Correlation between subsets of tumor-infiltrating immune cells and risk stratification in patients with cervical cancer

Clinicopathologic and prognostic significance of tumor-associated macrophages in cervical cancer: a systematic review and meta-analysis

OBJECTIVES

The role of tumor-associated macrophages (TAMs) in cervical cancer (CC) remains controversial. Here, we report a meta-analysis of the association between TAMs infiltration and clinical outcomes.

METHODS

PubMed, Embase, Web of Science, and CNKI were searched systematically from inception until December 20, 2023. Studies involving TAMs and prognosis, clinical, or pathological features were included. Quality assessments of the selected studies were assessed. The fixed-effect or random-effects model, standard mean difference (SMD), odds ratios (OR), or hazard ratios (HR) with 95% confidence intervals (CIs) were used as the effect size estimate.

RESULTS

26 eligible studies with 2,295 patients were identified. Our meta-analysis revealed that TAMs were overexpressed in CC (OR = 12.93, 95% CI = 7.73-21.61 and SMD = 1.58, 95% CI = 0.95-2.21) and that elevated TAM levels were strongly associated with lymph node metastasis (LNM) (SMD = 0.51, 95% CI = 0.90-2.01) and FIGO stages (SMD = 0.46, 95% CI = 0.08-0.85). Subgroup analysis indicated a significant positive correlation between LNM and TAMs density in tumor stroma, but not in cancer nests (SMD = 0.58, 95% CI = 0.31-0.58). Furthermore, in early stage, a stronger correlation exists between LNM and TAM density (SMD = 1.21, 95% CI = 0.75-1.66). In addition, it revealed that patients with high TAMs expression had poorer overall survival (OS) (HR = 2.55 95% CI = 1.59-4.07) and recurrence-free survival (RFS) (HR = 2.17, 95% CI = 1.40-3.35).

CONCLUSIONS

Our analyses suggest that a high density of TAMs predicts adverse outcomes in CC.

Disfunction of communication among immune cells in minimal-deviation adenocarcinoma of the cervix as an immunotherapeutic opportunity

Minimal deviation adenocarcinoma (MDA) of the uterine cervix, also referred to as malignant adenoma, is a rare subtype of cervical adenocarcinoma that exhibits histological characteristics resembling those of benign tumors, resulting in a low diagnostic rate and a lack of effective treatment options. The transcriptomic features of MDA at the single-cell resolution and within the tumor microenvironment (TME) remain unclear. In this study, we conducted single-cell transcriptomic analyses of MDA samples (Ca) and adjacent normal tissues (PCa). The present study reveals the prevalence of dendritic cells (DCs) and T cells in the carcinoma (Ca) of mammary ductal adenocarcinoma (MDA), with DCs undergoing significant metabolic reprogramming and immune stress. Additionally, our findings demonstrate the crucial involvement of DCs and T cells in the pathogenesis and metastatic progression of MDA, as evidenced by single-cell transcriptomic profiling of MDA and HPV samples. This resource provides a more profound understanding of the indolent nature of MDA and may prove useful in the development of MDA immunotherapy.

The construction of a prognostic model of cervical cancer based on four immune-related LncRNAs and an exploration of the correlations between the model and oxidative stress

Introduction: The immune-related lncRNAs (IRLs) are critical for the development of cervical cancer (CC), but it is still unclear how exactly ILRs contribute to CC. In this study, we aimed to examine the relationship between IRL and CC in detail. Methods: First, the RNAseq data and clinical data of CC patients were collected from The Cancer Genome Atlas (TCGA) database, along with the immune genes from the Import database. We used univariate cox and least absolute shrinkage and selection operator (lasso) to obtain IRLs for prediction after screening the variables. According to the expression levels and risk coefficients of IRLs, the riskscore were calculated. We analyzed the relationship between the model and oxidative stress. We stratified the risk model into two as the high and low-risk groups. We also evaluated the survival differences, immune cell differences, immunotherapeutic response differences, and drug sensitivity differences between the risk groups. Finally, the genes in the model were experimentally validated. Results: Based on the above analyses, we further selected four IRLs (TFAP2A.AS1, AP000911.1, AL133215.2, and LINC02078) to construct the risk model. The model was associated with oxidative-stress-related genes, especially SOD2 and OGG1. Patients in the high-risk group had a lower overall survival than those in the low-risk group. Riskscore was positively correlated with resting mast cells, neutrophils, and CD8+ T-cells. Patients in the low-risk group showed a greater sensitivity to immunosuppression therapy. In addition, we found that patients with the PIK3CA mutation were more sensitive to chemotherapeutic agents such as dasatinib, afatinib, dinaciclib and pelitinib. The function of AL133215.2 was verified, which was consistent with previous findings, and AL133215.2 exerted a pro-tumorigenic effect. We also found that AL133215.2 was closely associated with oxidative-stress-related pathways. Discussion: The results suggested that risk modeling might be useful for prognosticating patients with CC and opening up new routes for immunotherapy.

CNN stability training improves robustness to scanner and IHC-based image variability for epithelium segmentation in cervical histology

Background

In digital pathology, image properties such as color, brightness, contrast and blurriness may vary based on the scanner and sample preparation. Convolutional Neural Networks (CNNs) are sensitive to these variations and may underperform on images from a different domain than the one used for training. Robustness to these image property variations is required to enable the use of deep learning in clinical practice and large scale clinical research.

Aims

CNN Stability Training (CST) is proposed and evaluated as a method to increase CNN robustness to scanner and Immunohistochemistry (IHC)-based image variability.

Methods

CST was applied to segment epithelium in immunohistological cervical Whole Slide Images (WSIs). CST randomly distorts input tiles and factors the difference between the CNN prediction for the original and distorted inputs within the loss function. CNNs were trained using 114 p16-stained WSIs from the same scanner, and evaluated on 6 WSI test sets, each with 23 to 24 WSIs of the same tissue but different scanner/IHC combinations. Relative robustness (rAUC) was measured as the difference between the AUC on the training domain test set (i.e., baseline test set) and the remaining test sets.

Results

Across all test sets, The AUC of CST models outperformed "No CST" models (AUC: 0.940-0.989 vs. 0.905-0.986, p < 1e - 8), and obtained an improved robustness (rAUC: [-0.038, -0.003] vs. [-0.081, -0.002]). At a WSI level, CST models showed an increase in performance in 124 of the 142 WSIs. CST models also outperformed models trained with random on-the-fly data augmentation (DA) in all test sets ([0.002, 0.021], p < 1e-6).

Conclusion

CST offers a path to improve CNN performance without the need for more data and allows customizing distortions to specific use cases. A python implementation of CST is publicly available at https://github.com/TIGACenter/CST_v1.

Infiltration by Intratumor and Stromal CD8 and CD68 in Cervical Cancer

Background and Objectives: The tumor microenvironment (TME) plays a major role in neoplastic development. Various types of cells can be found in the TME. These cells can be classified into two groups, immunosuppressive and immunostimulatory types, depending on the function they perform in the antitumor immune response (IR). By interacting both with each other and with tumor cells, different immune mechanisms are activated or inhibited, which can suppress or promote the development and progression of cervical cancer (CC). Our aim was to investigate some of the main components of the cellular immune response in TME-tumor-infiltrating cytotoxic T cells (Tc, CD8+) and tumor-associated macrophages (TAMs, CD68+)-in patients with CC. Materials and Methods: We analyzed 72 paraffin-embedded tumor tissues of patients diagnosed and treated at Medical University Pleven, Bulgaria. Patients were classified according to the 2018 FIGO (International Federation of Gynaecology and Obstetrics) classification. From each patient, we selected one histological slide with hematoxylin eosin staining. In a microscopic evaluation, CD8+ T lymphocytes and CD68+-positive macrophages were counted in the tumor and stroma of five randomly selected fields at ×40 magnification (HPF). We analyzed the relationship between intratumoral and stromal CD8 and CD68 expression and FIGO stage and N status. Results: There was no significant association between the expression levels of intratumoral and stromal CD68+ cells in the different FIGO stages and according to the lymph nodes' involvement. For CD8+ cells, the association of stromal infiltration was also not found, but T intratumor infiltration was associated with a higher FIGO stage, despite the fact that the results did not reach significance (p = 0.063, Fisher test). Intratumoral CD8+ cells were significantly associated with positive N status, (p = 0.035). Discussion: The separation of tumor-infiltrating cytotoxic T cells and tumor-associated macrophages into intratumoral and stromal is inconsequential. In our study, the level of infiltration of CD68+ cells in tumors and stromata was not significantly associated with tumor progression or lymph node involvement. The results were different for CD8+ cells, in which levels of infiltration were associated with lymph nodes' statuses. Conclusions: The separate evaluation of CD68+ immune cells in the TME as intratumoral and stromal is not beneficial for defining prognoses, since the presence of these cells is not associated with the patient's stage. In our study, the presence of CD8+ cells was significantly associated with lymph node metastases. The prognostic value of the obtained results can be enriched with an additional study of the lymphocyte phenotype, including B and other subtypes of T lymphocytes, NK cells, as well as molecules involved in the immune response, such as HLA subtypes.

Increased Early Cancer Diagnosis: Unveiling Immune-Cancer Biology to Explain Clinical "Overdiagnosis"

Even though clinically small 'early' cancers represent many millions of cells biologically, when removed surgically, these often never recur or regrow, nor reduce the individual's lifespan. However, some early cancers remain quiescent and indolent; while others grow and metastasize, threatening life. Distinguishing between these different clinical behaviours using clinical/pathological criteria is currently problematic. It is reported that many suspicious lesions and early cancers are being removed surgically that would not threaten the patient's life. This has been termed 'overdiagnosis', especially in the sphere of cancer screening. Although a controversial and emotive topic, it poses clinical and public health policy challenges. The diagnostic differentiation between 'non-lethal' and 'lethal' tumor forms is generally impossible. One perspective gathering evidential support is that a dynamic balance exists between the immune response and malignant processes governing 'lethality', where many more cancers are produced than become clinically significant due to the immune system preventing their progression. Higher medical screening "diagnosis" rates may reflect lead-time effects, with more 'non-progressing' cancers detected when an early immune-cancer interaction is occurring. We present a model for this immune-cancer interaction and review 'excess' or 'overdiagnosis' claims that accompany increasingly sensitive diagnostic and screening technologies. We consider that immune tools should be incorporated into future research, with potential for immune system modulation for some early cancers.

Improved Prognosis of Treatment Failure in Cervical Cancer with Nontumor PET/CT Radiomics

Radiomics has been applied to predict recurrence in several disease sites, but current approaches are typically restricted to analyzing tumor features, neglecting nontumor information in the rest of the body. The purpose of this work was to develop and validate a model incorporating nontumor radiomics, including whole-body features, to predict treatment outcomes in patients with previously untreated locoregionally advanced cervical cancer. Methods: We analyzed 127 cervical cancer patients treated definitively with chemoradiotherapy and intracavitary brachytherapy. All patients underwent pretreatment whole-body ¹⁸F-FDG PET/CT. To quantify effects due to the tumor itself, the gross tumor volume (GTV) was directly contoured on the PET/CT image. Meanwhile, to quantify effects arising from the rest of the body, the planning target volume (PTV) was deformably registered from each planning CT to the PET/CT scan, and a semiautomated approach combining seed-growing and manual contour review generated whole-body muscle, bone, and fat segmentations on each PET/CT image. A total of 965 radiomic features were extracted for GTV, PTV, muscle, bone, and fat. Ninety-five patients were used to train a Cox model of disease recurrence including both radiomic and clinical features (age, stage, tumor grade, histology, and baseline complete blood cell counts), using bagging and split-sample-validation for feature reduction and model selection. To further avoid overfitting, the resulting models were tested for generalization on the remaining 32 patients, by calculating a risk score based on Cox regression and evaluating the c-index (c-index > 0.5 indicates predictive power). Results: Optimal performance was seen in a Cox model including 1 clinical biomarker (whether or not a tumor was stage III-IVA), 2 GTV radiomic biomarkers (PET gray-level size-zone matrix small area low gray level emphasis and zone entropy), 1 PTV radiomic biomarker (major axis length), and 1 whole-body radiomic biomarker (CT bone root mean square). In particular, stratification into high- and low-risk groups, based on the linear risk score from this Cox model, resulted in a hazard ratio of 0.019 (95% CI, 0.004, 0.082), an improvement over stratification based on clinical stage alone, which had a hazard ratio of 0.36 (95% CI, 0.16, 0.83). Conclusion: Incorporating nontumor radiomic biomarkers can improve the performance of prognostic models compared with using only clinical and tumor radiomic biomarkers. Future work should look to further test these models in larger, multiinstitutional cohorts.

The augment of regulatory T cells undermines the efficacy of anti-PD-L1 treatment in cervical cancer

Background

Immune checkpoint inhibitors have aroused great expectation of tumor eradication. However, the effect of anti-PD-L1 treatment for cervical cancer is unsatisfactory and the underlying antagonist to anti-PD-L1 efficacy is remained to be studied. Here, we investigated the anti-tumor effect of anti-PD-L1 treatment in cervical tumor model and identified the antagonist to the therapeutic efficacy of anti-PD-L1 treatment.

Results

We found that PD-L1 exhibited a moderate expression in both cervical tumor cell lines and clinical samples compared to other tumor types and the para-tumor tissue respectively. Interestingly, our results showed that the anti-PD-L1 treated mice were dichotomously divided into responsive and unresponsive group after five cycles of anti-PD-L1 treatment although all the mice had the same genome background. In addition, the unresponsive tumors showed less tumor necrosis area and higher immunosuppression activity induced by regulatory T cells (Tregs) population than the responsive ones. Furthermore, we found that anti-PD-L1 treatment autonomously upregulated Tregs proliferation and frequency in multiple immune organs, and, most importantly, Tregs depletion significantly depressed the tumor growth rate and tumor weight compared with either anti-PD-L1 or anti-CD25 treatment alone. Finally, we observed that the upregulating effector CD8⁺ T cell is associated with the better therapeutic effect of anti-PD-L1 therapy post Tregs depletion.

Conclusions

Anti-PD-L1 treatment upregulates Tregs frequency and proliferation in tumor model, and the depletion of Tregs may be a useful adjuvant strategy for anti-PD-L1 therapy of cervical cancer.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12865-021-00451-7.

Construction of Immune-Associated Nomogram for Predicting the Recurrence Survival Risk of Stage I Cervical Cancer

Background

Various studies reported that the prognosis of patients with cervical cancer (CC) was significantly associated with immunity, whereas limited studies have explored whether immune-associated genes could be classifiers for recurrence-free survival (RFS) of stage I CC. Thus, an improved immune-related gene signature for stage I CC patients' prognosis is urgently required.

Materials and Methods

We retrospectively analyzed the gene expression profiles of stage I CC patients in the GSE44001 set from the Gene Expression Omnibus (GEO) database. The stage I CC patients were randomly divided into the training group and the internal validation group. The training patients were adopted to develop a prognostic immune gene-based signature; meanwhile, the internal validation patients were used to validate the power of the selected immune gene-related signature using univariate Cox proportional hazard analysis, least absolute shrinkage and selection operator (LASSO), and multivariate Cox regression analysis. The accuracy and reliability of the immune gene-related signature were evaluated based on Kaplan-Meier analysis and time-dependent receiver operating characteristic (ROC) curves.

Results

High power of the 8-immune gene signature was found on the basis of ROC analysis (AUC at 1, 3, and 5 years were exhibited in the internal validation group (0.702, 0.715, and 0.728, respectively), external validation group (0.702, 0.825, and 0.842, respectively), and entire GEO dataset (0.840, 0.894, and 0.852, respectively)). Besides, C-index, ROC, calibration plots, and decision curve analysis (DCA) also acted well in our nomogram, suggestive of a high ability of the nomogram to elevate the prognostic prediction of stage I CC patients.

Conclusions

In this study, we successfully constructed an integrated 8-immune gene-based signature which could accurately identify patients with low prognostic risk from those with high prognostic risk. In addition, we developed an immune-related nomogram which can elevate the prognostic prediction of stage I CC patients.

Response to Neoadjuvant Chemotherapy in Locally Advanced Cervical Cancer: The Role of Immune-related Factors

BACKGROUND/AIM

Treatment of locally advanced cervical cancer (LACC) consists of concomitant chemoradiation or neoadjuvant chemotherapy (NACT) plus radical surgery (RS). This study analyzed the prognostic role of neutrophil to lymphocyte ratio (NLR), platelet to lymphocyte ratio (PLR), tumor infiltrating lymphocytes (TILs), and PD-L1 expression in LACC patients, treated with NACT+RS.

PATIENTS AND METHODS

We prospectively analyzed 37 LACC patients treated from December 2016 to September 2019. Patients were submitted to pelvic examination, biopsy and imaging.

RESULTS

In 65% of cases, a nodal involvement was present at pre-treatment MRI. All cancers showed the presence of stromal TILs and PD-L1 staining of inflammatory cells. No significant correlations were found between clinicopathological parameters and the number of TILs and PDL-1 at baseline. After NACT, 29 patients (78%) were submitted to RS; 28% of patients showed pathological complete response, 62% partial response and 10% stable disease. Seven (24%) patients reported a positive node. Patients with high levels of stromal TILs and low NLR and PLR showed a significantly better response to NACT. No significant correlation was observed between PD-L1 expression and response to NACT.

CONCLUSION

The number of TILs, the expression of PDL1, and NLR and PLR ratios correlate significantly with the response of LACC patients to NACT.

The Immune Microenvironment in Human Papilloma Virus-Induced Cervical Lesions-Evidence for Estrogen as an Immunomodulator

Globally, human papilloma virus (HPV) infection is a common sexually transmitted disease. However, most of the HPV infections eventually resolve aided by the body’s efficient cell-mediated immune responses. In the vast majority of the small group of patients who develop overt disease too, it is the immune response that culminates in regression of lesions. It is therefore a rarity that persistent infection by high-risk genotypes of HPV compounded by other risk factors progresses through precancer (various grades of cervical intraepithelial neoplasia—CIN) to cervical cancer (CxCa). Hence, although CxCa is a rare culmination of HPV infection, the latter is nevertheless causally linked to >90% of cancer. The three ‘Es’ of cancer immunoediting viz. elimination, equilibrium, and escape come into vogue during the gradual evolution of CIN 1 to CxCa. Both cell-intrinsic and extrinsic mechanisms operate to eliminate virally infected cells: cell-extrinsic players are anti-tumor/antiviral effectors like Th1 subset of CD4+ T cells, CD8+ cytotoxic T cells, Natural Killer cells, etc. and pro-tumorigenic/immunosuppressive cells like regulatory T cells (Tregs), Myeloid-Derived Suppressor Cells (MDSCs), type 2 macrophages, etc. And accordingly, when immunosuppressive cells overpower the effectors e.g., in high-grade lesions like CIN 2 or 3, the scale is tilted towards immune escape and the disease progresses to cancer. Estradiol has long been considered as a co-factor in cervical carcinogenesis. In addition to the gonads, the Peyer’s patches in the gut synthesize estradiol. Over and above local production of the hormone in the tissues, estradiol metabolism by the gut microbiome: estrobolome versus tryptophan non-metabolizing microbiome, regulates free estradiol levels in the intestine and extraintestinal mucosal sites. Elevated tissue levels of the hormone serve more than one purpose: besides a direct growth-promoting action on cervical epithelial cells, estradiol acting genomically via Estrogen Receptor-α also boosts the function of the stromal and infiltrating immunosuppressive cells viz. Tregs, MDSCs, and carcinoma-associated fibroblasts. Hence as a corollary, therapeutic repurposing of Selective Estrogen Receptor Disruptors or aromatase inhibitors could be useful for modulating immune function in cervical precancer/cancer. The immunomodulatory role of estradiol in HPV-mediated cervical lesions is reviewed.

Six-lncRNA Immune Prognostic Signature for Cervical Cancer

Background

This study searched for immune-related long noncoding RNAs (lncRNAs) to predict the prognosis of patients with cervical cancer.

Method

We obtained immunologically relevant lncRNA expression profiles and clinical follow-up data from cervical cancer patients from The Cancer Genome Atlas database and the Molecular Signatures Database. Cervical cancer patients were randomly divided into a training group, testing group and combined group. The immune prognostic signature was constructed by Least Absolute Shrinkage and Selection Operator Cox regression, prognosis was analyzed by Kaplan-Meier curves between different groups, and the accuracy of the prognostic model was assessed by receiver operating characteristic-area under the curve (ROC-AUC) analysis.

Results

A six-lncRNA immune prognostic signature (LIPS) was constructed to predict the prognosis of cervical cancer. The six lncRNAs are as follows: AC009065.8, LINC01871, MIR210HG, GEMIN7-AS1, GAS5-AS1, and DLEU1. A ROC-AUC analysis indicated that the model could predict the prognosis of cervical cancer patients in different subgroups. A Kaplan-Meier analysis showed that patients with high risk scores had a poor prognosis; these results were equally meaningful in the subgroup analyses. Risk scores differed depending on the clinical pathology and tumor grade and were independent risk factors for cervical cancer prognosis. Gene set enrichment analysis revealed an association between the LIPS and the immune response, Wnt signaling pathway, and TGF beta signaling pathway.

Conclusion

Our study shows that the six-LIPS can predict the prognosis of cervical cancer and contribute to decisions regarding the immunotherapeutic strategy.

Tackling Resistance to Cancer Immunotherapy: What Do We Know?

Cancer treatment has evolved tremendously in the last few decades. Immunotherapy has been considered to be the forth pillar in cancer treatment in addition to conventional surgery, radiotherapy, and chemotherapy. Though immunotherapy has resulted in impressive response, it is generally limited to a small subset of patients. Understanding the mechanisms of resistance toward cancer immunotherapy may shed new light to counter that resistance. In this review, we highlighted and summarized two major hurdles (recognition and attack) of cancer elimination by the immune system. The mechanisms of failure of some available immunotherapy strategies were also described. Moreover, the significance role of immune compartment for various established cancer treatments were also elucidated in this review. Then, the mechanisms of combinatorial treatment of various conventional cancer treatment with immunotherapy were discussed. Finally, a strategy to improve immune cancer killing by characterizing cancer immune landscape, then devising treatment based on that cancer immune landscape was put forward.

Cervical Cancer: Emerging Immune Landscape and Treatment

Immune cells are essential for defending the body’s balance and have increasingly been implicated in controlling tumor growth. In cervical cancer (CC), the immune landscape is extensively connected with human papillomavirus (HPV) status. Recent insights from studies have revealed that as a result of infection with HPV, immune cell populations such as lymphocytes or monocytes change during carcinogenesis. Immune therapy, in particular checkpoint inhibitors, those targeting PD-1 or PD-L1, has shown promising efficacy. This article reviews the immune landscape and immunotherapy of CC.

Immunologic impact of chemoradiation in cervical cancer and how immune cell infiltration could lead toward personalized treatment

We investigated the potential of tumor-infiltrating immune cells (ICs) as predictive or prognostic biomarkers for cervical cancer patients. In total, 38 patients treated with (chemo)radiotherapy and subsequent surgery were included in the current study. This unique treatment schedule makes it possible to analyze IC markers in pretreatment and posttreatment tissue specimens and their changes during treatment. IC markers for T cells (CD3, CD4, CD8 and FoxP3), macrophages (CD68 and CD163) and B cells (CD20), as well as IL33 and PD-L1, were retrospectively analyzed via immunohistochemistry. Patients were grouped in the low score or high score group based on the amount of positive cells on immunohistochemistry. Correlations to pathological complete response (pCR), cause-specific survival (CSS) and metastasis development during follow-up were evaluated. In analysis of pretreatment biopsies, significantly more pCR was seen for patients with CD8 = CD3, CD8 ≥ CD4, positive IL33 tumor cell (TC) scores, IL33 IC < TC and PD-L1 TC ≥5%. Besides patients with high CD8 scores, also patients with CD8 ≥ CD4, CD163 ≥ CD68 or PD-L1 IC ≥5% had better CSS. In the analysis of posttreatment specimens, less pCR was observed for patients with high CD8 or CD163 scores. Patients with decreasing CD8 or CD163 scores between pretreatment and posttreatment samples showed more pCR, whereas those with increasing CD8 or decreasing IL33 IC scores showed a worse CSS. Meanwhile, patients with an increasing CD3 score or stable/increasing PD-L1 IC score showed more metastasis during follow-up. In this way, the intratumoral IC landscape is a promising tool for prediction of outcome and response to (chemo)radiotherapy.