PD-1 and LAG-3-positive T cells are associated with clinical outcomes of relapsed/refractory multiple myeloma patients

This prospective observational study enrolled a total of 71 RRMM patients who came to Changshu Second People's Hospital and Changshu First People's Hospital during January 2018 to March 2021. The diagnosis of MM was according to the guidelines for the diagnosis and management of multiple myeloma in China (2020 revision) and the criteria of National Comprehensive Cancer Network (NCCN) [15, 16]. The RRMM was defined as patients with recurrence of MM after recovery, as well as patients with  ≥ 2 standard therapies resulting in no complete recovery. The patients who were diagnosed as MM and RRMM as described above were included. The exclusion criteria were: (1) patients with other cancers; (2) patients with severe systematic diseases such as severe heart, liver or renal dysfunctions; (3) patients who received treatment of chimeric antigen receptor T cells or autologous hematopoietic stem cell transplantation before the study. Additionally, 70 patients with first diagnosed MM (non-refractory) were included as the control with the same diagnostic criteria and exclusion criteria to the above. All patients signed the written informed consent. Besides, peripheral blood samples of 70 healthy individuals were also obtained from individuals who came for physical examination. The study observed the Helsinki Declaration. Study approval was obtained from the ethical committee of Changshu Second People's Hospital and Changshu First People's Hospital.

Briefly, peripheral elbow venous blood (5 ml) was obtained from all patients within 24 h after admission. The samples were collected in tubes with heparin and added with 5 ml phosphate buffer (PBS) and then centrifuged at 1500 g for 10 min. The cell layer was then collected, added with 5 ml PBS, following with centrifugation at 1200 g for 5 min. After removing the supernatant and washing with PBS, the samples were further centrifuged at 1000 g for 2 min to obtain the peripheral blood mononuclear cells. The cells with density of 1 × 10⁶/ml were maintained in RPMI-1640 Medium (Sigma-Aldrich, St. Louis, MO, USA).

For measurement of PD-1⁺ and LAG-3⁺ T cells, flow cytometry was performed as described elsewhere [17, 18]. Antibodies used in this study included anti-CD4, anti-CD8, anti-PD-1 and anti-LAG-3 (all purchased from Abcam, USA). The measurement was conducted on a FACS Calibur flow cytometry analyzer (BD Biosciences) using Diva software (version 6.1, BD Pharmingen).

Patients’ clinical characteristics including age, sex, body mass index (BMI), disease course, International Staging System (ISS) stage, Ig type, complications and mediation history were recorded. The study did not intervene in the treatment of any patient. All MM or RRMM patients received routine treatment according to patients’ conditions. All patients were followed up for 1 year and the survival condition was recorded.

Data distribution was measured by Kolmogorov–Smirnov analysis. Normally and non-normally distributed data were expressed as mean ± SD or median (range), respectively. Comparison between two groups was conducted by Student’s t test and Mann–Whitney U test for normally and non-normally distributed data, respectively. Rates were analyzed by Chi-squared test. The correlation among T cells and inflammatory factors was analyzed by Spearman rank correlation analysis. ROC curve was used for diagnostic analysis. Logistic regression was conducted for risk factors of RRMM. P  < 0.05 was regarded as statistically different. All data were calculated using SPSS 18.0.

As shown in Table 1, the 1-year mortality rate was markedly higher in RRMM patients compared with the MM patients and the healthy control (p < 0.05). However, no significant difference was found for age, sex, BMI, ISS stage, disease course, and Ig type between RRMM and MM patients.

The frequency of circulating CD4⁺ and CD8⁺ T cells expressing PD-1 and LAG-3 was then analyzed in different patients. It was found that in both CD4⁺ and CD8⁺ T cells, the frequencies of PD-1⁺ and LAG-3⁺ T cells were all markedly higher in RRMM compared with the MM patients and the healthy control (p < 0.05, Fig. 1). The typical flow cytometry plots of PD-1⁺/LAG-3⁺ T cells are shown in Fig. 2 with the original plots in supplementary data (Additional file 1: Fig. S1). Besides, the frequencies of PD-1⁺/LAG-3⁺ T cells were also remarkably higher in both CD4⁺ and CD8⁺ T cells of RRMM patients compared with the MM patients and healthy individuals (p < 0.05). Further analysis showed, in RRMM patients, frequencies of PD-1⁺ and LAG-3⁺ T, as well as PD-1⁺/LAG-3⁺ T cells in both CD4⁺ and CD8⁺ T cells showed no significant difference in ISS stage III patients compared with the patients with ISS stage I or II (p < 0.05, Fig. 3). However, deceased patients showed higher frequencies of the above T cell subtypes in both CD4⁺ and CD8⁺ T cells than the survival patients (p < 0.05, Fig. 4).

As shown in Fig. 5, it was found all inflammatory factors were remarkably higher in RRMM and MM patients than in the healthy control, while only serum levels of IL-6 and IL-17 were markedly higher in RRMM patients compared with the MM patients (p < 0.05). No significant difference was found for CRP, TNF-α and TGF-β between RRMM and MM patients. To further analyze the clinical significance of frequency of circulating T cells in RRMM and MM patients, Spearman’s correlation analysis was conducted between circulating T cells and inflammatory factors. Positive correlation was only observed among the IL-6, IL-17 and frequency of part of the circulating T cells in CD4⁺ and CD8⁺ T cells (Table 2).

Then ROC curves were used for measurement of the sensitivity and specificity of different circulating T cell subtypes for RRMM. It was found that among the T cell subtypes, the frequency of CD8⁺PD-1⁺LAG-3⁺ T cells showed the best sensitivity 82.61% and specificity 76.06% with AUC 0.881, cutoff value 8.015% (Fig. 6). Then we also analyzed the predictive value of T cell subtypes for patients’ mortality. As shown in Fig. 7, it was found the frequency of CD4⁺PD-1⁺ cells showed the best sensitivity 84.00% and specificity 97.35% with AUC 0.887, cutoff value 41.82%.

Finally, the risk factors for RRMM were analyzed using binary logistic regression by a backstep method. As shown in Table 3, it was found frequencies of CD4⁺PD-1⁺, CD8⁺PD-1⁺/LAG-3⁺, as well as IL-6, IL-17 and TNF-α were all risk factors for incidence of RRMM in all MM patients.