Letter to the editor with regard to the article entitled “Sézary syndrome and mycosis fungoides: An overview, including the role of immunophenotyping”,Cytometry Part B: Clinical Cytometry

当前位置： X-MOL 学术 › Cytom. Part B Clin. Cytom. › 论文详情

Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)

Letter to the editor with regard to the article entitled “Sézary syndrome and mycosis fungoides: An overview, including the role of immunophenotyping”
Cytometry Part B: Clinical Cytometry ( IF 2.3 ) Pub Date : 2020-10-08 , DOI: 10.1002/cyto.b.21962
Marie Roelens ₁ , Adèle de Masson _{2,

3} , Caroline Ram-Wolff _{2,

3} , Martine Bagot _{2,

3} , Hélène Moins-Teisserenc _{1,

4}

Affiliation

Pulitzer et al. published into Cytometry Part B: Clinical Cytometry on June 9, 2020, an overview on Sézary syndrome (SS) and mycosis fungoides (MF), with a focus on the latest advances concerning the role of immunophenotyping in their diagnosis. We were very interested in how the authors discussed the complexity of Sézary cells (SCs) and their extreme heterogeneity. Besides the detection of a malignant clone in blood and skin, using molecular based technologies, flow cytometry (FC) has been widely developed for the diagnosis and patients' follow‐up. We agree with the authors, claiming that the currently used FC markers (lack of CD7 and CD26) recommended by ISCL/EORTC, are not able to discriminate SCs from benign cells, which may have the same aberrant phenotypes with defects in non‐specific marker expression. However, since the first description of KIR3DL2/CD158k by Bagot et al. in 2001, we, and other teams, have repeatedly demonstrated the reliability of this marker for the detection of SCs in blood and skin. In healthy donors, KIR3DL2/CD158k expression is highly restricted on minor NK‐cell and CD8 + T‐cell subsets and on rare CD3 + CD4 + T‐cells. In SS patients, we found that the percentages and absolute numbers of KIR3DL2/CD158k cells were strongly correlated to the percentages and absolute numbers of clonal circulating SCs (as evidenced by the use of clonal‐specific anti‐TCR‐Vβ chain antibodies) and that (not NK, not CD8+ and CD4+/−) KIR3DL2/CD158k + lymphocytes corresponded to the malignant clonal cell population (Moins‐Teisserenc et al., 2015). We do not agree with the authors saying that KIR3DL2/CD158k has not been adequately analyzed in normal CD4+ T‐cells, as we reported extended characterization of the KIR3DL2 ̶ TCRV‐Vβ ̶ “benign” counterpart of SCs. In addition, analyses of 40 healthy donors showed that the mean percentage of CD4+ T‐cells expressing KIR3DL2/CD158k is 1.5% (min–max = 0–2.5%) (our unpublished data).

The review also cites the work of Boonk et al., evaluating, among other markers, KIR3DL2/CD158k for SS and reporting that the results were not conclusive, as SCs identification was not accurate enough. Of note, this study was conducted on frozen peripheral blood mononuclear cells (PBMCs) instead of freshly isolated PBMCs. Since then, a novel anti‐KIR3DL2/CD158k monoclonal antibody (murine IgG1), termed MOG1‐M‐K322‐13E4, was developed in 2014 by Innate Pharma (Marseille, France), which greatly improved the specificity and the sensitivity of SCs immunostaining, both on fresh and frozen samples. Since then, KIR3DL2/CD158k is part of the routine care of erythrodermic patients at Hospital Saint‐Louis, which is the French national reference center for CTCL.

In our paper reporting results of 254 consecutive patients at initial diagnosis with suspected CTCL, we propose to add KIR3DL2/CD158k in a novel algorithm for blood staging, together with CD7 and CD26 markers (Roelens et al. 2019). We also revisited the specificity and the sensitivity of CD7 and CD26 in the particular case of CD4 + T‐cell counts <1,600/μl at initial diagnosis and during therapy. Our results again, emphasize the reliability of KIR3DL2/CD158k, whereas the lack of CD26 remains informative only when results are given as percentages of total CD4+ T‐cells.

We acknowledge Pulitzer et al. to have mentioned our work describing SCs diversity in naive/memory maturation phenotypes and molecular signature, as well as cytokine/chemokine receptor expression but we would like to stress out that we already described this heterogeneity for the first time in 2015, with SCs expressing naive markers in some patients, while SCs have been largely described elsewhere as TCM cells exclusively (Moins‐Teisserenc et al., 2015). In addition, we demonstrated that skin‐derived SCs showed a more advanced maturation profile than blood‐derived SCs, with effector memory and resident memory phenotypes, which are not the hallmark of skin infiltrating T‐cells as usually described in MF.

Finally, Pulitzer et al. argued that distinction between SCs and normal/reactive CD4+ T‐cells significantly improves when a more extensive phenotypic profile (and not just individual parameters) is considered. This would make sense if at least one of the markers mentioned remained relatively constant over time. In our hands, benign CD4+ T‐cells in SS patients often display aberrant phenotypes, not only concerning CD7 and CD26, but also the usual markers of naive/memory cells and exhaustion, which resemble to SCs (Figure 1) (Roelens et al., personal communication). Although surprising, these phenotypes may play a role in the well documented immunodepression of SS patients. In addition, benign and malignant CD4+ T‐cells often evolve in the same way under treatment. As these cells may be affected by the same variations of cytokine environment, such common phenotypes question the reliability of extended phenotype using only nonspecific markers for SCs. We are currently monitoring these populations with regard to the tumor burden evolution (manuscript in preparation).

**FIGURE 1**
Open in figure viewerPowerPoint

Example of an extended immune‐profile by FC viewed sequentially in a Sézary patient: (a) Sézary cells defined as CD4+TCRVβ+KIR3DL2+, (b) “Benign” CD4+T‐cells defined as TCRVβ ̶ KIR3DL2 ̶ CD4+ T‐cells; (c) immune‐profile of an age‐matched Healthy Donor (HD). Note the percentage of KIR3DL2+ cells in HD (1% of total CD4+ T‐cells). In this example, the lymphocyte subset defined as CD4+CD45RA‐CD26‐CD27+CD28+PD1+ represents 58.5% and 49.2% of total SCs and “Benign” CD4+T‐cells, respectively, in the Sézary patient, whereas it represents only 3.8% of total CD4+ T‐cells from the HD (right panel). Therefore, even if extended phenotypes are performed, the distinction between malignant and benign cells remains challenging in the absence of specific positive markers [Color figure can be viewed at wileyonlinelibrary.com]

As opposed to FC markers recommended by the authors (mainly negative/down‐regulated cell‐surface molecules), tracking SCs with KIR3DL2/CD158k does not take into‐account benign cells which may have the same aberrant phenotypes with defects in nonspecific marker expression. Indeed, the decrease of such T‐cells may not indicate clearance of neoplastic cells and conversely, the presence of these cells may underestimate the rate of blood response.

In conclusion, the early diagnosis and follow‐up of SS/MF is still challenging, because of a lack, in most laboratories, of a truly specific surface marker for malignant cells. In such circumstances some patients are diagnosed with a delay, even treated with inappropriate therapies which worsens their prognosis.

In response to Pulitzer et al., we recommend to introduce in further studies KIR3DL2/CD158k as a positive marker for SCs identification using FC.

中文翻译：

给编辑的信，标题为“塞氏综合征和真菌病：概述，包括免疫表型的作用”

普利策等。于2020年6月9日发表在《细胞计数法B部分：临床细胞计数法》中，概述了塞扎里综合症（SS）和真菌病（MF），重点是关于免疫表型在其诊断中的作用的最新进展。我们对作者如何讨论Sézary细胞（SC）的复杂性及其极端异质性非常感兴趣。除了检测血液和皮肤中的恶性克隆外，使用基于分子的技术，流式细胞术（FC）已广泛用于诊断和患者随访。我们同意作者的观点，声称ISCL / EORTC建议使用的当前使用的FC标记（缺少CD7和CD26）无法将SC与良性细胞区分开，良性细胞可能具有相同的异常表型，但非特异性标记存在缺陷表达。然而，自Bagot等人首次对KIR3DL2 / CD158k进行描述以来。在2001年，我们和其他团队反复证明了该标记物在血液和皮肤中SC的检测中的可靠性。在健康的供体中，KIR3DL2 / CD158k的表达高度限制于次要的NK细胞和CD8 + T细胞亚群以及稀有的CD3 + CD4 + T细胞。在SS患者中，我们发现KIR3DL2 / CD158k细胞的百分比和绝对数量与克隆循环SC的百分比和绝对数量密切相关（如使用克隆特异性抗TCR-Vβ链抗体所证明的那样），并且（不是NK，不是CD8 +和CD4 +/-）KIR3DL2 / CD158k +淋巴细胞对应于恶性克隆细胞群（Moins-Teisserenc等，反复证明了该标记物在检测血液和皮肤中SC方面的可靠性。在健康的供体中，KIR3DL2 / CD158k的表达高度限制于次要的NK细胞和CD8 + T细胞亚群以及稀有的CD3 + CD4 + T细胞。在SS患者中，我们发现KIR3DL2 / CD158k细胞的百分比和绝对数量与克隆循环SC的百分比和绝对数量密切相关（如使用克隆特异性抗TCR-Vβ链抗体所证明的那样），并且（不是NK，不是CD8 +和CD4 +/-）KIR3DL2 / CD158k +淋巴细胞对应于恶性克隆细胞群（Moins-Teisserenc等，曾多次证明该标记物可用于检测血液和皮肤中的SC。在健康的供体中，KIR3DL2 / CD158k的表达高度限制于次要的NK细胞和CD8 + T细胞亚群以及稀有的CD3 + CD4 + T细胞。在SS患者中，我们发现KIR3DL2 / CD158k细胞的百分比和绝对数量与克隆循环SC的百分比和绝对数量密切相关（如使用克隆特异性抗TCR-Vβ链抗体所证明的那样），并且（不是NK，不是CD8 +和CD4 +/-）KIR3DL2 / CD158k +淋巴细胞对应于恶性克隆细胞群（Moins-Teisserenc等，2015）。我们不同意作者所说的在正常CD4 + T细胞中未对KIR3DL2 / CD158k进行充分分析的原因，因为我们报道了KIR3DL2̶TCRV-VβSC SC的“良性”对应物的扩展特征。此外，对40位健康供体的分析表明，表达KIR3DL2 / CD158k的CD4 + T细胞的平均百分比为1.5％（最小-最大= 0-2.5％）（我们未发表的数据）。

该评论还引用了Boonk等人的工作，除其他标记外，还评估了SS的KIR3DL2 / CD158k并报告了结果尚不确定，因为SC的识别不够准确。值得注意的是，这项研究是针对冷冻的外周血单个核细胞（PBMC），而不是新鲜分离的PBMC。此后，Innate Pharma（法国马赛）于2014年开发了一种名为MOG1-M-K322-13E4的新型抗KIR3DL2 / CD158k单克隆抗体（鼠IgG1），极大地提高了SCs免疫染色的特异性和敏感性。，无论是新鲜样品还是冷冻样品。从那以后，KIR3DL2 / CD158k成为法国圣路易斯医院的全国参考中心圣路易斯医院的红皮病患者常规护理的一部分。

在我们的254例首次诊断为可疑CTCL的连续诊断患者的报告结果中，我们建议在一种新的血液分期算法中将KIR3DL2 / CD158k与CD7和CD26标记一起添加（Roelens等人2019）。我们还重新探讨了CD7和CD26在CD4 + T细胞计数<1,600 /μl的特殊情况下在初诊和治疗期间的特异性和敏感性。我们的结果再次强调了KIR3DL2 / CD158k的可靠性，而仅当结果以总CD4 + T细胞的百分比给出时，CD26的缺乏仍然是有益的。

我们承认普利策等人。提到我们的工作描述了SC在原始/记忆成熟表型和分子标记以及细胞因子/趋化因子受体表达方面的多样性，但我们想强调的是，我们已经在2015年首次描述了这种异质性，其中SC表达原始在某些患者中，SCs是一种标记物，而在其他地方，SCs在很大程度上仅被描述为TCM细胞（Moins-Teisserenc等人，2015年）。此外，我们证明皮肤源性SCs比血液源性SCs具有更高级的成熟特征，具有效应记忆和常驻表型，这不是MF中通常描述的皮肤浸润T细胞的标志。

最后，普利策等。有人认为，考虑到更广泛的表型特征（而不仅仅是个别参数）时，SC与正常/反应性CD4 + T细胞之间的区别会大大改善。如果提到的至少一个标记在一段时间内保持相对恒定，这将是有道理的。在我们手中，SS患者的良性CD4 + T细胞经常表现出异常的表型，不仅涉及CD7和CD26，而且还显示幼稚/记忆细胞和衰竭的常见标志物，类似于SC（图1）（Roelens等。，个人通讯）。尽管令人惊讶，但这些表型可能在充分记录的SS患者的免疫抑制中起作用。此外，良性和恶性CD4 + T细胞在治疗中通常以相同的方式进化。由于这些细胞可能会受到细胞因子环境的相同变化的影响，因此此类常见表型仅使用SC的非特异性标记物质疑扩展表型的可靠性。我们目前正在监测这些人群的肿瘤负荷演变情况（准备中的手稿）。