Intratumoral expression of FoxP3-positive regulatory T-cells in T-cell lymphoma: no correlation with survival ARTICLE Intratumoral expression of FoxP3-positive regulatory T-cells in T-cell lymphoma: no correlation with survival Josefine Lundberga,c, David Berglunda, Daniel Molinb and Amelie Kinchc aDepartment of Immunology, Genetics and Pathology, Section of Clinical Immunology, Uppsala University, Uppsala, Sweden; bDepartment of Immunology, Genetics and Pathology, Section of Experimental and Clinical Oncology, Uppsala University, Uppsala, Sweden; cDepartment of Medical Sciences, Section of Infectious Diseases, Uppsala University, Uppsala, Sweden ABSTRACT Background. In cancer, regulatory T-cells (Tregs) were previously believed to inhibit tumor immunity, leading to reduced survival. However, in hematologic malignancies, including T-cell lymphoma (TCL), a correlation between increased numbers of tumor-infiltrating Tregs and a favorable prognosis has been reported. We aimed to investigate the expression of the Treg biomarker forkhead box protein 3 (FoxP3) in TCL in immunocompetent individuals and explore a possible correlation to overall survival. Methods. In total, 35 diagnostic biopsies of TCL were stained using a FoxP3-specific monoclonal anti- body (clone 236A/E7). Visual scoring was performed by counting positive cells in 15 high-power fields. Clinical data were collected retrospectively from medical records. Results. All the TCLs contained FoxP3þ cells, median 342 FoxP3þ cells/mm2 (range 1–3047). The degree of intratumoral expression of FoxP3 varied between the different subtypes of TCL, with the highest frequency found in angioimmunoblastic TCL. The frequency of intratumoral FoxP3þ cells had no impact on overall survival; neither when using a cutoff value of 200 FoxP3þ cells/mm2 (P ¼ 0.84) nor with FoxP3 as a continuous variable (P ¼ 0.63). Conclusions. Intratumoral Tregs are frequently found in TCL in immunocompetent individuals. In this heterogeneous group of TCL, there was no correlation between the density of intratumoral FoxP3þ cells and overall survival. ARTICLE HISTORY Received 25 October 2018 Revised 27 November 2018 Accepted 29 November 2018 KEYWORDS FoxP3; outcome; T-cell lymphoma; Tregs; tumor microenvironment Introduction Regulatory T-cells (Tregs) are a specialized subpopulation of T-cells with the capacity to suppress anti-self immune responses, thus maintaining peripheral tolerance and pre- venting autoimmune diseases (1). Tregs are also believed to play a vital part in inducing tumor-specific immune tolerance (2). Natural Tregs are produced in the thymus (3) and express CD4, CD25, and the transcription factor forkhead box protein 3 (FoxP3) (4). Peripheral CD4þ FoxP3� cells may also upregu- late FoxP3 in the presence of cytokines such as TGF-b and IL-2; this population of extrathymically generated FoxP3þ T- cells are known as peripherally derived Tregs (pTregs) (5). The transcription factor Helios (6) and the cell surface mol- ecule neuropilin-1 (7) have been suggested as markers to distinguish nTregs from pTregs, but the usefulness of these markers in humans has been subject to debate (8,9), and consensus has not yet been reached. Despite the fact that CD4þ CD25� T-cells have been shown to express FoxP3 upon activation, and that the levels of expression do not necessarily correlate to suppressive function in these cells, FoxP3 is still considered the most reliable molecular marker of Tregs (4, 10). Regarding the role of Tregs in cancer, several studies have shown a correlation between increased numbers of intratu- moral Tregs and decreased survival (e.g. ovarian carcinoma (11), prostate cancer (12), and non-small cell lung cancer (13)). In contrast, a higher frequency of intratumoral Tregs has been associated with improved survival in colorectal car- cinoma (14) as well as in several types of lymphoma—for example, follicular lymphoma, Hodgkin lymphoma, and ger- minal center-like diffuse large B-cell lymphoma (15). In previ- ous studies of Tregs in T-cell lymphoma (TCL), a positive impact on survival with increased numbers of intratumoral Tregs has been reported in mycosis fungoides (MF), unspeci- fied cutaneous TCL (CTCL) (16), and extranodal NK/T-cell lymphoma (ENKTL) (17). One study compared lymph node biopsies from patients with angioimmunoblastic TCL (AITL) or follicular lymphoma with reactive lymph nodes and found significantly lower numbers of Tregs in the AITL biopsies, which, according to the authors, might at least in part explain the autoimmune symptoms typical for AITL and thus the poor prognosis (18). Another study on adult T-cell leuke- mia/lymphoma (ATLL) focused on the expression of FoxP3 in the malignant cell population and reported that patients with FoxP3-negative tumors showed a tendency toward CONTACT Amelie Kinch amelie.kinch@medsci.uu.se Department of Medical Sciences, Section of Infectious Diseases, Uppsala University, S-751 85 Uppsala, Sweden � 2019 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. UPSALA JOURNAL OF MEDICAL SCIENCES 2019, VOL. 124, NO. 2, 105–110 https://doi.org/10.1080/03009734.2018.1555195 http://crossmark.crossref.org/dialog/?doi=10.1080/03009734.2018.1555195&domain=pdf http://orcid.org/0000-0002-9646-7283 http://orcid.org/0000-0003-3420-8590 http://creativecommons.org/licenses/by/4.0/ https://doi.org./10.1080/03009734.2018.1555195 http://www.tandfonline.com better survival in comparison with those with strong or inter- mediate intratumoral expression of FoxP3 (19). In a previous study we found that none of the T-cell post- transplant lymphoproliferative disorders were FoxP3þ (20). We speculated that this finding in part might be due to the immunosuppressive treatment that the patients typically receive after solid organ transplantation, and we questioned if TCL from patients without any previous immunosuppres- sion would differ in their expression of FoxP3, which led us to initiate this present study. We hypothesized that increased numbers of intratumoral Tregs may have an impact on sur- vival in different types of TCL, and in this study we have cor- related the number of FoxP3þ cells in biopsies of different subtypes of TCL to outcome. Materials and methods We included 35 TCL patients diagnosed 1999–2002 with suf- ficient tissue available for further analysis from a Swedish–Danish case-control study (SCALE; Scandinavian Lymphoma Etiology) (21). The 35 tissue samples were re- evaluated in the SCALE study by experienced hematopathol- ogists according to the World Health Organization (WHO) classification of tumors in hematopoietic and lymphatic tis- sue. After re-examination of the pathological reports in com- bination with the clinical data, we reclassified one of the biopsies from primary cutaneous CD30þ TCL to TCL not otherwise specified. Another biopsy was reclassified as highly malignant TCL not otherwise specified, after consulting an experienced hematopathologist, since the original diagnosis of ‘peripheral large cell TCL’ was not consistent with any of the diagnostic entities of the WHO classification. Clinical data for each patient were collected retrospectively from the dif- ferent Swedish hospitals where the patients had been pri- marily treated for their TCL. Immunohistochemistry for FoxP3 From formalin-fixed, paraffin-embedded diagnostic biopsies of TCL, 4 lm thick tissue sections were sliced. PT-link was used for performing deparaffinization and antigen retrieval. Immunohistochemical (IHC) staining was performed with Dako autostainer plus with clone 236A/E7 (eBioscience) used as pri- mary antibody to detect FoxP3, diluted 1:100 in antibody-solu- tion (Dako). In order to visualize the primary antibody, EnVision DAB-kit (Dako) was used. The samples were scanned and examined using Aperio ImageScope v12.1.0.5029 (Leica Biosystems). In the majority of the samples the positive cells were distributed in a heterogeneous manner, therefore the stained cells were counted manually in 15 arbitrarily chosen fields in each biopsy, at 40� magnification. The area of each field was measured. The number of positive cells per mm2 in the total analyzed area was calculated. Statistics Comparison of frequencies of FoxP3þ cells in different sub- types of TCL was made using the Mann–Whitney U-test. Overall survival (OS) was defined as the time from initial diagnosis of TCL until death or last date of follow-up, which was 19 February 2016. Survival curves were generated using the Kaplan–Meier method. Difference in survival was calcu- lated using Cox proportional hazards regression model with FoxP3 as a continuous variable or log-rank test with a cutoff value of 200 FoxP3þ cells/mm2. Statistical significance was defined as P < 0.05. All statistical analyses were performed using Statistica software (Version 13, Stat Soft Inc., Tulsa, OK, USA). Table 1. Clinical and biological characteristics of the patient group. Characteristics Number (%) (Available data) Sex (35/35) Male 21 (60%) Female 14 (40%) Age at TCL diagnosis (35/35) 18–20 2 (6%) 41–50 8 (23%) 51–60 13 (37%) 61–70 11 (31%) >70 1 (3%) TCL localization (35/35) Nodala 15 (43%) Extranodal 20 (57%) Bone marrow 6 (17%) Skin 7 (20%) Small intestine 5 (14%) Otherb 6 (17%) Stage (Ann Arbor) (31/35) I 9 (29%) II 4 (13%) III 11 (35%) IV 7 (23%) Performance status (WHO/ECOG) (35/35) 0–1 33 (94%) 2–3 2 (6%) IPI (26/35) Low risk 11 (42%) Low/high intermediate risk 14 (54%) High risk 1 (4%) Serum LDH concentration (28/35) Elevated 15 (54%) Within normal range 13 (46%) Recurrence (35/35) Yes 19 (54%) No 16 (46%) Subtypes of TCL (35/35) Anaplastic large cell lymphoma 12 (34%) Peripheral TCL, not otherwise specified 6 (17%) Unspecified TCL 5 (14%) Angioimmunoblastic TCL 3 (9%) Enteropathy-associated TCL 3 (9%) Cutaneous TCL 2 (6%) Mycosis fungoides 2 (6%) T-cell prolymphocytic leukemia 2 (6%) Cytotoxic chemotherapy treatment (first-line) (35/35) CHOP/CHOEP 22 (63%) VACOP-B 4 (11%) ABVD 1 (3%) Other 1 (3%) None 7 (20%) aNodal localization: malignancy in lymph nodes only. bExtranodal localization, other: pleura, lung, meninges, bone marrow, testis. ABVD ¼ adriamycin (doxorubicin), bleomycin, vinblastine, dacarbazine; CHOEP ¼ addition of etoposide to the standard CHOP-regimen; CHOP ¼ cyclophosphamide, hydroxydaunorubicin, oncovin (vincristine), pred- nisone or prednisolone; ECOG ¼ Eastern Cooperative Oncology Group; IPI ¼ International Prognostic Index; LDH ¼ lactate dehydrogenase; TCL ¼ T-cell lymphoma; VACOP-B ¼ VP-16 (etoposide), adriamycin (doxorubicin), cyclophos- phamide, oncovin (vincristine), prednisone, bleomycin; WHO ¼ World Health Organization. 106 J. LUNDBERG ET AL. Ethics approval The study was approved by the Regional Ethical Review Board in Stockholm, Sweden, and was conducted in accord- ance with the Declaration of Helsinki. Results Clinical characteristics and subtypes of TCL The median age at the time of diagnosis was 57 years (range 18–73) (Table 1). There were 21 men and 14 women. The majority of the patients (57%) had extranodal localization of TCL, most commonly the skin (20%). The most common sub- type of TCL was anaplastic large cell lymphoma (ALCL; n ¼ 12, 34%). Other subtypes included peripheral TCL, not otherwise specified (PTCL-NOS; n ¼ 6, 17%), unspecified TCL (n ¼ 5, 14%), AITL (n ¼ 3, 9%), enteropathy-associated TCL (EATL; n ¼ 3, 9%), CTCL (n ¼ 2, 6%), MF (n ¼ 2, 6%), and T-cell prolymphocytic leukemia (T-PLL, n ¼ 2, 6%). Expression of FoxP3 in TCL All 35 tissue samples expressed FoxP3 to some degree. Overall, the median expression of FoxP3 was 342 positive cells/mm2 (mean 610 positive cells/mm2), and the range was large (1–3047 positive cells/mm2). The degree of intratumoral expression of FoxP3 tended to vary between the different subtypes of TCL (Figure 1). The highest frequency of FoxP3þ cells was detected in AITL (n ¼ 3, median 1057 and range 849–1642 positive cells/mm2) compared with all other sub- types, although this difference did not reach statistical sig- nificance (P ¼ 0.053). T-PLL (n ¼ 2) tended to have the lowest frequency of FoxP3þ cells/mm2 (P ¼ 0.14). Survival At the end of follow-up 21 patients had died, 18 of whom due to lymphoma. The median follow-up time for the 14 sur- viving patients was 11.8 years (range 5.7–14.4 years). One- and five-year overall survival were 89% and 46%, and one- and five-year relapse-free survival were 72% and 44% for all patients. There was no difference in OS between cases with high or low expression of FoxP3, neither when analyzed as a dichotomous variable (> or < 200 FoxP3þ cells per mm2, P ¼ 0.84, Figure 2), nor as a continuous variable (P ¼ 0.63). This result did not change when the cases of MF, which Figure 1. Degree of intratumoral expression of FoxP3þ cells/mm2 in the eight different subtypes of T-cell lymphoma. AITL: angioimmunoblastic T-cell lymphoma; ALCL: anaplastic large cell lymphoma; CTCL: unspecified cutaneous T-cell lymphoma; EATL: enteropathy-associated T-cell lymphoma; MF: mycosis fungoides; PTCL-NOS: peripheral T-cell lymphoma, not otherwise specified; T- PLL: T-cell prolymphocytic leukemia; TCL-NOS: unspecified T-cell lymphoma. Figure 2. The frequency of intratumoral FoxP3þ cells (more or less than 200 FoxP3þ cells/mm2) in T-cell lymphomas had no impact on overall survival. Figure 3. Degree of intratumoral expression of FoxP3 in the 35 biopsies, with cutoff level set arbitrarily at 200 FoxP3þ cells/mm2. UPSALA JOURNAL OF MEDICAL SCIENCES 107 have a better prognosis than the rest of the TCLs, were omit- ted from the survival analysis (log-rank test, P ¼ 0.55; Cox proportional hazards, P ¼ 0.76). Considering that no differ- ence in OS was detected when analyzing FoxP3 as a continu- ous variable, it is of minor consequence that the cutoff was set arbitrarily based on visual assessment of a scatter chart (Figure 3) when FoxP3 was used as a dichotomous variable. Discussion In this study we found that all TCLs expressed FoxP3 to some degree but that the frequency varied considerably between different subtypes. There was no association between the number of intratumoral Tregs and OS. Many studies have shown a correlation between a high density of Tregs in tumor tissue and a worse prognosis in different malignancies (11–13), but at the same time sev- eral studies have shown the opposite (14–17). The nega- tive impact of Tregs on outcome in cancers may be explained as one of several immune evasion strategies by tumors, where the tumors escape the immune system by suppressing antitumoral effector cells (11,22–24). In the malignancies where Tregs instead are associated with a favorable outcome, a possible explanation could be that Tregs, due to their anti-inflammatory capacity, reduce tis- sue damage and restrict inflammation-related carcinogen- esis (1,25,26). In TCL, the function of Tregs seems particularly complex, partly because the malignant cells in these tumors are lymphoid and thus potentially could be suppressed by Tregs, and partly because some studies have suggested that the malignant cells could in fact be Treg-derived (19,27–29). Berger et al. (29) suggested that CTCLs are tumors of malignant Tregs, based on findings that CTCL-cells could adopt Treg phenotype and function after stimulation by apoptotic material in vitro. Gjerdrum et al. (16) showed that only a small minority of CTCL cases showed expression of FoxP3 in the malignant cells while all contained FoxP3þ cells in the tumor microenvironment to some degree. Of note, higher numbers of intratumoral Tregs were associated with improved survival in these TCLs. There were only two MF and two CTCL unspecified in our case series, and we have not used the same method for evaluation; this complicates comparisons between the two studies, but the frequencies of Tregs seem similar. Regarding ALCL, PTCL-NOS, and AITL, we found higher frequencies of FoxP3þ cells than previous studies (15,18,27). The largest subgroup in our case series was ALCL (n ¼ 12), where we found a mean of 621 FoxP3þ cells/mm2 as compared with 15 FoxP3þ cells/mm2 in the four cases of ALCL in the study by Tzankov et al. (15). Bonzheim et al. (27) detected FoxP3þ cells in four of six ALKþ and in one of 14 ALK- ALCL, but the specific number of positive cells was not reported. This is in sharp contrast to our case series where the majority of ALCL were ALK- (seven of nine with known ALK status) and despite this were FoxP3þ. However, the limited number of cases pre- vents a meaningful comparison between ALKþ and ALK- ALCL. In PTCL-NOS we found a mean of 965 FoxP3þ cells/ mm2 compared with 34 FoxP3þ cells/mm2 in the 27 cases in the study by Tzankov et al. (15), whereas Bonzheim et al. (27) only detected expression of FoxP3 in one of 14 cases of PTCL-NOS, and that was in the malignant popula- tion. Regarding AITL, we detected a mean of 1183 FoxP3þ cells/mm2 in three cases as compared with a mean of 61 FoxP3þ cells/mm2 in 23 cases by Tzankov et al. (15) and mean of 90 FoxP3þ cells/high-power field (HPF; 400�) in 30 cases by Bruneau et al. (18). Bonzheim et al. (27) found no FoxP3þ tumor cells and only a few FoxP3þ cells in the reactive infiltrate in 23 cases of AITL. The differences in expression of FoxP3 can in part be explained by the use of different antibodies for the detection of FoxP3. Tzankov et al. (15) used the mouse monoclonal antibody 22510 (Abcam), Bonzheim et al. (27) used a rabbit polyclonal antibody from Abcam (clone number not speci- fied), whereas Bruneau et al. (18) and Gjerdrum et al. (16) used the same clone as we did, namely the mouse monoclo- nal 236A/E7 (Abcam). Clone 236A/E7 has been reported to be predictive of improved outcome in follicular lymphoma (30), Hodgkin lymphoma (31), and CTCL (16). In our case ser- ies there was no correlation between expression of FoxP3þ and outcome. An association between high numbers of intra- tumoral Tregs and superior survival has been reported for some types of TCLs, e.g. CTCL (16) and ENKTL (17), but not for the largest subgroups in our case series, ALCL and PTCL- NOS, to the best of our knowledge. In a previous study with the use of the same antibody as in the present study, we found that 13 cases of T-cell post- transplant lymphoproliferative disorder occurring after solid organ transplantation were completely negative for expres- sion of FoxP3 and in one case FoxP3þ cells were detected at a low density (10 positive cells/mm2) (20). This finding is in contrast with the observation in the present study where all TCL expressed FoxP3 to some degree. A plausible explan- ation for this striking difference is the influence of the immunosuppressive drugs in solid organ trans- plant recipients. The limitations of this study include the small subgroups of TCL subtypes, which hinders comparisons between groups. Further, the FoxP3þ cells in the biopsies have not been defined as belonging to either the neoplastic or react- ive cell population within the tumors. The frequency of intra- tumoral Tregs is based on the expression of FoxP3, which has limitations but still is considered the most reliable bio- marker for Tregs (10). Strengths include that the biopsies were collected prospectively on a national level and re-eval- uated by experienced hematopathologists. Furthermore, 15 arbitrarily chosen HPFs were manually counted which is more than the standard five HPFs. In conclusion, in this study we found distinctly higher levels of expression of FoxP3 in AITL, PTCL-NOS, and ALCL when compared with previous studies of these lymphomas in immunocompetent individuals. The finding that all of the TCLs showed expression of FoxP3 to some degree is interesting when compared with the results of our previ- ous study of Tregs in posttransplant lymphoproliferative 108 J. LUNDBERG ET AL. disorder, where all but one of the T-cell posttransplant lymphoproliferative disorders were entirely lacking expres- sion of FoxP3. Thus, the role of Tregs seems to differ between TCLs in immunocompetent and immunosup- pressed hosts. In this heterogeneous group of TCL, there was no correlation between the density of intratumoral FoxP3þ cells in biopsies of TCL and OS. Declaration of interest D.M. has received honoraria from Roche, Merck, Bristol-Meyers Squibb, and Takeda. The other authors report no conflicts of interest. Funding The work was supported by the Lion's Cancer Research Foundation. Notes on contributors Josefine Lundberg, Master of Science in Medicine, is at the beginning of her clinical internship at Nyk€opings Hospital, Sweden. David Berglund, MD, PhD, trained in immunology and transplant/recon- structive surgery and is an associate professor at the Department of Immunology, Genetics and Pathology, Section of Immunology, Uppsala University, Uppsala, Sweden. Daniel Molin, MD, PhD, is a senior consultant at the Department of Oncology, Uppsala University Hospital and associate professor at the Department of Immunology, Genetics and Pathology, Section of Experimental and Clinical Oncology, Uppsala University, Uppsala, Sweden. Amelie Kinch, MD, PhD, is a senior consultant at the Department of Infectious Diseases, Uppsala University Hospital and researcher at the Department of Medical Sciences, Section of Infectious Diseases, Uppsala University, Uppsala, Sweden. ORCID Daniel Molin http://orcid.org/0000-0002-9646-7283 Amelie Kinch http://orcid.org/0000-0003-3420-8590 References 1. Mougiakakos D, Choudhury A, Lladser A, Kiessling R, Johansson CC, Regulatory T cells in cancer. Advances in Cancer Research [Internet]. Elsevier; 2010 [cited 2018 Nov 23]. p. 57–117. Available from: http://linkinghub.elsevier.com/retrieve/pii/ S0065230X1007003X. 2. Liu C, Workman CJ, Vignali DAA. Targeting regulatory T cells in tumors. Febs J. 2016;283:2731–48. 3. Itoh M, Takahashi T, Sakaguchi N, Kuniyasu Y, Shimizu J, Otsuka F, et al. Thymus and autoimmunity: production of CD25 þ CD4þ nat- urally anergic and suppressive T cells as a key function of the thy- mus in maintaining immunologic self-tolerance. J Immunol. 1999; 162:5317 4. Fontenot JD, Gavin MA, Rudensky AY. Foxp3 programs the devel- opment and function of CD4 þ CD25þ regulatory T cells. Nat Immunol. 2003;4:330–6. 5. Zheng SG, Wang J, Wang P, Gray JD, Horwitz DA. IL-2 Is Essential for TGF-beta to Convert Naive CD4 þ CD25- Cells to CD25 þ Foxp3þ Regulatory T Cells and for Expansion of These Cells. J Immunol. 2007;178:2018–27. 6. Thornton AM, Korty PE, Tran DQ, Wohlfert EA, Murray PE, Belkaid Y, et al. Expression of Helios, an Ikaros transcription factor family member, differentiates thymic-derived from peripherally induced Foxp3þ T regulatory cells. J Immunol. 2010;184:3433–41. 7. Yadav M, Louvet C, Davini D, Gardner JM, Martinez-Llordella M, Bailey-Bucktrout S, et al. Neuropilin-1 distinguishes natural and inducible regulatory T cells among regulatory T cell subsets in vivo. J Exp Med. 2012;209:1713–22. 8. Szurek E, Cebula A, Wojciech L, Pietrzak M, Rempala G, Kisielow P, et al. Differences in expression level of helios and neuropilin-1 do not distinguish thymus-derived from extrathymically-induced CD4 þ Foxp3þ regulatory T cells. PLoS One. 2015;10:e0141161. 9. Elkord E. Helios should not be cited as a marker of human thy- mus-derived Tregs. Commentary: Helios þ and Helios � cells coex- ist within the natural FOXP3þ T regulatory cell subset in humans. Front Immunol. 2016;7:276. 10. Sakaguchi S, Miyara M, Costantino CM, Hafler DA. FOXP3þ regula- tory T cells in the human immune system. Nat Rev Immunol. 2010;10:490–500. 11. Curiel TJ, Coukos G, Zou L, Alvarez X, Cheng P, Mottram P, et al. Specific recruitment of regulatory T cells in ovarian carcinoma fosters immune privilege and predicts reduced survival. Nat Med. 2004;10:942. 12. Flammiger A, Weisbach L, Huland H, Tennstedt P, Simon R, Minner S, et al. High tissue density of FOXP3þ T cells is associated with clinical outcome in prostate cancer. Eur J Cancer. 2013;49:1273–9. 13. Tao H, Mimura Y, Aoe K, Kobayashi S, Yamamoto H, Matsuda E, et al. Prognostic potential of FOXP3 expression in non-small cell lung cancer cells combined with tumor-infiltrating regulatory T cells. Lung Cancer. 2012;75:95–101. 14. Salama P, Phillips M, Grieu F, Morris M, Zeps N, Joseph D, et al. Tumor-infiltrating FOXP3þ T regulatory cells show strong prog- nostic significance in colorectal cancer. Jco. 2009;27:186–92. 15. Tzankov A, Meier C, Hirschmann P, Went P, Pileri SA, Dirnhofer S. Correlation of high numbers of intratumoral FOXP3þ regulatory T cells with improved survival in germinal center-like diffuse large B-cell lymphoma, follicular lymphoma and classical Hodgkin’s lymphoma. Haematologica 2008;93:193–200. 16. Gjerdrum LM, Woetmann A, Odum N, Burton CM, Rossen K, Skovgaard GL, et al. FOXP3þ regulatory T cells in cutaneous T-cell lymphomas: association with disease stage and survival. Leukemia 2007;21:2512–8. 17. Peng R-J, Huang Z-F, Zhang Y-L, Yuan Z-Y, Xia Y, Jiang W-Q, et al. Circulating and tumor-infiltrating Foxp3(þ) regulatory T cell subset in Chinese patients with extranodal NK/T cell lymphoma. Int J Biol Sci. 2011; 7:1027–36. 18. Bruneau J, Canioni D, Renand A, Marafioti T, Paterson JC, Martin- Garcia N, et al. Regulatory T-cell depletion in angioimmunoblastic T-Cell lymphoma. Am J Pathol. 2010;177:570–4. 19. Roncador G, Garcia JF, Garcia JF, Maestre L, Lucas E, Menarguez J, et al. FOXP3, a selective marker for a subset of adult T-cell leukae- mia/lymphoma. Leukemia 2005;19:2247–53. 20. Berglund D, Kinch A, Edman E, Backlin C, Enblad G, Larsson E, et al. Expression of intratumoral forkhead box protein 3 in post- transplant lymphoproliferative disorders: clinical features and sur- vival outcomes. Transplantation 2015;99:1036–42. 21. Smedby KE, Hjalgrim H, Melbye M, Torrång A, Rostgaard K, Munksgaard L, et al. Ultraviolet radiation exposure and risk of malignant lymphomas. J Natl Cancer Inst. 2005;97:199–209. 22. Zou W. Regulatory T cells, tumour immunity and immunotherapy. Nat Rev Immunol. 2006;6:295–307. 23. Perrone G, Ruffini PA, Catalano V, Spino C, Santini D, Muretto P, et al. Intratumoural FOXP3-positive regulatory T cells are associ- ated with adverse prognosis in radically resected gastric cancer. Eur J Cancer. 2008;44:1875–82. 24. Bates GJ, Fox SB, Han C, Leek RD, Garcia JF, Harris AL, et al. Quantification of regulatory T cells enables the identification of high-risk breast cancer patients and those at risk of late relapse. J Clin Oncol. 2006;24:5373–80. 25. Oleinika K, Nibbs RJ, Graham GJ, Fraser AR. Suppression, subversion and escape: the role of regulatory T cells in cancer progression: regu- latory T cells in cancer progression. Clin Exp Immunol. 2013;171:36–45. UPSALA JOURNAL OF MEDICAL SCIENCES 109 http://linkinghub.elsevier.com/retrieve/pii/S0065230X1007003X http://linkinghub.elsevier.com/retrieve/pii/S0065230X1007003X 26. Shang B, Liu Y, Jiang S, Liu Y. Prognostic value of tumor-infiltrat- ing FoxP3þ regulatory T cells in cancers: a systematic review and meta-analysis. Sci Rep. 2015;5:15179. 27. Bonzheim I, Geissinger E, Tinguely M, Roth S, Grieb T, Reimer P, et al. Evaluation of FoxP3 expression in peripheral T-cell lymph- oma. Am J Clin Pathol. 2008;130:613–9. 28. Wang J, Ke X-Y. The four types of Tregs in malignant lymphomas. J Hematol Oncol. 2011;4:50. 29. Berger CL, Tigelaar R, Cohen J, Mariwalla K, Trinh J, Wang N, et al. Cutaneous T-cell lymphoma: malignant proliferation of T-regula- tory cells. Blood 2005;105:1640–7. 30. Carreras J, Lopez-Guillermo A, Fox BC, Colomo L, Martinez A, Roncador G, et al. High numbers of tumor-infiltrating FOXP3- positive regulatory T cells are associated with improved overall survival in follicular lymphoma. Blood 2006;108:2957–64. 31. Greaves P, Clear A, Coutinho R, Wilson A, Matthews J, Owen A, et al. Expression of FOXP3, CD68, and CD20 at diagnosis in the microenvironment of classical Hodgkin lymphoma is predictive of outcome. J Clin Oncol. 2013;31:256–62. 110 J. LUNDBERG ET AL. Abstract Introduction Materials and methods Immunohistochemistry for FoxP3 Statistics Ethics approval Results Clinical characteristics and subtypes of TCL Expression of FoxP3 in TCL Survival Discussion Declaration of interest Notes on contributors References