Introduction

Sezary syndrome (SS), commonly referred to as the leukemic variant of mycosis fungoides (MF), constitutes fewer than 5% of cutaneous T cell lymphoma CTCL [1]. The WHO/EORTC classifies SS as disease process characterized pathologically by the presence of Sezary cells (neoplastic T‐cells) in any or all of the skin, lymph nodes, and peripheral blood, and on physical examination by generalized lymphadenopathy, erythroderma [2]. SS is associated with very poor survival and a variable disease course, which may involve rapid progression in a subset of patients, with 5-year survival approximated at 11% [3,4].

“Large Cell Transformation” (LCT) can be diagnosed by the presence of large cells (defined as lymphocytes ≥ 4 times the size of normal lymphocytes) in ≥ 25% of the dermal infiltrate, or a specimen that demonstrates large cell nodules [5,6]. LCT can take place in up to 20% of advanced MF cases, and generally correlates with a more aggressive disease course [7]. Along with CTCL, patients with SS may also undergo LCT, and the median survival time for large cell transformed-SS ranges from 19–36 months [8,9]. Unfortunately, our current understanding of the molecular mechanisms and clinicopathologic features that correlate with progression to LCT is limited [7].

Due to the low incidence of this disease, and the poor prognosis associated with its transformation, there is a stark paucity of data regarding risk factors for the progression of SS to LCT. For this reason, there is a distinct need to better understand the clinicopathologic risk factors that might signal SS transformation to LCT prior to diagnosis. This knowledge is an essential first step to developing targeted management and therapies for these patients, and preventing the complications associated with more aggressive disease courses. Here, we seek to retrospectively characterize the clinicopathologic risk factors that may predispose patients with SS to LCT. We present the following article/case in accordance with the STROBE reporting checklist.

Materials and methods

This study was conducted with the approval of the Institutional Review Board at the University of Michigan. The need for informed consent was waived in this study and all data were anonymized prior to access. We retrospectively evaluated all SS patient records available at the Michigan Medicine from 2010–2021. The patients were queried from the University of Michigan electronic health record database and identified using free text parameters that included any of the words ‘progression’, or ‘large cell’ associated with a diagnosis of ‘Sezary Syndrome’, or ‘cutaneous T-cell lymphoma’ (‘CTCL’). Only patients with at least one histopathologically confirmed biopsy specimen for each diagnosis, reviewed by a dermatopathologist/hematopathologist with cutaneous lymphoma expertise to confirm the findings (A.H.), were included in the study. Cases which met the above criteria were then clinically confirmed as SS by N.K.J., R.B., and T.T. via detailed electronic health record review, which included immunohistochemical and molecular data alongside the consensus of cutaneous lymphoma experts in dermatology and hematology (T.T. and R.A.W.).

All patients were classified according to the International Society for Cutaneous Lymphomas (ISCL) and European Organization of Research and Treatment of Cancer (EORTC) revised criteria of 2007 [6]. Disease staging was determined by documented physical examination, peripheral blood smear, flow cytometry, complete blood count and chemistries, and imaging that included computed tomography (CT) and/or positron emission tomography (PET) scan of the chest, abdomen, and pelvis.

Results

Clinical features

Medical records from 794 entries in 337 unique patients were initially retrieved using the aforementioned search strategy. Of these, 85 patients were found to have either SS or LCT diagnosis. The authors manually reviewed these 85 patient records and excluded any patients that did not have a retrievable confirmed histopathologic or hematological diagnosis of SS. After applying exclusion criteria, 28 patients were included in the final analysis (Fig 1). The median age at diagnosis of SS was 72 (range 46–99). The diagnosis of SS preceded the diagnosis of LCT in all cases in which transformation occurred. At the time of data analysis, 18 of the 28 (64%) patients in the study had died, and 10 (36%) were alive. Additional malignant diagnoses were present in 2 patients. One patient (NLCT group, deceased secondary to lung adenocarcinoma) had a history of metastatic lung adenocarcinoma, concurrent with their SS diagnosis, and one patient had a history of metastatic melanoma concurrent with their SS diagnosis and preceding chronic lymphocytic leukemia (POT-1 mutation, LCT group, deceased secondary to metastatic melanoma). Concise demographic tables and summary statistics can be visualized in Table 1A.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Flow diagram detailing inclusion and exclusion of patients at each stage of the collection.

https://doi.org/10.1371/journal.pone.0277655.g001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. A. Concise demographic details for each patient included in the study.

“NA” in the “Days from SS Dx to death” column indicates that the patient was still alive at last follow up. B. Concise flow cytometric and clinicopathologic characteristics for each patient included in the study.

https://doi.org/10.1371/journal.pone.0277655.t001

Survival and time-to-next-treatment analysis

Survival analysis was conducted by taking into account the time to death from diagnosis of SS (OSF S1 Fig) to LCT, comparing patients that underwent LCT to those that did not. There was no significant difference in OS between groups when considering time to diagnosis from SS to LCT (HR 1.297, 95% CI 0.491–3.424, p = 0.600). Upon analysis of CD8+ T cell counts relative to the median expression of all samples (“High” or “Low”), there were no differences in OS from SS diagnosis to death based on relative absolute CD8+ T cell count at diagnosis of SS (OSF S2 Fig, HR 1.726, 95% CI 0.6275–4.748, p = 0.29) or at most recent blood draw (OSF S2B Fig, HR 0.9202, 95% CI 0.3359–2.521, p = 0.872). When considering TTNT, defined as mean time to next systemic therapy, there was a significant difference observed between the two groups, with the LCT group demonstrating a significantly shorter TTNT than the NLCT group (Fig 2, HR 5.868, 95% CI 0.0623–0.4661, p = 0.0013), as would be expected in more aggressive disease phenotypes.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Average time-to-next-treatment (TTNT) analysis measured in days of large cell transformation (LCT) versus non-large cell transformation (NLCT) groups.

The LCT group (blue line) demonstrated significantly decreased mean TTNT compared to the NLCT group (yellow line, HR 5.868, 95% CI 0.0623–0.4661, p = 0.0013).

https://doi.org/10.1371/journal.pone.0277655.g002

Discussion

There are few studies which have explored the prognostic value of selected clinicopathologic factors to predict which patients with SS will eventually undergo LCT, although notably, LCT has been confirmed as being independently correlated with poor prognosis in multiple previous studies of cutaneous T cell lymphomas, including SS [5,8,10–16]. Studies investigating the clinical outcomes of LCT have determined that advanced stage at the time of LCT diagnosis is associated with poor prognosis [5,8]. Additionally, LCT at initial diagnosis of SS has been shown to be correlated with decreased survival [5,8,11]. Our study, however, demonstrated no difference in overall survival between LCT and NLCT groups, possibly secondary to small sample size, patient comorbidities, variability in treatments offered over the study’s nearly 10-year span, or socioeconomic factors that could not be controlled for. Elevated LDH has also been recognized as another independent poor prognostic marker for survival in patients with advanced-stage SS and LCT [5,17]. Furthermore, Novelli et al. found in a retrospective analysis of 134 cases that a TBSA> 50% involvement by either MF or SS was significantly correlated with decreased survival [18].

Our study suggests that there are clinical and pathologic factors that may predict the development of LCT. Increased extent of skin lesions (increased TBSA involvement) has previously been surmised to predict reduced disease-specific survival (DSS) and OS in patients with transformed MF, but these have not been studied specifically in SS [19]. Our study demonstrates that TBSA is predictive of LCT in SS. Because TBSA is seen as a marker of disease progression, TBSA may be a promising indicator of aggressive disease, rather than only a modifiable disease factor that is correlated with LCT. The same may be said for the presence of ulceration on pathologic specimens, as it correlated with increased TBSA involvement in our dataset.

It is well-accepted that CD8+ T cells play an important role in mediating anti-tumor immunity by eliminating tumor cells through recognition of tumor-associated antigens. In cancers such as colorectal, lung, breast, melanoma and glioblastoma, the infiltration of T cells, and particularly CD8+ T cells, is correlated with better prognosis [20,21]. There are promising immunotherapies in development and trials for various cancers that aim to boost CD8+ T cell-mediated anti-tumor immunity, which include adaptive cell transfer of tumor-reactive T cells (either native or engineered to express tumor specific T cell receptors or chimeric antigen receptors), dendritic cell cancer vaccines (DCVax), and immune checkpoint blockades (including anti-PD-1, anti-PD-L1, and anti-CTLA-4 antibodies) [22,23]. Our analysis poses an interesting question concerning the role of these non-neoplastic cells in SS. Previously, it has been demonstrated that a higher CD8+ T cell infiltrate in the skin is correlated with better prognosis in patients across plaque-stage MF, tumor stage MF, and CD30- PCTCL, and is correlated with less aggressive disease [24]. These results are supported by Hoppe et al. who describe a relationship between higher CD8+ T cell proportions and better survival in MF [25]. To the authors’ knowledge, this relationship is yet to be studied in SS. Our study demonstrated the stark contrast in absolute CD8+ T cell number in the peripheral blood between patients in the LCT and NLCT groups, with significantly decreased absolute counts in patients who underwent transformation, both before and after diagnosis of LCT. This observation may indicate that decreased peripheral CD8+ T cell count may also allow for persistence or continued resistance against LCT in SS. This is consistent with the demonstrated role of CD8+ T cells as anti-tumoral effector cells. Given the retrospective nature of the study, we were unable to obtain sequencing data for the markers of T cell activation that may have offered a more robust analysis of the functional capabilities of the T cells present in the peripheral blood. Further in vitro and in vivo studies are needed to understand these results in a mechanistic context.

Our study took advantage of the electronic health records to identify patients to be included in the analysis, however, this is a retrospective analysis in a disease with low prevalence, which may explain some of the unusually high hazard ratios seen in our results, and additionally may affect the interpretation of results. Larger studies may reveal patterns that we were not able to appreciate with a small number of patients. Given the low incidence of LCT, addressing this limitation may be difficult to overcome. Thus, future studies utilizing in vitro and in vivo laboratory experimentation may be required confirm the validity of our hypothesized mechanisms.

In conclusion, our 28-patient study identified the risk factors for LCT in patients diagnosed with SS. Here, we identify clinical and pathologic factors that may confer prognostic value for the incidence of LCT and assess their impact on OS. These include maximum TBSA %, peak LDH, presence of ulceration, and decreased levels of CD8+ cells in the peripheral blood. Although some of these factors may be clinical surrogates for disease progression, the significance of absolute CD8+ cell counts in the peripheral blood provides new insight into LCT risk in SS and implicates potentially novel therapeutic approaches in management. Greater absolute CD8+ cell count in the peripheral blood may serve as a protective factor against the development of LCT in SS. Furthermore, lesser numbers of CD8+ T cells may allow for persistence of the disease. Clinicians may use this insight in stratifying patient risk and deciding whether to escalate therapy, in conjunction with the other variables found to be significant in the study, such as maximum TBSA involvement, LDH level, or ulceration on biopsy specimen. Therapy targeted at increasing the number of CD8+ T cells in the peripheral circulation may serve to protect patients from developing transformation of their SS. These results indicate the need for further study into the mechanisms of tumor progression based on immune cell populations in patients with SS and need for a larger multi-center trial to create robust prognostic indices.

Supporting information