Management of T cell responses by anesthetic drugs-propofol & isoflurane in perioperative breast cancer patients: A prospective hospital-based study

Study setting

This prospective study recruited individuals with breast cancer presenting to the Department of Surgical Oncology, Chittaranjan National Cancer Institute, Kolkata, who were advised for surgery [modified radical mastectomy (MRM)/breast conserving surgery (BCS)] according to the considered inclusion and exclusion criteria¹⁸. Recruitment of patients in this study was administered through informed consent form. Patients were block-randomized in two arms- volatile anesthestic isoflurane (n=50) in one arm and intravenous anaesthestic propofol (n=50) in the other. The study was approved by the Institutional Ethics Committee. The study adhered to the Indian Council of Medical Research’s ethical guidelines for biomedical research on human participants (2017)¹⁹ and was registered at Clinical Trials Registry-India (CTRI/2020/11/02886 dated Nov 3, 2020).

Protocol for anesthetic management

The protocol for anesthetic management was same as reported before¹⁸. In the isoflurane group induction was achieved with thiopentone, 3-5 µg/kg and maintenance was done by using isoflurane 50 per cent in nitrous oxide to achieve minimum alveolar concentration of 1.0. For the propofol group induction was same but the maintenance was done using propofol with a target-controlled infusion pump, at an effect-site concentration of 2-3 µg/ml.

Collection of blood samples

Perioperative blood samples (preoperative, 1 day before surgery; intraoperative, 1 h after surgical incision and postoperative, 48 h after surgery) were collected from the study participants in both anesthetic arms¹⁸.

Typically, these anesthetic drugs are rapidly metabolized and eliminated from the body. The terminal elimination phase for propofol lasts for 1.5-31h²⁰ whereas in case of isoflurane the elimination half-life of serum fluoride levels (metabolite of isoflurane) has been estimated to approximately 21 h²¹. Therefore, in this study, we have investigated the T cell immune responses for 1 h (intra) and 48 h (post) after incision.

Monitoring of anesthetic agents in serum samples

propofol was detected in serum samples using high-performance liquid chromatography (HPLC)^22,23. Fluoride concentration during the pre, intra and postoperative periods after isoflurane anesthesia were measured directly from serum samples using fluoride ion selective electrode technique²¹ (Supplementary Material, Supplementary Fig. 1).

Flow cytometry for detection of lymphocyte subtypes

The immunophenotype of the lymphocytes was analyzed with flow cytometry^24-26. Whole blood (100 µl) was incubated with a cocktail of fluorescent-tagged antibodies, in dark for 45 min at 4°C, suspended in PBS and was analyzed in a flow cytometer. The detailed protocol and gating strategy have been depicted in Supplementary Material (Supplementary Fig. 2 and 3).

Estimation of cytokines

ELISA kits [Ray Biotech Peachtree Corners, GA, USA] were used to measure serum concentrations of interleukin (IL)-2, IL-12, IL-10, tumor necrosis factor (TNF)-α and interferon (IFN)-γ at 450 nm in a microplate reader¹⁸.

In silico analyses

Grid-based Ligand Docking with Energetics (GLIDE, Schrödinger Release 2021-1: Maestro, Schrödinger, LLC, New York, NY, 2021) was used for docking analysis of isoflurane and propofol against CD4 and CD8 targeted proteins in extra precision mode. LigPlot+ ( https://www.ebi.ac.uk/thornton-srv/software/LigPlus/) and the ligand interactions module of Schrödinger were used to show the presence of intermolecular bonds between protein-drug complexes. Desmond program (Schrödinger Release 2021-1: Maestro, Schrödinger, LLC, New York, NY, 2022) was used for molecular dynamic simulations (MDS) of the Holo-1: CD4-isoflurane complex, Holo-2: CD4-propofol complex, Holo-3: CD8-isoflurane complex, and Holo-4: CD8-propofol complex, to understand the dynamic behaviour, mode of binding and inhibitor specificity for all the systems (details provided in Supplementary Material).

Statistical analysis

To compare the clinicopathological characteristics, Student’s t-test and Chi-square test were performed. The comparison of repeated measurement indicators at different observation time among the isoflurane and propofol group was performed using a linear mixed effect model. The model was used with patient as a random effect and with arm, timepoints and arm-by-timepoints as fixed effects. The arm-by-timepoints interaction indicates whether the change over time differed between the two anesthetic arms²⁷. Post-hoc analyses were used with Bonferroni correction wherever applicable. The same linear mixed effect model was applied to investigate the effect of the anesthetic arm on Th cells with adjustment of confounding factors including age, height, weight, type of surgery, histopathology, grade, stage, molecular subtypes, ASA classification, duration of anesthesia and surgery, and pain score. Data were analyzed using Statistical Package for Social Sciences (IBM SPSS, ver. 25.0, Chicago, IL, USA). A value of P< 0.05 was considered significant.

Participant clinicopathological characteristics

Participants were recruited from December, 2020 to July, 2023 and followed up till August 2024 (Fig. 1). The clinicopathological characteristics of the enrolled participants with breast cancer (Table I) showed no significant difference between the two anesthetic groups in terms of age, weight, height, American Society of Anesthesiologists (ASA) classification, histopathology, estrogen receptor (ER)/ progesterone receptor (PR) / human epidermal growth factor receptor 2 (Her2) nu status, grade, stage, type of surgery, duration of surgery, duration of anesthesia and the post-operative numerical pain score. We also monitored the presence of anesthetic agents in serum samples. propofol and fluoride were detected only in the intraoperative serum samples (Supplementary Fig. 1).

Close

Fig. 1.

CONSORT flow diagram for recruitment of breast cancer patients in the study.

Export to PPT

Impact of anesthetic agents on Th and Tc cells

isoflurane significantly suppressed the frequency of Th (CD3+CD8-CD4+) cells during the intraoperative than the preoperative period which was evident from the ‘intra-group’ analysis (Fig. 2; panels A and B). On the other hand, Th cells were significantly increased in postoperative compared to pre and intraoperative periods of propofol group (Fig. 2A and B). ‘Inter-group’ analysis showed the intraoperative Th cell depletion by isoflurane was significant compared to propofol (β-coefficient: -8.75; 95% CI: -13.00 to -4.51; Fig. 2C). Further subgroup analysis showed that the isoflurane-induced suppression of Th cell frequency during intraoperative period was prominent in stage II breast cancer as observed from ‘intra-group’ comparisons (Fig. 2D) and ‘inter-group’ comparisons (β: -8.51; 95% CI: -13.96 to -3.06; Fig. 2E). The intraoperative suppressive effect of isoflurane on Th cell frequency was also evident in the molecular subtype ER+PR+HER2- during the ‘intra-group’ comparison between intraoperative and preoperative period (Fig. 2F) and in the molecular subtype ER+PR+HER2- (β: -8.29; 95% CI: -14.37 to -2.21) as well as in ER-PR-HER2- (β: -11.10; 95% CI: -25.10 to 2.90) during the ‘inter-group’ comparison between isoflurane and propofol (Fig. 2G). According to the type III fixed effects of linear mixed model analysis (Supplementary Table I) it was further confirmed that Th cell frequency was significantly impacted by the interaction of the anesthetic arm with the timepoints. However, the age of the individuals also showed significant effect on the Th cells but not as an interaction of age with arm. The other confounding factors did not have any significant effect on the Th cell frequency of the individuals with breast cancer.

Close

Fig. 2.

Flowcytometric analysis and frequency of CD3+CD8-CD4+ (Th) and CD3+CD8+CD4- (Tc) cells (%) in peripheral blood of perioperative breast cancer patients administered with anesthetic agents, Iso or Pro. (A) Th cell frequency in representative female anesthetized with Iso/Pro; (B) the effect of Iso/Pro on Th cells at different time points with intra-group analysis; (C) with inter-group analysis; (D) the effect of Iso/Pro on Th cells according to tumor stage (II) with intra-group analysis; and (E) with inter-group analysis; (F) the effect of Iso/Pro on Th cells according to breast cancer molecular subtype (ER+PR+HER-) with intra-group analysis; and (G) with inter-group analysis (ER+PR+HER- and ER-PR-HER2-); (H) Tc cell frequency in representative female anesthetized with Iso/Pro; (I) the effect of Iso/Pro on Tc cells at different timepoints with intra-group analysis; and (J) with inter-group analysis; (K) the effect of Iso/Pro on Th cells according to tumor stage (II); and (L) molecular subtypes (ER+PR+HER-, ER+PR-HER-); (M) ratio of CD4+/CD8+ T cells in perioperative breast cancer patients. The graphs were plotted based on the mean±SD. Iso: isoflurane; Pro: propofol; Pre: preoperative; Intra: intraoperative; Post: postoperative.

Export to PPT

Tc cells (CD3+CD4-CD8+) were also significantly reduced in count during the postoperative than preoperative period by both isoflurane and propofol (Fig. 2H and I) and the ‘inter-group’ comparison showed no differential effect of the anesthetics (Fig. 2J). However, subgroup analysis showed that isoflurane inflicted significant reduction in Tc cell frequency during the postoperative period compared to intraoperative period particularly in stage II cases but the same was not evident with propofol (Fig. 2K). Similarly, the decrease in Tc cells during postoperative period compared to intraoperative period by isoflurane was also observed across the molecular subtypes ER+PR+HER2- and ER+PR-HER2- whereas propofol did not induce any such changes (Fig. 2L). Additionally, isoflurane significantly increased the ratio of CD4+/CD8+ T cells postoperatively than intraoperative period, while the same ratio was maintained with propofol (Fig. 2M).

Effect of isoflurane/propofol on T cell activity markers, B cells and inflammatory cytokines

In order to activate a Tc or a Th cell to proliferate and differentiate into an effector cell, an antigen presenting cell (APC) provides two signals – (i) through foreign peptide bound to major histocompatibility complex (MHC) on the surface of APC which signals through T cell receptor (TCR); and (ii) through co-stimulatory molecules - CD80 and CD86, which are identified by the co-receptor protein CD28 on the surface of the T cell²⁸. Therefore, we checked the effect of isoflurane/propofol on the TCR and CD28 activity and did not observe any significant difference between the anesthetic groups (Supplementary Fig. 4).

The Th cells inflict a direct anti-tumor response and help B lymphocytes to produce antibodies²⁹. Therefore, we investigated the impact of the anesthetic agents on B cells. The ‘intra-group’ analysis revealed that the increase in B cell (CD3-CD19+) frequency was significant in the postoperative as compared to the intraoperative period in the isoflurane group whereas the same was higher in the postoperative than the preoperative period in the propofol group (Fig. 3A, B). However, ‘inter-group’ analysis showed significant decrease of B cell frequency by isoflurane compared to propofol during the intraoperative period (β: -7.51; 95% CI: -15.46 to 0.44; Fig. 3C). Subgroup analysis exhibited that isoflurane-induced increase in B cells during the postoperative period compared to pre and intraoperative period was significant only in breast cancer stage II cases (Fig. 3D). However, comparison between isoflurane and propofol showed that B cells were significantly reduced by isoflurane than propofol during the intraoperative period (β: -7.14; 95% CI: -16.47 to 2.18; Fig. 3E). The increment of B cell frequency by propofol during the postoperative than preoperative period was significant only in the ER-PR-HER2- breast cancer molecular subtype (Fig. 3F).

Close

Fig. 3

Flowcytometric analysis of CD3-CD19+ B cells and expression of serum inflammatory markers in peripheral blood of perioperative breast cancer patients administered with anesthetic agents, Iso or Pro. (A) B cell frequency in representative female anesthetized with Iso/Pro; (B) comparative effect of Iso/Pro on B cells at different timepoints with intra-group analysis; and (C) with inter-group analysis; (D) the effect of Iso/Pro on B cells according to tumor stage (II) with intra-group analysis; and (E) with inter-group analysis; (F) the effect of Iso/Pro on B cells according to breast cancer molecular subtype (ER-PR-HER2-); (G and H) comparative effect of Iso /Pro on serum inflammatory markers such as IL-12 and IL-10 according to timepoint with intra-group analysis respectively; (I and J) IL-12 and IL-10 with inter-group analysis respectively; (K) TNF-α; ( L) IL-2; and (M) IFN-γ in perioperative breast cancer patients. The graphs were plotted based on the mean± SD.

Export to PPT

A panel of cytokines including IL-12, IL-10, TNF-α, IL-2 and IFN-γ involved in innate and adaptive immune responses mediated through T and B cells were investigated in isoflurane/propofol groups at different timepoints. Both IL-12 (Fig. 3G) and IL-10 (Fig. 3I) were increased with isoflurane in the intra than preoperative periods and reduced with propofol in the post than preoperative period. However, inter-group analysis revealed significant increase in both IL-12 and IL-10 levels during intra (IL-12 β: 13.16; 95% CI: 8.73 to 17.60; IL-10 β: 1.33; 95% CI: 0.39 to 2.27) and postoperative periods (IL-12 β: 8.70; 95% CI: 4.26 to 13.13; IL-10 β: 1.53; 95% CI: 0.59 to 2.47) in the isoflurane group than propofol (Fig. 3I and J). TNF-α was significantly inhibited postoperatively in both the anesthetic groups (Fig. 3K) whereas IL-2 (Fig. 3L) and IFN-γ (Fig. 3M) did not show any significant alteration across isoflurane/propofol groups.

Molecular docking and trajectory analysis of isoflurane and propofol with CD4 and CD8

The binding of the anesthetics with CD4 and CD8 can modulate their activity, altering immune responses. CD4 and CD8 co-receptors play crucial roles in T cell activation by binding to MHC and recruiting lymphocyte-specific protein tyrosine kinases (lck), which are essential for T cell signaling³⁰. Any alteration in these interactions due to anesthetics can impact T cell function and the overall immune response. Therefore, next we investigated molecular docking and MDS of the anesthetics, isoflurane and propofol with CD4 and CD8. Molecular docking revealed that propofol elicited better binding energy against CD4/CD8 with higher binding scores than isoflurane (Table II). Propofol showed better binding affinity and higher scores compared to isoflurane against the said targets (Fig. 4; panels A and F). The 100 nanoseconds (ns) MDS trajectory revealed a stable root mean square deviation (RMSD) from 0 to 75 ns in Holo-1 but later exhibited higher deviation, which may be due to conformational changes. Holo-2 maintained a stable trajectory throughout the time frame. Holo-3 showed consistent deviations after 20 ns compared to Holo-4. Overall, propofol’s binding appeared to better stabilize CD4 and CD8 than isoflurane, as shown by lower RMSD values (Fig. 4B and G). Root mean square fluctuation (RMSF) analysis further elucidated the impact of ligand binding on residue mobility. The Apo state showed greater fluctuations, while specific residues in Holo states (notably in Holo-2 and Holo-4) exhibited reduced mobility, indicating constrained motions due to ligand interactions (Fig. 4C and H). Radius of gyration (rGyr) measurements indicated Holo-2 with more compactness compared to Holo-1. Similarly, Holo-3 and Holo-4 showed varying degrees of compactness, with Holo-4 being notably stable. These observations aligned with the RMSF results, highlighting the influence of ligand binding on the structural integrity of the proteins (Fig. 4D and I). Solvent accessible surface area (SASA) analysis during later simulation stages demonstrated that many residues transitioned from accessible to buried states upon ligand binding. In SASA, Holo-2 and Holo-4 depicted decreased values which signified its shift towards a buried state compared to Holo-1 and Holo-3 (Fig. 4E and J). Collectively, these results affirmed that propofol bonded more effectively to CD4 and CD8, contributing to their stability through reduced residue mobility, enhanced compactness, and significant changes in protein surface orientation. Additionally, post-MDS conferred that conventional H-bonds were broken down between CD4/CD8 and isoflurane but were retained with propofol (details provided in the Supplementary Material, Supplementary Fig. 5).

Close

Fig. 4.

Molecular docking and molecular dynamic (MD) simulations of Iso and Pro with CD4 and CD8. Intermolecular hydrogen bonding, electrostatic and hydrophobic interactions formed between (A) CD4-Iso/ Pro; (F) CD8-Iso/ Pro complex; MD simulation of Iso and Pro with CD4 and CD8. Conformational stability of Apo (blue) and Holo (red) states of CD4 and CD8 proteins throughout 100 nanoseconds (ns) time period of MD simulations as evidenced by Holo-1, -2, -3, and -4 profile of MDS backbone- (B and G) RMSD; (C and H) Cα-RMSF (protein residues that may interact with the ligand were marked with green-colored vertical bars); (D and I) rGyr; and (E and J) SASA analysis. Iso: isoflurane; Pro: propofol; Holo1, CD4-isoflurane complex; Holo-2, CD4-propofol complex; Holo-3, CD8-isoflurane complex; Holo-4, CD8- propofol complex; rGyr: radius of gyration, SASA: solvent accessible surface analysis.

Export to PPT