  • Carolyn J. Presley ,

    Roles Funding acquisition, Writing – review & editing

    ‡ These authors also contributed equally to this work

    Affiliation Department of Internal Medicine, Division of Medical Oncology, College of Medicine, Ohio State University Comprehensive Cancer Center, The Ohio State University, Columbus, Ohio, United States of America

    ⨯
  • Peter G. Shields ,

    Contributed equally to this work with: Barbara L. Andersen, John Myers, Peter G. Shields, William E. Carson

    Roles Funding acquisition, Supervision, Writing – review & editing

    Affiliation Department of Internal Medicine, Division of Medical Oncology, College of Medicine, Ohio State University Comprehensive Cancer Center, The Ohio State University, Columbus, Ohio, United States of America

    ⨯
  • William E. Carson

    Contributed equally to this work with: Barbara L. Andersen, John Myers, Peter G. Shields, William E. Carson

    Roles Conceptualization, Investigation, Methodology, Writing – original draft, Writing – review & editing

    william.carson@osumc.edu

    Affiliation Department of Surgery, Division of Surgical Oncology, College of Medicine, Ohio State University Comprehensive Cancer Center, The Ohio State University, Columbus, Ohio, United States of America

    ⨯

Abstract

Lung cancer is a product of inflammation and a dysfunctional immune system, and depression has similar dysregulation. Depression disproportionately affects lung cancer patients, having the highest rates of all cancers. Systemic inflammation and depression are both predictive of non-small cell lung cancer (NSCLC) survival, but the existence and extent of any co-occurrence is unknown. Studied is the association between systemic inflammation ratio (SIR) biomarker levels and patients’ depressive symptoms, with the hypothesis that depression severity would be significantly associated with prognostically poor inflammation. Newly diagnosed stage-IV non-small cell lung cancer (NSCLC; N = 186) patients were enrolled (ClinicalTrials.gov Identifier: NCT03199651) and blood draws and depression self-reports (Patient Health Questionnaire-9) were obtained. For SIRs, cell counts of neutrophils (N), lymphocytes (L), and platelets (P) were abstracted for ratio (R) calculations for NLR, PLR, and the Advanced Lung cancer Inflammation Index (ALI). Patients were followed and biomarkers were tested as predictors of 2-year overall survival (OS) to confirm their relevance. Next, multivariate linear regressions tested associations of depression with NLR, PLR, and ALI. Overall 2-year mortality was 61% (113/186). Cox model analyses confirmed higher NLR [hazard ratio (HR) = 1.91; p = 0.001] and PLR (HR = 2.08; p<0.001), along with lower ALI (HR = 0.53; p = 0.005), to be predictive of worse OS. Adjusting for covariates, depression was reliably associated with biomarker levels (p ≤ 0.02). Patients with moderate/severe depressive symptoms were 2 to 3 times more likely to have prognostically poor biomarker levels. Novel data show patients’ depressive symptoms were reliably associated with lung-relevant systemic inflammation biomarkers, all assessed at diagnosis/pretreatment. The same SIRs were found prognostic for patients’ 2-year OS. Intensive study of depression, combined with measures of cell biology and inflammation is needed to extend these findings to discover mechanisms of depression toxicity for NSCLC patients’ treatment responses and survival.

Citation: Andersen BL, Myers J, Blevins T, Park KR, Smith RM, Reisinger S, et al. (2023) Depression in association with neutrophil-to-lymphocyte, platelet-to-lymphocyte, and advanced lung cancer inflammation index biomarkers predicting lung cancer survival. PLoS ONE 18(2): e0282206. https://doi.org/10.1371/journal.pone.0282206

Editor: Tomasz Urbanowicz, Poznan University of Medical Sciences Faculty of Medical Sciences: Uniwersytet Medyczny im Karola Marcinkowskiego w Poznaniu Wydzial Medyczny, POLAND

Received: September 21, 2022; Accepted: February 7, 2023; Published: February 24, 2023

Copyright: © 2023 Andersen et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data are within the paper and its Supporting Information files.

Funding: This study was funded by the Ohio State University Comprehensive Cancer Center and Pelotonia through grants awarded to PS and DC. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: BLA, JM, TB, KRP, RMS, SR, and WEC have no competing interests to declare. DPC reports personal fees from Abbvie, Adaptimmune, Agenus, Amgen, Ariad, AstraZeneca, Biocept, Boehringer Ingelheim, Celgene, Clovis, Daiichi Sankyo, Inc. (DSI), EMD Serono, Flame Biosciences, Foundation Medicine, G1Therapeutics/Intellisphere, GenePlus, Genentech/Roche, Glaxo-Smith-Kline, Gloria Biosciences, Gritstone, Guardant Health, Humana, Incyte, Inivata, Inovio, Janssen, Kyowa Kirin, Loxo Oncology, Merck, MSD, Nexus Oncology, Novartis, Oncocyte, Palobiofarma, Pfizer, prIME Oncology, Stemcentrx, Takeda Oncology, and Teva; and grants and personal fees from Bristol Myers-Squibb (BMS) outside the submitted work during the conduct of the study. This does not alter our adherence to PLOS ONE policies on sharing data and materials.

Introduction

Immunotherapy and targeted treatments are yielding significantly longer survival for patients with advanced (stage IV) non-small cell lung cancer (NSCLC) [1–3], but it is unknown why a substantial portion of patients fail to respond to treatment. Causes are multifactorial, but an unexplored contributor to premature death may be toxic, biobehavioral (psychological, behavioral, biological) characteristics of co-morbid depression.

Several lines of evidence point to the plausibility of depression influencing disease progression. In the general case, depression assessed at diagnosis is prognostic for mortality in cancer patients [4], with the strongest effects found in lung cancer [5]. Moreover, the continuing trajectory of depressive symptoms—from diagnosis through two years—predicts overall NSCLC survival [6]. Discovery of such effects is made possible, unfortunately, because of the high incidence of depression in NSCLC patients. Of all cancer patients, those with lung cancer are among the most impaired [5,7,8], with an estimated 36% of NSCLC with moderate to severe depressive symptoms at diagnosis, many of whom also have comorbid anxiety [9], a clinical presentation more resistant to psychological treatment [10].

In addition to psychologic aspects, depression’s biologic dysregulation may contribute to disease progression. Lung cancer is a product of a dysfunctional immune system, as evidenced by tobacco/smoking-induced inflammation [11], an inflammatory tumor microenvironment [12], and robust, systemic inflammatory responses (SIRs) [12,13]. In this context, there is “overwhelming” evidence [14] that severe depressive symptoms and major depressive disorder (MDD) covary with elevations of proinflammatory cytokines [15], decreased adaptive immune responses [16–19], and others [20]. Further, biologic therapies which increase inflammation (e.g. interferon-alfa treatment) can cause MDD [21,22]. Thus, studies suggest that inflammation with depression co-occurs with that arising from lung cancer.

Study of the health consequences of depression in lung cancer patients is timely as new therapies come on line and treatment guidelines rapidly change [23]. Psychosocial studies of lung patients have been few and come largely from prior decades of chemotherapy-only treatments [24], making examination association of depressive symptoms and systemic inflammation responses (SIRs) novel. Also important is discovery of the nature and biologic mechanisms of depressive symptom toxicity; depression is a rate limiting factor impacting patients’ quality of life and, potentially, treatment response.

To do this, biomarkers of inflammation were studied in relationship to depressive symptoms. Systemic inflammatory responses (SIRs)—neutrophils, lymphocytes, monocytes, and platelets—are key factors in proliferation, angiogenesis, immunosuppression, and nutritional depletion [25]. Briefly, the inflammatory response is characterized by increases in circulating neutrophils (N) accompanied by falls in circulating lymphocytes (L). The neutrophil-to- lymphocyte ratio (NLR) is viewed as reflecting the inflammatory imbalance of pro-tumor efficacy (N) and anti-tumor capacity (L) of the host [25,26]. The platelet-to-lymphocyte ratio (PLR) is important as platelet elevation accelerates tumor progression [27,28]. Unique is the advanced lung cancer inflammation (ALI) index, which considers nutritional status with NLR. SIRs are prognostic biomarkers for multiple tumor types and multiple meta analyses have confirmed elevated NLR [29] and PLR [29] and lower ALI [29] levels at diagnosis to predict NSCLC overall survival (OS).

Newly diagnosed NSCLC patients (N = 186) enrolled in an observational cohort (NCT03199651) participated. Procedures included routine laboratory work and patient-reported depressive symptom assessment [Patient Health Questionnaire-9 (PHQ-9)] [30,31]. First, NLR, PLR, and ALI were validated as predictors of OS at 2-years. If confirmed, associations of PHQ-9 scores with SIRs were next examined, with the hypothesis that depression severity would be significantly associated with higher NLR and PLR and lower ALI levels. Individually, patients with moderate to severe depressive symptoms were expected to have prognostically poor levels of systemic inflammation in comparison to patients without depressive symptoms. If confirmed, the findings would provide support for biobehavioral aspects of depression potentially “adding” to the inflammatory dysregulation already present in those with advanced NSCLC.

Materials and methods

Study design

A prospective cohort design was used. An NCI-designated Comprehensive Cancer Center enrolled advanced (stage IV) NSCLC patients at diagnosis (NCT03199651) and followed patients up to 5 years, withdrawal, or death. Accrual occurred from July 2017 to February 2020.

Procedures

A university institutional review board approved the study and procedures in accordance with ethical standards and the 1964 Helsinki declaration. Inclusion criteria were as follows: newly diagnosed with pathology confirmed stage IV NSCLC (any type); any Eastern Cooperative Oncology Group (ECOG) performance status; any illness/condition comorbidity; age ≥ 18 years; English-speaking; willingness to provide access to medical records, provide biospecimens, and respond to patient reported outcome (PRO) measures. Exclusion criteria were as follows: prior treatment with definitive chemo-radiotherapy for lung cancer; diagnosis >90 days prior to enrollment; receipt of lung cancer treatment for over one month; presence of disabling hearing, vision, or psychiatric conditions (e.g., psychosis) preventing consent or completion of PROs in English. Also excluded were second opinion or consult only cases (patients returning to his/her community physician for treatment).

Research personnel completed informed consent face-to-face (verbal, with written materials) during a patient’s first clinic appointment with a thoracic oncologist and written informed consent was obtained and the signature witnessed. Cell count and albumin data for biomarker calculations were retrieved from electronic medical record reports. Lab data on or immediately after the patients’ first visit (mean = 11.82 days) were used. Interviewers from a professional survey center contacted patients by telephone and assessed PROs, for which patients received $15. Study flow is provided (Fig 1).

thumbnail
Download:
Fig 1. Study flow.

https://doi.org/10.1371/journal.pone.0282206.g001

Measures

Depressive symptoms.

The American Society of Clinical Oncology’s recommended measure was used, the Patient Health Questionnaire-9 (PHQ-9) [30,31]. Nine items assessing symptoms of major depressive disorder (MDD) are rated on a Likert frequency scale, ranging from 0 = not at all to 3 = nearly every day. Items are summed for a total score ranging from 0 to 27. Symptom level classifications are as follows: 0–7 (none/mild), 8-14 (moderate), 15–19 (moderate to severe), and 20–27 (severe). Cronbach’s alpha was .80.

Systemic inflammation biomarkers.

1) Neutrophil-to-lymphocyte ratio (NLR) was calculated by dividing a patient’s absolute neutrophil count by the absolute lymphocyte count. 2) Platelet-to-lymphocyte ratio (PLR) was calculated by dividing platelet count (109/L) by the absolute lymphocyte count. 3) Advanced lung cancer inflammation index (ALI) was calculated by multiplying body mass index (BMI) by the quotient of albumin (g/dL) and NLR (i.e., ALI = BMI × Albumin/NLR), where BMI = weight (kg)/height (m)2. Considering methods and findings from prior studies and meta analyses of SIRS as predictors of OS, the following accepted cutoffs were used for the survival analyses: 5 for NLR [32,33], 185 for PLR [34], and 24 for ALI [35].

Covariates.

Variables considered were as follows: age, sex, race, partner status, education level, employment status, and smoking history. Also considered were lung cancer type, performance status, and first line treatment to be received as abstracted from medical records. The Eastern Cooperative Oncology Group Performance Status (ECOG PS) [36] score was assigned by the treating physician. ECOG PS can range from 0 to 5: 0 = fully active; 1 = restricted in physically strenuous activity but ambulatory and able to carry out light work, 2 = ambulatory and capable of all self-care, but unable to work, 3 = capable of only limited self-care, confined to bed or chair more than 50%, 4 = completely disabled, confined to bed or chair, and, 5 = deceased.

Power

A priori power analyses estimated the sample size necessary to detect a significant association of PHQ-9 scores, whether measured as continuous or categorical, with SIRs at baseline. Based on a significance level of alpha = .05, a sample size of 91 would yield an estimated power equal to 0.80 in linear regression with the six parameters and a medium effect size (Cohen’s F2 = 0.15). Sample size N = 186 translates to power >.95. This sample size is equivalent to or larger than those in meta analyses of the relationship between these biomarkers and NSCLC OS [29].

Statistical analysis

Demographic and disease characteristics of the sample were summarized and scores on depression and biomarker ratio levels and cutoffs provided. In order to avoid overfitting the survival models, covariates for analysis were selected based on a cutoff of p < 0.20 from univariate models S1 and S2 Tables. Additionally, smoking status and treatment type were included in survival models to account for their effect on survival. Demographic variables such as age, sex, race, and marital status were selected as covariates for multiple linear regressions. Baseline ECOG Performance Status was also selected to serve as a proxy for patient fitness. A cut-off of p < 0.20 was used to select additional covariates for multiple linear regressions. Cox proportional models were used to assess the association between biomarkers as defined by cutoff levels and overall survival (OS), adjusting for relevant covariates.

Next, multiple linear regressions tested the association of patient depressive symptoms (PHQ-9) with elevated NLR, elevated PLR, and lower ALI, adjusting for relevant covariates. NLR, PLR, and ALI were log-transformed due to right-skewed distributions. Descriptive analyses show the percentage of patients with no/low versus high depressive symptoms stratified by low versus high SIR levels. A p-value of 0.05 was used as a cutoff for significance. Analyses were performed using R 4.1.1. [37].

Results

Sample description

Baseline demographics, smoking status, lung cancer type, and performance status for N = 186 are provided (Table 1). Characteristic of advanced NSCLC patients, the sample was older (mean 63 years), unemployed (74.7%), previously/currently smoking (84.9%), with the majority having either adenocarcinoma or squamous histologic subtype (89.8%). Patients were diagnosed a median of 34 days prior to the baseline assessment. The majority of patients were subsequently treated (91%), with chemotherapy alone (30%), immunotherapy alone (22%), chemotherapy+immunotherapy (19%), targeted therapy for mutations (e.g., EGFR, ALK) (16%), targeted therapy chemotherapy (2%), or other treatment (1%).

thumbnail
Download:
Table 1. Demographic and disease characteristics of NSCLC sample (N = 186).

https://doi.org/10.1371/journal.pone.0282206.t001

The median value for the PHQ-9 was 5.5 (Q1 = 3, Q3 = 9), corresponding to a mild level of depressive symptoms for the sample. Using classifications for individuals, 64.5% of patients reported depressive symptoms at the none/mild level (n = 120), 27.4% at the moderate level (n = 51), 4.8%% at the moderate to severe level (n = 9), and 3.2% at the severe level (n = 6).

SIR biomarker median and quartiles (Q1, Q3) were as follows: NLR 5.39 (Q1 = 3.33, Q3 = 10.70), PLR 224.45 (Q1 = 149.89, Q3 = 341.84), and ALI 18.24 (Q1 = 9.17, Q3 = 34.50). Using the cutoff values (NLR≥5, PLR≥185, and ALI≥24), the majority of the sample had elevated NLR (53%) and PLR (63%). Likewise, 62% of patients had high inflammation as indexed by low ALI.

SIR biomarkers as prognostic for NSCLC overall survival (OS)

The median follow-up for the sample was 8 months, with an inter-quartile range of 3 to 22 months and a range of 0 to 24 months. Overall mortality at 2-years was 61% (113/186). Survival analyses used the SIR cutoffs noted above and adjusted for covariates (i.e., ECOG-PS, age, educational level, smoking status, and treatment received at baseline). Analyses determined statistically significant associations between each biomarker and overall survival (Table 2).

thumbnail
Download:
Table 2. Associations between systemic inflammation ratios (SIRs) and overall survival.

https://doi.org/10.1371/journal.pone.0282206.t002

After adjusting for covariates, the hazard ratio (HR) for NLR is 1.91 (95% CI: 1.29, 2.84; p = 0.001). That is, patients with elevated NLR at diagnosis (higher inflammation) were approximately twice as likely to die at any time point compared to those with a lower inflammation ratio. Patients with baseline NLR<5 (47%) had an estimated one-year survival probability of 0.76 (95% CI: 0.67, 0.85), while those with NLR≥5 (53%) had a one-year survival probability of 0.46 (95% CI: 0.37, 0.57). The log-rank test was significant (p-value = 0.001) indicating a marked difference in survival (Fig 2).

thumbnail
Download:
Fig 2. Kaplan Meier survival curve for NSCLC patients stratified by NLR systemic inflammation, NLR < 5 and NLR ≥ 5 (n = 184), showing worse survival probability (p = 0.001) for patients having greater inflammation (i.e., above NLR cutoff).

https://doi.org/10.1371/journal.pone.0282206.g002

The hazard ratio for elevated PLR is 2.08 (95% CI: 1.34, 3.22; p = 0.001), comparable to the association between NLR and survival. Patients with baseline PLR<185 (43%) had a one-year survival probability of 0.78 (95% CI: 0.69, 0.89), while those with a baseline PLR≥185 (57%) had a one-year survival probability of 0.49 (95% CI: 0.41, 0.59). The log-rank test was significant (p = 0.005), indicating a significant difference in survival between groups (Fig 3).

thumbnail
Download:
Fig 3. Kaplan Meier survival curve for NSCLC patients stratified by PLR systemic inflammation, PLR < 185 and PLR ≥ 185 (n = 184), showing worse survival probability (p = 0.005) for patients having greater inflammation (i.e., above PLR cutoff).

https://doi.org/10.1371/journal.pone.0282206.g003

The hazard ratio for ALI is 0.53 (95% CI: 0.34, 0.82; p = 0.005), indicating that those with higher levels of ALI (i.e., indicative of lower inflammation) had better survival. Patients with baseline ALI<24 (62%) were approximately twice as likely to die than those with higher ALI, having an estimated survival probability of 0.50 (95% CI: 0.41, 0.60). Patients with baseline ALI≥24 (38%) had a one year estimated survival probability of 0.77 (95% CI: 0.77, 0.87). The log-rank test indicated a significant difference in survival (p < 0.005) between groups (Fig 4).

thumbnail
Download:
Fig 4. Kaplan Meier survival curve for NSCLC patients stratified by ALI systemic inflammation, ALI < 24 and ALI ≥ 24 (n = 182), showing worse survival probability (p<0.005) for patients having greater inflammation (i.e., below ALI cutoff).

https://doi.org/10.1371/journal.pone.0282206.g004

In summary, the relevance of the three SIRs was confirmed, with higher levels of inflammation predicting lower OS in this NSCLC sample.

Relationship of depressive symptoms with SIR biomarkers

Analyses supported the prediction of significant relationships between patients’ depressive symptoms and SIRs, both obtained at diagnosis, adjusted for covariates (Table 3). That is, the PHQ-9 had a significant positive association with NLR (p = 0.02) and PLR (p = 0.02). In both cases, elevations in depressive symptoms were associated with greater elevations of neutrophils and platelets in relationship to lymphocytes. Also, PHQ-9 had an inverse, significant association with ALI (p = 0.009). High levels of depressive symptoms were associated with lower ALI values.

thumbnail
Download:
Table 3. Associations between depression symptoms (PHQ-9) and systemic inflammation ratios at baseline.

https://doi.org/10.1371/journal.pone.0282206.t003

To illustrate significant associations at the patient level, figures display the relationship between depression [PHQ-9<8 (none/mild) vs ≥8 (moderate to severe)] and SIRs (e.g., NLR<5 vs. NLR ≥5), yielding 4 patient groups determined by cutoffs. For patients with no/mild depressive symptoms, they were as likely to be found above (50%) as they were to be below (50%) the NLR cutoff (Fig 5, left side). Although not 50/50, similar percentages of patients with no/mild depressive symptoms were found above (56%) versus below (42%) the PLR cutoff (Fig 6, left side) and below (58%) versus above (42%) the ALI cutoff (Fig 7, left side).

thumbnail
Download:
Fig 5. NSCLC patients classified into PHQ/NLR subgroups.

Patients with no/low depressive symptoms were as likely to have low NLRs (<5; 50%) as have high NLRs (≥5; 50%). In contrast, for patients with moderate/severe depressive symptoms, significantly more patients had prognostically worse, high NLRs (≥5; 58%) rather than low NLRs (<5; 42%). Percentage data are provided with error bars.

https://doi.org/10.1371/journal.pone.0282206.g005

thumbnail
Download:
Fig 6. NSCLC patients classified into PHQ/PLR subgroups.

Patients with no/low depressive symptoms were as likely to have low PLRs as to have high PLRs (left side). In contrast, for patients with moderate/severe depressive symptoms, significantly more patients had prognostically worse, high PLRs rather than low PLRs (right side). Percentage data are provided with error bars.

https://doi.org/10.1371/journal.pone.0282206.g006

thumbnail
Download:
Fig 7. NSCLC patients classified into PHQ/ALI subgroups.

Patients with no/low depressive symptoms were as likely to have low ALIs as to have high ALIs (left side). In contrast, for patients with moderate/severe depressive symptoms, significantly more patients had prognostically worse, low ALIs rather than high ALIs (right side). Percentage data are provided with error bars.

https://doi.org/10.1371/journal.pone.0282206.g007

In contrast, patients with moderate to severe depressive symptoms had, disproportionately, high inflammation rather than low, an effect seen across SIRs. For the depressed, 58% were above the NLR cutoff versus 42% below the cutoff (Fig 5, right side). The discrepancy is 2 to 3 times greater for PLR and ALI ratios. For the depressed, 77% were above the PLR cutoff versus 23% below (Fig 6, right side). The effect is the same, though reversed, for the depressed patients and ALI; 70% were below the cutoff verses 30% above (Fig 7, right side).

Conclusions

Systemic inflammatory biomarkers—NLR, PLR, ALI—are known predictors of advanced NSCLC survival [38]. This is supported by new data from a large sample of patients assessed at diagnosis and followed for two years. Novel data show patients’ baseline depressive symptoms to be significantly associated with the same biomarker predictors, adjusting for demographic, smoking, and functional status variables. The relationship is strongest for those with moderate to severe depressive symptoms, being 1.3 to 3 times more likely to also have prognostically poor inflammation levels. Biological and psychological domains often viewed as disparate were found to covary in a significant manner, an effect replicated across the three biomarkers.

Immune-based therapies are standard of care in patients with non-small cell lung cancer [23]. A meta-analysis of eight randomized clinical trials demonstrated that the administration of chemotherapy in combination with antibodies that bind to and block the T cell PD-1 and CTLA-4 inhibitory pathways (so called checkpoint inhibitors-CPI) significantly improve overall and progression-free survival [39]. Inflammatory immune processes could alter and possibly subvert the anti-tumor immune response generated by CPI. Indeed, Ayers et al. (2021) showed elevated NLR to correlate with shorter treatment times and overall survival in lung patients receiving PD-1/PD-L1 inhibitors [33]. A large proportion of patients—35% found here—had significant depressive symptoms prior to receipt of immune-based therapy. The mechanism whereby depression leads to reduced treatment response or poorer survival is likely multi-factorial, but the current literature supports a model in which attendant inflammation exerts an inhibitory effect on immune function. These data support further investigation of depression and its mechanisms in relationship the efficacy of current therapeutic modalities.

Close inspection of the data highlighted individual differences in the association between depression and inflammation elevations. That is, data show the majority of depressed patients had prognostically poor PLR (Fig 6) and ALI (Fig 7) levels, double to triple the number of persons without depressive symptoms. The numbers are even more compelling when contrasted with currently accepted indicators of poor survival. That is, there were more depressed/elevated inflammation patients in this study than patients with low performance status (ECOG ≥ 2; 13%), overweight status (29%), and high school education (34%). Also, depressed NSCLC patients have a constellation of negative emotions and cognitions heightening vulnerability and lessening resilience [40]. For example, they perceive, personally, that little can be done control the disease and are pessimistic about treatment control as well. Theirs is a unique struggle with lung cancer, one different from peers, still concerned but able to cope, but doing so without a depression burden [40].

It is unknown if the depression/inflammation association would remain once cancer treatment begins. However, we have shown the trajectory of depression from diagnosis through two years (18 assessments) predicted NSCLC patients’ survival (HR = 1.09, 95% CI = 1.03–1.15, p = .002), above and beyond baseline depression, sociodemographics, smoking status, cell type, and receipt of targeted treatments and immunotherapies [6]. Taken together, data support psychological, behavioral, and biologic toxicities of depression capable of influencing treatment response and/or survival.

The validity of the depression and SIR associations found here is supported by pathophysiology studies of psychiatric disorders [14,40,41] including mood disorders [15,42,43]. In NLR studies with individuals with depression but not cancer, the relationship between depression severity and NLR elevation is evident. Patients diagnosed with major depressive disorder (MDD) but not yet treated have been found with significant elevations of NLR in comparison to matched controls without MDD, an effect replicated [44–46]. Further, study of major depressive disorder (MDD) “severity” has found patients with MDD who previously attempted suicide had significantly higher NLRs compared to matched MDD patients without a suicide history [47].

The non-cancer studies are important for interpreting the present findings, as in combination, they show a dose response relationship between depression severity ad elevated NLR. This data provides a conceptual replication, i.e. the elevated depression/elevated inflammation effect is the same, but now observed in a cancer sample. PLR and ALI have not been studied in MDD patients, but here the depression/NLR effect was replicated in PLR and ALI analyses, adding reliability and validity the findings. And, it is likely that similar effects would be found with MDD patients. A plausible interpretation of this data is a differential effect of depression severity which “adds” to patients’ basal level of inflammation arising from other sources. This may be a contributing mechanism to the uniquely high rates of depression in found LC patients at diagnosis [5,7,8] and the toxicity of the depression trajectory thereafter predicting lower survival [6].

Strengths and weaknesses are considered. Uniform timing of data collection during the difficult diagnostic period was clinically and methodologically important. The sample was at least equal to if not larger than those found in multiple SIR/OS meta analyses [27,29–33,39] and sufficient for this first test of depression in relationship to lung biomarkers [48]. The SIRs are accurate, precise, and robust survival predictors across tumor types [49]. The PHQ-9 has equivalent psychometric strengths [50], and the homogeneity of the sample and the robust range of depressive symptoms found added power. Diagnostic interviews were not done, with the number of patients with MDD unknown. However, MDD criterion symptoms, e.g., low mood, anhedonia, cognitive difficulties, hopelessness, and suicidality, were endorsed by the patients, as has been the case other studies of depressed NSCLC patients [6,9,40,51]. The sample had comparatively low ethnic and racial diversity (18% vs. 22% nationally in the US) [52], potentially limiting generalizability.

In conclusion, new therapies for NSCLC are changing previously bleak, 5-month survival prognosis [53] to 2 years and longer [54]. However, variability in patient response to new therapies exists [2,3], with a pressing need to identify biomarkers predictive of first line therapy failure. The clinical significance of depression/systemic inflammation associations found here require further study as individually, the variables are prognostic for NSCLC survival [6,55,56]. Intensive study of depression among patients with NSCLC, combined with measures of cell biology, inflammation, and immunity, is needed to extend these findings and discover their mechanisms, with the long term aim to improve patients’ quality of life, treatment responses, and longevity.

Supporting information

S1 Table. Overall p-values from univariate linear analysis.

https://doi.org/10.1371/journal.pone.0282206.s001

(PDF)

S2 Table. Overall p-values from univariate cox regression.

https://doi.org/10.1371/journal.pone.0282206.s002

(PDF)

Acknowledgments

We thank the Recruitment, Intervention, and Survey Shared Resource at The Ohio State University Comprehensive Cancer Center for data management, Strategic Research Group for PRO assessments, and John Covarrubias for early assistance.

References

  1. 1. Siegel RL, Miller KD, Fuchs HE, Jemal A. Cancer statistics, 2022. CA Cancer J Clin. 2022;72: 7–33. pmid:35020204
  2. 2. Chen R, Manochakian R, James L, Azzouqa A-G, Shi H, Zhang Y, et al. Emerging therapeutic agents for advanced non-small cell lung cancer. J Hematol OncolJ Hematol Oncol. 2020;13: 1–23. pmid:32448366
  3. 3. Majeed U, Manochakian R, Zhao Y, Lou Y. Targeted therapy in advanced non-small cell lung cancer: current advances and future trends. J Hematol Oncol. 2021;14: 1–20.
  4. 4. Wang X, Wang N, Zhong L, Wang S, Zheng Y, Yang B, et al. Prognostic value of depression and anxiety on breast cancer recurrence and mortality: a systematic review and meta-analysis of 282,203 patients. Mol Psychiatry. 2020;25: 3186–3197. pmid:32820237
  5. 5. Walker J, Magill N, Mulick A, Symeonides S, Gourley C, Toynbee M, et al. Different independent associations of depression and anxiety with survival in patients with cancer. J Psychosom Res. 2020;138: 110218. pmid:32911441
  6. 6. Andersen BL, McElroy JP, Carbone DP, Presley CJ, Smith RM, Shields PG, et al. Psychological symptom trajectories and non-small cell lung cancer survival: A joint model analysis. Psychosom Med. 2022;84: 215–223. pmid:34629425
  7. 7. Wong MC, Lao XQ, Ho K-F, Goggins WB, Shelly LA. Incidence and mortality of lung cancer: global trends and association with socioeconomic status. Sci Rep. 2017;7: 1–9.
  8. 8. Lee H, Singh GK. The association between psychological distress and cancer mortality in the United States: Results from the 1997–2014 NHIS-NDI Record Linkage Study. Ann Behav Med. 2021;55: 621–640. pmid:33410477
  9. 9. Andersen BL, Valentine TR, Lo SB, Carbone DP, Presley CJ, Shields PG. Newly diagnosed patients with advanced non-small cell lung cancer: a clinical description of those with moderate to severe depressive symptoms. Lung Cancer. 2020;145: 195–204. pmid:31806360
  10. 10. Brothers BM, Yang H-C, Strunk DR, Andersen BL. Cancer patients with major depressive disorder: testing a biobehavioral/cognitive behavior intervention. J Consult Clin Psychol. 2011;79: 253. pmid:21341891
  11. 11. Walser T, Cui X, Yanagawa J, Lee JM, Heinrich E, Lee G, et al. Smoking and lung cancer: the role of inflammation. Proc Am Thorac Soc. 2008;5: 811–815. pmid:19017734
  12. 12. Brown D, Zingone A, Yu Y, Zhu B, Candia J, Cao L, et al. Relationship between circulating inflammation proteins and lung cancer diagnosis in the National Lung Screening Trial. Cancer Epidemiol Prev Biomark. 2019;28: 110–118. pmid:30297515
  13. 13. Mandaliya H, Jones M, Oldmeadow C, Nordman II. Prognostic biomarkers in stage IV non-small cell lung cancer (NSCLC): neutrophil to lymphocyte ratio (NLR), lymphocyte to monocyte ratio (LMR), platelet to lymphocyte ratio (PLR) and advanced lung cancer inflammation index (ALI). Transl Lung Cancer Res. 2019;8: 886–894. pmid:32010567
  14. 14. Drevets WC, Wittenberg GM, Bullmore ET, Manji HK. Immune targets for therapeutic development in depression: towards precision medicine. Nat Rev Drug Discov. 2022;21: 224–244. pmid:35039676
  15. 15. Dowlati Y, Herrmann N, Swardfager W, Liu H, Sham L, Reim EK, et al. A meta-analysis of cytokines in major depression. Biol Psychiatry. 2010;67: 446–457. pmid:20015486
  16. 16. Osimo EF, Baxter LJ, Lewis G, Jones PB, Khandaker GM. Prevalence of low-grade inflammation in depression: A systematic review and meta-analysis of CRP levels. Psychol Med. 2019;49: 1958–1970. pmid:31258105
  17. 17. Majd M, Saunders EF, Engeland CG. Inflammation and the dimensions of depression: A review. Front Neuroendocrinol. 2020;56: 100800. pmid:31654681
  18. 18. Sforzini L, Nettis MA, Mondelli V, Pariante CM. Inflammation in cancer and depression: A starring role for the kynurenine pathway. Psychopharmacology (Berl). 2019;236: 2997–3011. pmid:30806743
  19. 19. Miller AH, Raison CL. The role of inflammation in depression: from evolutionary imperative to modern treatment target. Nat Rev Immunol. 2016;16: 22–34. pmid:26711676
  20. 20. Maeng SH, Hong H. Inflammation as the potential basis in depression. Int Neurourol J. 2019;23: S63–71. pmid:31795605
  21. 21. Chiu WC, Su YP, Su KP, Chen PC. Recurrence of depressive disorders after interferon-induced depression. Transl Psychiatry. 2017;7: e1026–e1026. pmid:28170005
  22. 22. Lotrich FE. Major depression during interferon-α treatment: Vulnerability and prevention. Dialogues Clin Neurosci. 2022;4: 417–425.
  23. 23. Ettinger DS, Wood DE, Aisner DL, Akerley W, Bauman JR, Bharat A, et al. Non–small cell lung cancer, Version 3.2022, NCCN Clinical Practice Guidelines in Oncology. J Natl Compr Canc Netw. 2022;20: 497–530. pmid:35545176
  24. 24. Sullivan DR, Forsberg CW, Ganzini L, Au DH, Gould MK, Provenzale D, et al. Longitudinal changes in depression symptoms and survival among patients with lung cancer: a national cohort assessment. J Clin Oncol. 2016;34: 3984. pmid:27996350
  25. 25. Zhou Q, Su S, You W, Wang T, Ren T, Zhu L. Systemic inflammation response index as a prognostic marker in cancer patients: A systematic review and meta-snalysis of 38 cohorts. Dose-Response. 2021;19: 1–14.
  26. 26. Prame Kumar K, Nicholls AJ, Wong CH. Partners in crime: Neutrophils and monocytes/macrophages in inflammation and disease. Cell Tissue Res. 2018;371: 551–565. pmid:29387942
  27. 27. Xu H, He A, Liu A, Tong W, Cao D. Evaluation of the prognostic role of platelet-lymphocyte ratio in cancer patients treated with immune checkpoint inhibitors: A systematic review and meta-analysis. Int Immunopharmacol. 2019;77:1–9. pmid:31677498
  28. 28. Zhang N, Jiang J, Tang S, Sun G. Predictive value of neutrophil-lymphocyte ratio and platelet-lymphocyte ratio in non-small cell lung cancer patients treated with immune checkpoint inhibitors: A meta-analysis. Int Immunopharmacol. 2020;85: 1–9. pmid:32531712
  29. 29. Jin J, Yang L, Liu D, Li W. Association of the neutrophil to lymphocyte ratio and clinical outcomes in patients with lung cancer receiving immunotherapy: a meta-analysis. BMJ Open. 2020;10: e035031. pmid:32499266
  30. 30. Andersen BL, DeRubeis RJ, Berman BS, Gruman J, Champion VL, Massie MJ, et al. Screening, assessment, and care of anxiety and depressive symptoms in adults with cancer: an American Society of Clinical Oncology guideline adaptation. J Clin Oncol. 2014;32: 1605–1619. pmid:24733793
  31. 31. Kroenke K, Spitzer RL, Williams JB. The PHQ‐9: validity of a brief depression severity measure. J Gen Intern Med. 2001;16: 606–613. pmid:11556941
  32. 32. Sacdalan DB, Lucero JA, Sacdalan DL. Prognostic utility of baseline neutrophil-to-lymphocyte ratio in patients receiving immune checkpoint inhibitors: A review and meta-analysis. OncoTargets Ther. 2018;11: 955–965. pmid:29503570
  33. 33. Ayers KL, Ma M, Debussche G, Corrigan D, McCafferty J, Lee K, et al. A composite biomarker of neutrophil-lymphocyte ratio and hemoglobin level correlates with clinical response to PD-1 and PD-L1 inhibitors in advanced non-small cell lung cancers. BMC Cancer. 2021;21: 1–11.
  34. 34. Ding N, Pang Z, Shen H, Ni Y, Du J, Liu Q. The prognostic value of PLR in lung cancer: A meta-analysis based on results from a large consecutive cohort. Sci Rep. 2016;6: 1–9.
  35. 35. Hua X, Chen J, Wu Y, Sha J, Han S, Zhu X. Prognostic role of the advanced lung cancer inflammation index in cancer patients: A meta-analysis. World J Surg Oncol. 2019;17: 1–9.
  36. 36. Oken MM, Creech RH, Tormey DC, Horton J, Davis TE, McFadden ET, et al. Toxicity and response criteria of the Eastern Cooperative Oncology Group. Am J Clin Oncol. 1982;5: 649–656. pmid:7165009
  37. 37. Team RC. R: A language and environment for statistical computing. 2021.
  38. 38. Kazandjian D, Gong Y, Keegan P, Pazdur R, Blumenthal GM. Prognostic value of the lung immune prognostic index for patients treated for metastatic non–small cell lung cancer. JAMA Oncol. 2019;5: 1481–1485. pmid:31343662
  39. 39. Meng L-F, Huang J-F, Luo P-H, Huang S-X, Wang H-L. The efficacy and safety of immune checkpoint inhibitor plus chemotherapy in patients with advanced non-small-cell lung cancer: A meta-analysis. Invest New Drugs. 2022; 1–8. pmid:35412172
  40. 40. Valentine TR, Presley CJ, Carbone DP, Shields PG, Andersen BL. Illness perception profiles and psychological and physical symptoms in newly diagnosed advanced non-small cell lung cancer. Health Psychol. 2022;41: 379–388. pmid:35604701
  41. 41. Zulfic Z, Weickert CS, Weickert TW, Liu D, Myles N, Galletly C. Neutrophil–lymphocyte ratio–a simple, accessible measure of inflammation, morbidity and prognosis in psychiatric disorders? Australas Psychiatry. 2020;28: 454–458. pmid:32174125
  42. 42. Medina-Rodriguez EM, Lowell JA, Worthen RJ, Syed SA, Beurel E. Involvement of innate and adaptive immune systems alterations in the pathophysiology and treatment of depression. Front Neurosci. 2018; 547: 1–14. pmid:30174579
  43. 43. Kim Y-K, Na K-S, Myint A-M, Leonard BE. The role of pro-inflammatory cytokines in neuroinflammation, neurogenesis and the neuroendocrine system in major depression. Prog Neuropsychopharmacol Biol Psychiatry. 2016;64: 277–284. pmid:26111720
  44. 44. Demir S, Atli A, Bulut M, İbiloğlu AO, Güneş M, Kaya MC, et al. Neutrophil–lymphocyte ratio in patients with major depressive disorder undergoing no pharmacological therapy. Neuropsychiatr Dis Treat. 2015;11: 2253–2258. pmid:26347335
  45. 45. Demircan F, Gözel N, Kılınç F, Ulu R, Atmaca M. The impact of red blood cell distribution width and neutrophil/lymphocyte ratio on the diagnosis of major depressive disorder. Neurol Ther. 2016;5: 27–33. pmid:26686339
  46. 46. Özyurt G, Binici NC. Increased neutrophil-lymphocyte ratios in depressive adolescents is correlated with the severity of depression. Psychiatry Res. 2018;268: 426–431. pmid:30130709
  47. 47. Ekinci O, Ekinci A. The connections among suicidal behavior, lipid profile and low-grade inflammation in patients with major depressive disorder: a specific relationship with the neutrophil-to-lymphocyte ratio. Nord J Psychiatry. 2017;71: 574–580. pmid:28800269
  48. 48. Group BDW, Atkinson AJ Jr, Colburn WA, DeGruttola VG, DeMets DL, Downing GJ, et al. Biomarkers and surrogate endpoints: Preferred definitions and conceptual framework. Clin Pharmacol Ther. 2001;69: 89–95. pmid:11240971
  49. 49. Winther-Larsen A, Aggerholm-Pedersen N, Sandfeld-Paulsen B. Inflammation scores as prognostic biomarkers in small cell lung cancer: A systematic review and meta-analysis. Syst Rev. 2021;10: 1–14.
  50. 50. Mitchell AJ, Yadegarfar M, Gill J, Stubbs B. Case finding and screening clinical utility of the Patient Health Questionnaire (PHQ-9 and PHQ-2) for depression in primary care: A diagnostic meta-analysis of 40 studies. BJPsych Open. 2016;2: 127–138. pmid:27703765
  51. 51. Presley CJ, Arrato NA, Janse S, Shields PG, Carbone DP, Wong ML, et al. Functional disability among older versus younger adults with advanced non–small-cell lung cancer. JCO Oncol Pract. 2021;17: e848–e858. pmid:33939536
  52. 52. U.S. Census Bureau. United States Population Estimates: Quick Facts. 2021 Jul. Available: https://www.census.gov/quickfacts/fact/table/US#.
  53. 53. Noone AM, Howlader N, Krapcho M al, Miller D, Brest A, Yu M, et al. SEER cancer statistics review, 1975–2015. Bethesda MD Natl Cancer Inst. 2018;4.
  54. 54. Garon EB, Hellmann MD, Rizvi NA, Carcereny E, Leighl NB, Ahn M-J, et al. Five-year overall survival for patients with advanced non‒small-cell lung cancer treated with pembrolizumab: results from the phase I KEYNOTE-001 study. J Clin Oncol. 2019;37: 2518.
  55. 55. Li M, Spakowicz D, Burkart J, Patel S, Husain M, He K, et al. Change in neutrophil to lymphocyte ratio during immunotherapy treatment is a non-linear predictor of patient outcomes in advanced cancers. J Cancer Res Clin Oncol. 2019;145: 2541–2546. pmid:31367835
  56. 56. Yun N, Rouhani SJ, Gilmore B, Ritz EM, Bestvina CM, Tarhoni I, et al. The prognostic value of neutrophil-to-lymphocyte ratio in patients with epidermal growth factor receptor mutated advanced non-small cell lung cancer (NSCLC). J Clin Oncol. 2020;Suppl 15: e21675–e21675.