Search

Abstract Background Histological images show strong variance (e.g. illumination, color, staining quality) due to differences in image acquisition, tissue processing, staining, etc. This can impede downstream image analysis such as staining intensity evaluation or classification. Methods to reduce these variances are called image normalization techniques. Methods In this paper, we investigate the potential of CycleGAN (cycle consistent Generative Adversarial Network) for color normalization in hematoxylin-eosin stained histological images using daily clinical data with consideration of the variability of internal staining protocol variations. The network consists of a generator network G B that learns to map an image X from a source domain A to a target domain B, i.e. G B :X A →X B . In addition, a discriminator network D B is trained to distinguish whether an image from domain B is real or generated. The same process is applied to another generator-discriminator pair (G A ,D A ), for the inverse mapping G A :X B →X A . Cycle consistency ensures that a generated image is close to its original when being mapped backwards (G A (G B (X A ))≈X A and vice versa). We validate the CycleGAN approach on a breast cancer challenge and a follicular thyroid carcinoma data set for various stain variations. We evaluate the quality of the generated images compared to the original images using similarity measures. In addition, we apply stain normalization on pathological lymph node data from our institute and test the gain from normalization on a ResNet classifier pre-trained on the Camelyon16 data set. Results Qualitative results of the images generated by our network are compared to original color distributions. Our evaluation indicates that by mapping images to a target domain, the similarity training images from that domain improves up to 96%. We also achieve a high cycle consistency for the generator networks by obtaining similarity indices greater than 0.9. When applying the CycleGAN normalization to HE-stain images from our institute the kappa-value of the ResNet-model that is only trained on Camelyon16 data is increased more than 50%. Conclusions CycleGANs have proven to efficiently normalize HE-stained images. The approach compensates for deviations resulting from image acquisition (e.g. different scanning devices) as well as from tissue staining (e.g. different staining protocols), and thus overcomes the staining variations in images from various institutions.The code is publicly available at https://github.com/m4ln/stainTransfer_CycleGAN_pytorch . The data set supporting the solutions is available at https://doi.org/10.11588/data/8LKEZF .

In recent years an increase of functional CD4+CD25+ regulatory T cells (Treg cells) has been established for patients with solid tumors, acute leukemias, and lymphomas. We have reported an expanded pool of CD4+CD25high Treg cells in patients with chronic lymphatic leukemia (CLL), multiple myeloma (MM) as well as its premalignant precursor monoclonal gammopathy of undetermined significance (MGUS). In healthy individuals, low-level expression of CD127 on T cells in addition to the expression of FOXP3 has been associated with Treg cells. Here, we demonstrate that the expanded FOXP3+ T-cell population in patients with colorectal cancer, CLL, MGUS, MM, follicular lymphoma, and Hodgkin's disease are exclusively CD127low Treg cells and were strongly suppressive. A significant portion of CD127lowFOXP3+ Treg cells expressed only low levels of CD25 suggesting that the previously reported expansion of CD25+ Treg cells underestimates the true expansion. The assessment of CCR7 and CD45RA expression on the expanded CD4+CD127lowFOXP3+ Treg cells revealed an increase of both naïve as well as central and effector memory Treg cells in peripheral blood. Our data strongly support superiority of combined CD127 and FOXP3 analysis in comparison to CD25 and FOXP3 assessment for further quantification of Treg cells in malignant diseases.

Regulatory T cells (T(reg) cells) are increased in context of malignancies and their expansion can be correlated with higher disease burden and decreased survival. Initially, interleukin 2 (IL-2) has been used as T-cell growth factor in clinical vaccination trials. In murine models, however, a role of IL-2 in development, differentiation, homeostasis, and function of T(reg) cells was established. In IL-2 treated cancer patients a further T(reg)-cell expansion was described, yet, the mechanism of expansion is still elusive. Here we report that functional T(reg) cells of a naïve phenotype--as determined by CCR7 and CD45RA expression--are significantly expanded in colorectal cancer patients. Treatment of 15 UICC stage IV colorectal cancer patients with IL-2 in a phase I/II peptide vaccination trial further enlarges the already increased naïve T(reg)-cell pool. Higher frequencies of T-cell receptor excision circles in naïve T(reg) cells indicate IL-2 dependent thymic generation of naïve T(reg) cells as a mechanism leading to increased frequencies of T(reg) cells post IL-2 treatment in cancer patients. This finding could be confirmed in naïve murine T(reg) cells after IL-2 administration. These results point to a more complex regulation of T(reg) cells in context of IL-2 administration. Future strategies therefore might aim at combining IL-2 therapy with novel strategies to circumvent expansion and differentiation of naïve T(reg) cells.