Dear Editor, Sézary syndrome (SS) is a rare cutaneous T‐cell lymphoma (CTCL) grouped with mycosis fungoides (MF) in the international classification and staging criteria of CTCL. 1 , 2 Large‐cell transformation (LCT) has been widely described in MF and associated with reduced overall survival, suggesting the importance of early and sequential histological screening of LCT in MF. 3 However, LCT has never been studied and characterized in a large cohort of SS. Another concern is that LCT in MF was defined in the 1980s using diagnostic criteria for LCT in follicular lymphoma. 4 Although widely used since, the reliability of these criteria has never been specifically studied in SS. Additionally, the presence of large circulating Sézary cells (SCs) based on cytomorphological and flow cytometry analysis was independently associated with poor outcome and might predict LCT occurrence in skin. 5 , 6 Nevertheless, the prognostic impact of structure parameters [forward scatter (FSC) and side scatter (SSC)] of circulating cells in cytometry and correlation with LCT remains to be determined in SS.
The main objective of our study was to characterize LCT in SS. All patients with SS diagnosed at Saint‐Louis hospital (Paris, France) between 1998 and 2020 according to European Organisation for Research and Treatment of Cancer–World Health Organization criteria were included. The gating strategy of KIR3DL2+ SC among lymphocytes was previously described. 7 Circulating KIR3DL2+ SC ≥ 200 mm−3 was used to define KIR3DL2‐positive status. 8 For each patient, all skin biopsy samples performed were included. LCT was histologically defined by the presence in the lymphocytes’ infiltrate of > 25% or aggregates/nodules of large cells (more than four times the diameter of a small lymphocyte). 3 , 4 Haematoxylin‐, eosin‐ and safran‐stained slides were then digitized and an analysis using HALO software was performed. All blood samples from patients with flow cytometry data between 2015 and 2020 were included for FSC/SSC analysis. This study received the Institutional Review Committee agreement (LYMPHOTEQ reference: CPP 2019‐AO1158‐49).
In total, 117 patients were included with a median follow‐up of 41 months (interquartile range 1–81). Overall, 6% (six of 100) and 16% (18 of 112) of patients were diagnosed with LCT on skin biopsy samples at diagnosis and during follow‐up, respectively. Interobserver reliability between two independent pathologists was excellent [k = 0·88; 95% confidence interval (CI) 0·78–0·98]. Considering all skin biopsy samples at diagnosis, CD30 > 10% and Ki67 > 20% was more frequent in LCT+ samples than in LCT− samples (67% vs. 9%; P = 0·003 and 100% vs. 22%; P = 0·016, respectively).
We then compared visual histopathological analysis with digital pathology analysis of skin biopsy samples (n = 189). Mean cell surface was significantly higher in LCT+ than in LCT− skin biopsy sample images [27 μm2 (SD 3·3) vs. 22 μm2 (SD 2·3); P < 0·001].
Subsequently, we compared cell size between blood and skin compartments on 231 blood samples from 112 patients with SS. The maximal mean FSC value of circulating tumour cells in patients who were LCT+ prior to LCT occurrence was not significantly higher than that found in patients who were LCT− [561 273 (SD 99 359) vs. 531 839 (SD 68 040)] (P = 0·37). Among patients with a maximum mean FSC value < 600 000 during follow‐up, 17% (10 of 60) subsequently presented LCT vs. 14% (one of seven) of patients with a maximum mean FSC value ≥ 600 000 (P > 0·99). We obtained similar results for SSC.
Finally, median overall survival was shorter in patients who were LCT+ than in those who were LCT− at diagnosis (35 months vs. 80 months) (HR 9·5, 95% CI 1·9–47·1; P = 0·006). In multivariate analysis, age > 60 years (HR 4·46, 95% CI 1·14–17·41; P = 0·031), elevated LDH level (HR 2·63, 95% CI 1·03–6·72; P = 0·044), CD4 + CD26− circulating cell ≥ 10 000 mm−3 (HR 3·71, 95% CI 1·23–11·19; P = 0·020) and LCT at diagnosis (HR 4·77, 95% CI 1·11–20·4; P = 0·035) were independently associated with shorter overall survival (Table 1). During follow‐up, median overall survival after LCT occurrence was 21 months, with a 5‐year survival of 12% (95% CI 1–38).
Table 1.
Univariate and multivariate survival analysis
| Univariate analysis | Multivariate analysis | ||||||
|---|---|---|---|---|---|---|---|
| Variables | No. of patients with complete data | No. of patients (%) | Median survival, months | HR (95% CI) | P‐values | HR (95% CI) | P‐values |
| Sex | 117 | 1·3 (0·67–2·4) | 0·470 | ||||
| Male | 53 (45) | 77 | – | – | |||
| Female | 64 (56) | 150 | |||||
| Age, years | 117 | 3·3 (1·7–6·3) | < 0·001 | 4·46 (1·14–17·41) | 0·031 | ||
| > 60 | 38 (32) | 62 | |||||
| ≤ 60 | 79 (68) | 258 | |||||
| Stage | 103 | 7·4 (1·3–41·3) | 0·022 | 0·66 (0·16–2·7) | 0·562 | ||
| IVA2 | 10 (9) | 39 | |||||
| IVA1 | 93 (91) | 80 | |||||
| LDH | 97 | 2·4 (1·2–5·1) | 0·018 | 2·63 (1·03–6·72) | 0·044 | ||
| Elevated | 43 (44) | 42 | |||||
| Normal | 54 (56) | 92 | |||||
| Circulating KIR3DL2+ SC mm−3 | 87 | 5·9 (1·3–26·2) | 0·036 | – | – | ||
| ≥ 10 000 | 7 (8) | 39 | |||||
| < 10 000 | 80 (92) | 75 | |||||
| Circulating CD4 + CD26− cells mm−3 | 83 | 6·8 (1·7–27·7) | 0·007 | 3·71 (1·23–11·19) | 0·020 | ||
| ≥ 10 000 | 12 (14) | 35 | |||||
| < 10 000 | 71 (86) | 75 | |||||
| FSC baseline | 96 | 1·6 (0·4–6·4) | 0·537 | – | – | ||
| ≥ 600 000 | 9 (9) | NR | |||||
| < 600 000 | 87 (91) | 109 | |||||
| Large‐cell transformation at diagnosis | 100 | 9·5 (1·9–47·1) | 0·006 | 4·77 (1·11–20·4) | 0·035 | ||
| Yes | 6 (6) | 35 | |||||
| No | 94 (94) | 80 | |||||
| CD30 expression in skin at diagnosis | 63 | 0·8 (0·3–2·2) | 0·649 | – | – | ||
| > 10% | 9 (14) | 45 | |||||
| ≤ 10% | 54 (86) | 43 | |||||
| Ki67 in skin at diagnosis | 52 | 1·7 (0·5–6·3) | 0·323 | – | – | ||
| > 20% | 14 (27) | 41 | |||||
| ≤ 20% | 38 (73) | 55 | |||||
HR, hazard ratio; NR, not reached; LDH, lactate dehydrogenase; CI, confidence interval; FSC, forward scatter; SC, Sézary cell. P‐values < 0·05 are provided in bold [log‐rank (univariate) and Cox regression (multivariate) statistical test].
In conclusion, our study characterized for the first time LCT in a large homogeneous cohort of patients with SS using histological, digital pathology and flow cytometry approaches. LCT incidence in SS is lower than for MF. Studies specifically investigating LCT in advanced‐stage MF reported variable cumulative incidences ranging from 20% to 55% and large international cohorts have reported a variable incidence of LCT from 4·7% (all stages) to 20% (advanced stages) of MF/SS but without distinguishing the two entities. 2
Moreover, we have shown that LCT at diagnosis was an independent unfavourable prognostic factor that could not be predicted by the presence of large circulating tumour cells. In parallel, patients with visually diagnosed LCT on skin biopsy had lymphoid cells with a significantly higher mean surface in digital pathology analysis. The criteria used in MF for LCT diagnosis proved reliable and reproducible.
Thus, histological evaluation of LCT in SS, assisted by digital pathology, is a major means of prognostic staging at diagnosis and during follow‐up, allowing the early selection of patients that might benefit from therapeutic changes.
Author contributions
Christophe Bontoux: Data curation (equal); formal analysis (equal); investigation (equal); methodology (equal); software (equal); visualization (equal); writing – original draft (equal); writing – review and editing (equal). Adèle de Masson: Methodology (equal); supervision (equal); validation (equal); writing – review and editing (equal). Nicolas Thonnart: Data curation (equal); formal analysis (equal); software (equal). Caroline Ram‐Wolff: Data curation (equal); validation (equal). Flavien Caraguel: Formal analysis (equal); software (equal); validation (equal); visualization (equal). Luciana Batista: Formal analysis (equal); software (equal); validation (equal); visualization (equal). Sabrina Carpentier: Formal analysis (equal); software (equal); validation (equal); visualization (equal). Hélène Moins‐Teisserenc: Data curation (equal); validation (equal). Jacqueline Rivet: Data curation (equal); validation (equal). Marie‐Dominique Vignon‐Pennamen: Data curation (equal); validation (equal). Anne Marie‐Cardine: Data curation (equal); formal analysis (equal); investigation (equal); methodology (equal); software (equal); validation (equal); visualization (equal). Martine Bagot: Project administration (equal); resources (equal); validation (equal). Maxime Battistella: Conceptualization (equal); investigation (equal); methodology (equal); project administration (equal); resources (equal); supervision (equal); validation (equal); writing – review and editing (equal).
Funding sources: none.
Conflicts of interest: the authors declare they have no conflicts of interest.
Data availability: the data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.
Ethics statement: this study received Institutional Review Committee agreement (LYMPHOTEQ reference: CPP 2019‐AO1158‐49).
References
- 1. Olsen E, Whittaker S, Willemze R et al. Primary cutaneous lymphoma: recommendations for clinical trial design and staging update from the ISCL, USCLC, and EORTC. Blood 2021; DOI: 10.1182/blood.2021012057. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Scarisbrick J, Prince H, Vermeer M et al. Cutaneous lymphoma international consortium study of outcome in advanced stages of mycosis fungoides and Sézary syndrome: effect of specific prognostic markers on survival and development of a prognostic model. J Clin Oncol 2015; 33:3766–73. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Vergier B, de Muret A, Beylot‐Barry M et al. Transformation of mycosis fungoides: clinicopathological and prognostic features of 45 cases. Blood 2000; 95:2212–18. [PubMed] [Google Scholar]
- 4. Salhany KE, Cousar JB, Greer JP et al. Transformation of cutaneous T cell lymphoma to large cell lymphoma. A clinicopathologic and immunologic study. Am J Pathol 1988; 132:265–77. [PMC free article] [PubMed] [Google Scholar]
- 5. Clark RA, Shackelton JB, Watanabe R et al. High‐scatter T cells: a reliable biomarker for malignant T cells in cutaneous T‐cell lymphoma. Blood 2011; 117:1966–76. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Vonderheid EC, Jakubowski J, Hou JS. High‐scatter lymphocytes in the blood of erythrodermic cutaneous T‐cell lymphoma: evidence for large‐cell transformation? Clin Lymphoma Myeloma Leuk 2020; 20:624–31.e2. [DOI] [PubMed] [Google Scholar]
- 7. Moins‐Teisserenc H, Daubord M, Clave E et al. CD158k is a reliable marker for diagnosis of Sézary syndrome and reveals an unprecedented heterogeneity of circulating malignant cells. J Invest Dermatol 2015; 135:247–57. [DOI] [PubMed] [Google Scholar]
- 8. Roelens M, de Masson A, Ram‐Wolff C et al. Revisiting the initial diagnosis and blood staging of mycosis fungoides and Sézary syndrome with the KIR3DL2 marker. Br J Dermatol 2020; 182:1415–22. [DOI] [PubMed] [Google Scholar]