Evaluation of the diagnostic and prognostic value of PDL1 expression in Hodgkin and B-cell lymphomas

    Evaluation of the diagnostic and prognostic value of PDL1-expression in Hodgkin- and B-cell lymphomas Thomas Menter, Andrea Bodmer-Haecki, Stephan Dirnhofer, Alexandar Tzankov PII: DOI: Reference:

S0046-8177(16)30014-4 doi: 10.1016/j.humpath.2016.03.005 YHUPA 3855

To appear in:

Human Pathology

Received date: Revised date: Accepted date:

9 December 2015 23 February 2016 1 March 2016

Please cite this article as: Menter Thomas, Bodmer-Haecki Andrea, Dirnhofer Stephan, Tzankov Alexandar, Evaluation of the diagnostic and prognostic value of PDL1-expression in Hodgkin- and B-cell lymphomas, Human Pathology (2016), doi: 10.1016/j.humpath.2016.03.005

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

RI P

T

Evaluation of the diagnostic and prognostic value of PDL1expression in Hodgkin- and B-cell lymphomas

SC

Thomas Menter1, 2, Andrea Bodmer-Haecki1, Stephan Dirnhofer1, Alexandar Tzankov1 1

MA NU

Institute of Pathology, University Hospital Basel, Schönbeinstrasse 40, 4031 Basel, Switzerland 2

AC

CE

PT

ED

Department of Histopathology, Hammersmith Hospital Campus, Imperial College Healthcare NHS Trust, Du Cane Road, W12 0HS London, United Kingdom

Correspondence to:

Prof. Dr. med. Alexandar Tzankov Institute of Pathology Schönbeinstrasse 40 4031 Basel Switzerland Tel.: +41-61-3286880 Fax: +41-61-2653194 E-mail: [email protected]

ACCEPTED MANUSCRIPT Abstract Activation of the programmed death 1 (PD1)/PD1 ligand (PDL1) pathway is important for

T

tumor cells to escape from immune control. The clinical efficacy of therapeutic modulation

RI P

of the PD1-PDL1 pathway has been recently shown in classical Hodgkin lymphoma (cHL), but little is known about the frequency, diagnostic and prognostic importance of PDL1

SC

expression in lymphomas.

MA NU

The available anti-PDL1 antibody clones E1L3N and SP142 were compared and a large cohort of HL (n=280) and B-cell lymphomas (n=619) was examined for PDL1 using E1L3N. The results were correlated with the expression of other phenotypic markers, interphase FISH data of the 9p24.1region (PDL1 locus) and with the clinical outcome.

ED

PDL1 was expressed on >5% of tumor cells in 70% of cHL, 54% of nodular lymphocytic

PT

predominant HL and 35% of primary mediastinal B-cell lymphomas; in the latter, PDL1 expression correlated with PDL1 gains (ρ=0.573). PDL1 was expressed in 31% of primary

CE

diffuse large B-cell lymphomas (DLBCL), while most other entities did not express PDL1. In

AC

cHL, expression of PDL1 correlated with increased numbers of granzyme+ T-cells (ρ=0.251) and CD68+ macrophages (ρ=0.221) but with decreased numbers of FoxP3+ T-cells (ρ=0.145). In activated B cell-like (ABC) DLBCL, PDL1 positively correlated with PD1+ T-cells, while an inverse correlation with FoxP3+ T-cells was seen in the germinal center B cell-like DLBCL. PDL1 expression can be diagnostically valuable in some gray zones around DLBCL and cHL; it identifies an “immune escape”-cluster of cHL and ABC-DLBCL with increased granzyme+ and PD1+ T-cells and macrophages and decreased regulatory T-cells.

Key words: PDL1, Hodgkin lymphoma, B-cell lymphoma, immunohistochemistry, diagnostic marker

ACCEPTED MANUSCRIPT INTRODUCTION The programmed death 1 (PD1)/PD-ligand (PDL) pathway is an important checkpoint for the

T

regulation of T-cell mediated immune responses [1]. It consists of the transmembrane

RI P

protein PD1/CD279 itself and its two ligands PDL1 (B7-H1, CD274) and PDL2 (B7-DC, CD273). These PDL activate PD1, which results in a reversible inhibition of T-cell activity and

SC

proliferation also known as T-cell exhaustion or anergy. This mechanism plays an important

MA NU

physiological role in preventing placenta infiltration by T-cells [2] and maintaining selftolerance [3], which has been verified in animal studies of pd1 knock-out mice developing several autoimmune diseases [4].

Unfortunately, malignancies also make use of the immunosuppressive effects of the PD1-

ED

PDL pathway [5], which is to a part reflected by high levels of PD1 positive T-cells infiltrating

PT

tumors. Besides, several tumors are known to express PDL1, being one of the mechanisms of building up a defense line against tumor infiltrating lymphocytes [6]. This applies to solid

CE

tumors like lung or breast cancer but also to hematolymphoid neoplasms like

AC

angioimmunoblastic T-cell lymphoma, follicular lymphoma (FL), diffuse large B-cell lymphoma (DLBCL), primary mediastinal B-cell lymphoma (PMBCL) and classical Hodgkin lymphoma (cHL), especially the nodular sclerosis subtype [7, 8]. Amplifications of the PDL1 gene locus on 9p24.1 are recurrent in the last two entities [9] further underscoring the importance of that pathway. PDL1-expression by immunohistochemistry has been demonstrated in some of the lymphomas listed above only in smaller cohorts or with antibodies shown to more poorly perform so far [10, 11]. Due to the fact that intrinsic cancer-specific T-cells might be thwarted by PD1-PDL1 interactions [12], targeting PDL1 has become a new approach in tumor therapy [13]. Recently, first very promising results of PD1-

ACCEPTED MANUSCRIPT PDL1 blockade have been reported in cHL, melanoma and non-small cell lung cancer [14– 16].

T

So far, the question whether expression of PD1 and PDL1 by immunohistochemistry might

RI P

also be of diagnostic or prognostic importance in lymphomas has not been answered on larger collectives. It was our aim to investigate the expression of PDL1 at large scale in in cHL

SC

and various B-cell lymphoma subtypes to draw conclusions both for diagnostic and potential

MA NU

clinico-pathological applications.

MATERIALS AND METHODS

Selection and tissue microarrays construction

ED

A total number of 899 cases encompassing different lymphoma entities were selected and

PT

examined on tissue microarrays (TMA) and whole slides (Table 1); 15 nodular lymphocytepredominant Hodgkin lymphomas (NLPHL), 7 Burkitt lymphomas (BL), 35 mantle cell

CE

lymphomas (MCL) and 11 T-cell and histiocyte rich B-cell lymphomas (THRBCL) were

AC

analyzed on whole mount slides. The study was approved by the ethics committee of Northwestern and Central Switzerland (EKNZ 2014-252).

Immunohistochemistry In order to assess the optimal staining for PDL1, two different antibodies [clone E1L3N (Cell Signaling, Danvers, MA, USA) and clone SP142 (Roche/Ventana, Rotkreuz, Switzerland)] were evaluated on non-small cell lung cancer tissue as well as reactive lymph node and tonsil tissue (figure 1). The staining pattern of both antibodies was as expected: membranous/submembranous with occasional dots, corresponding to the PD1-PDL1 interaction sites (talk given by Dr. Banks from Roche/Ventana at the 2015 SH Workshop on

ACCEPTED MANUSCRIPT Immunodeficiency and Dysregulation at Long Beach, CA, on October 30 th); in reactive lymph nodes the sinus-lining cells as well as macrophages stained positively, as to be assumed.

T

Finally the E1L3N antibody clone was decided to be used for further evaluations primary

RI P

because of the better signal-to-noise results; in addition this antibody has already been used by other groups showing reliable staining results [17]. Slides were processed on an immunostainer

(Benchmark,

Ventana/Roche,

SC

automated

Tucson,

AZ,

USA).

MA NU

Immunohistochemical staining procedures of all antibodies used in this study are listed in Table 2. This table also contains the staining protocols including the references of staining results obtained in other studies of our group that were used for correlation analyses in the present study.

ED

Only expression of PDL1 on tumor cells was considered in the present study. PDL1 positivity

PT

was defined as at least 5% (every 20th) positively staining tumor cells. One fifth of cases were scored by two observers (TM and AT) to assess reproducibility. To be considered evaluable,

CE

cases, especially TMA cases, had to exhibit at least 25% of tissue available for morphologic

AC

analysis and at least one positively staining tumor-infiltrating macrophage as positive internal control. In addition they had to contain unequivocal tumor cells, and, in cases of cHL, at least five Hodgkin and Reed-Sternberg cells (HRSC) or, in cases of NLPHL, five L&H cells, respectively. Tumor cells for scoring were morphologically easily identifiable in cases of BL, cHL, DLBCL, marginal zone lymphomas (MZL) with a high grade component, NLPHL and PMBCL, while in so called “low grade” B-cell lymphomas tumor-cell rich areas, as assessed on consecutive slides stained with tumor-cell typical markers like cyclin D1, CD20 or BCL6, were particularly considered. Data on PD1 expression as well as on 9p24.1gains (PDL1 locus) was gathered from previous studies on the same TMA from our group [7, 8, 18]. To verify/falsify PDL1 expression on tumor cells in cases of THRBCL, MZL and MCL, a PDL1 (AEC-

ACCEPTED MANUSCRIPT based red read-out)/BCL6 (DAB-based brown read-out) double staining (for THRBCL) and PDL1/PAX5 double staining (for MZL and MCL) was applied and performed analogously as

RI P

T

described elsewhere [19].

Statistical evaluation

SC

All statistical analyses were performed using the Statistical Package of Social Sciences (IBM

MA NU

SPSS version 22.0, Chicago, IL, USA) for Windows. The degree of agreement between observers (reproducibility) was evaluated by interclass correlation coefficients, using reliability Cronbach’s Alpha analysis; -values>0.75 implying an excellent agreement. The chi-square test was applied to identify quantitative differences between groups. The

ED

Spearman rank correlation was used to analyze relationships between markers. The

PT

diagnostic performance of PDL1 was assessed by receiver operating characteristic (ROC)curve, setting entities with higher expression of PDL1 as test variables and plotting sensitivity

CE

versus 1-specificity with special consideration of the respective area under the ROC

AC

(AUROC). Overall survival (OS) was measured from registration to death or last follow-up, event-free survival (EFS) from registration to relapse, death of any cause, or to last followup. The probabilities of survival were determined using the Kaplan–Meier method, and differences were compared using the log-rank test. All p-values were two-sided and considered statistically significant if <0.05. No adjustment for multiple testing was applied for secondary analyses because they were considered hypothesis generating and exploratory.

RESULTS Staining results

ACCEPTED MANUSCRIPT The Cronbach’s α-value was 0.896 for overall PDL1 staining evaluation; thus, the staining assessment was regarded reproducible. The PDL1 staining was strong to moderate,

T

membranous and submembranous, occasionally dotted (for explanation of the dots see

RI P

Materials and Methods/Immunohistochemistry). Figure 2 shows representative examples of PDL1 staining results in several lymphoma entities. In THRBCL and HL there were many

SC

reactive non-tumor cells expressing PDL1. In THRBCL, a double staining for BCL6 and PDL1

MA NU

with both clones used (E1L3N and SP142) was used to discriminate tumor cells from the surrounding non-tumor cells as was the case in MZL and MCL applying PAX5 and PDL1 double staining, while in cases of HL tumor cells were discriminated from non-tumor cells based on morphology since tumor cells had unequivocal morphologic characteristics

PT

ED

allowing discernment. All staining results are summarized in Table 1.

PDL1 expression in Hodgkin lymphomas

CE

PDL1 expression in HRSC was observed in the majority of cases, and in positive cases – on

AC

almost all tumor cells (Figure 2A). However, the proportion of positive cases varied from 65% in the nodular sclerosis subtype (87/134) and 67% in the lymphocyte depleted subtype (4/6) and unclassifiable cases (6/9) to 81% in the mixed cellularity subtype (60/74) and 90% in the lymphocyte rich subtype (9/10). Expression in NLPHL was lower (54%; 7/13; Figure 2B).

PDL1 expression in B-cell (non-Hodgkin) lymphomas PDL1 was expressed in 31% of DLBCL cases on a considerable amount of tumor cells (Figure 2C). A similar rate of PDL1-expression (35%) was seen in PMBCL (Figure 2D). PDL1 was detected in about 5% of FL and 10% of MZL with a high grade component (Figure 2E and F); yet in FL only a small proportion of tumor cells stained weakly for PDL1 (Figure 2G). In MZL

ACCEPTED MANUSCRIPT the expression of PDL1 on B-cells was verified by means of PAX5 and PDL1 double staining. Expression of PDL1 was very low in small lymphocytic B-cell lymphomas, low grade MZL and

T

primary testicular DLBCL cases. All cases of lymphoplasmacytic lymphoma and MCL were

RI P

negative for PDL1; in MCL this was falsified by means of PAX5 and PDL1 double staining on

SC

whole mount sections.

MA NU

Evaluation of whole mount slides

All cases of BL, MCL and THRBCL examined on whole mount slides did not express PDL1 on tumor cells. THRBCL showed a high number of tumor infiltrating T-cells and macrophages expressing PDL1. Double immunostaining for BCL6 and PDL1 confirmed the negativity of

ED

tumor cells in THRBCL (Figure 2H). As these results were in contrast to previous studies, both

PT

ant-PDL1 antibody clones (E1L3N and SP142) were used for the double stainings with both

CE

showing the same results.

AC

Correlation of PDL1 expression with other markers and survival in both Hodgkin and B-cell (non-Hodgkin) lymphomas Hodgkin lymphomas

There was a negative correlation observed between PDL1 positivity on HRSC and FoxP3 positive tumor-infiltrating lymphocyte (TIL) amounts (=-0.145, p=0.036). Granzyme positive TIL and CD68 positive tumor-infiltrating macrophages showed a positive correlation with PDL1 expression on HRSC (=0.251, p=0.0001; =0.221, p=0.005). No correlation was observed with FISH results regarding aberrations on 9p24.1, Epstein-Barr virus (EBV) status, rosetting of PD1 positive reactive T-cells as well as clinical details (gender, B-symptoms,

ACCEPTED MANUSCRIPT disease stage or overall survival). However, in EBV negative cHL cases, PDL1 expression showed a trend towards worse EFS (p=0.078; Figure 3).

RI P

T

B-cell (non-Hodgkin) lymphomas

In DLBCL cases, an inverse correlation was seen between germinal center expressed

SC

transcript 1 (GCET1) and PDL1 (=-0.186, p=0.005) thus pointing towards a more common expression of PDL1 in the non-germinal center-, activated B cell-like (ABC) subtype of DLBCL.

MA NU

A positive correlation between PDL1 expression on tumor cells of ABC-DLBCL and PD1 positive TIL amounts (=0.294, p=0.034) and a negative correlation between PDL1 expression on tumor cells of germinal center B cell-like (GCB) DLBCL and FoxP3 positive TIL

ED

amounts was observed (=-0.414, p=0.005).

In PMBCL, a strong correlation coefficient between PDL1 expression and gains of the PDL1

PT

locus was perceived (=0.573, p=0.087).

CE

No correlation between PDL1 expression and clinical outcomes was seen in all non-Hodgkin

AC

lymphomas in contrast to the findings in cHL.

Diagnostic value of PDL1 in discriminating different entities Accounting positivity for PDL1 in lesional cells can theoretically be helpful in the differential diagnosis of cHL versus THRBCL, PDL1 expression favoring cHL (AUROC 0.836, p=0.00016), and in the differential diagnosis of NLPHL versus THRBCL, PDL1 favoring NLPHL (AUROC 0.769, p=0.026). In B-cell lymphomas, PDL1 expression theoretically favors DLBCL compared to other entities (AUROC 0.687p=0.001) as well as PMBCL (AUROC 0.692, p=0.005).

DISCUSSION

ACCEPTED MANUSCRIPT In this study, we comprehensively analyzed the expression of PDL1 in both Hodgkin- and Bcell (non-Hodgkin) lymphomas. To this end, we selected an optimally working antibody

T

giving best signal-to-noise staining results and assessed the reproducibility of staining

RI P

evaluation, which showed excellent results. PDL1 expression was studied in a large cohort of TMA cases as well as on whole mount slides. Besides, a double staining for BCL6 and PDL1

SC

was established for THRBCL in order to facilitate the examination of the rare neoplastic cells,

MA NU

and a double staining for PAX5 and PDL1 was established for MZL and MCL to unequivocally show expression of PDL1 on B-cells in the former and lacking expression of PDL1 on B-cells in the latter. PDL1 data were correlated with known expression data obtained in previous studies and with clinical data of patients. So far, PDL1 expression in lymphomas has not been

ED

studied at such a large scale [20, 21]; thus, by means of the present study we considerably

PT

extend and add new knowledge on this topic. We could demonstrate that PDL1 can theoretically be used in the diagnostic setting for

CE

discriminating between THRBCL and cHL and THRBCL and NLPHL with PDL1 expression

AC

favoring the diagnosis of cHL and NLPHL, respectively. This particular finding of our study might be a helpful addendum in occasional difficult differential diagnoses [22]. With respect to other B-cell lymphomas, the diagnostic value of PDL1 seems to be low. In contrast to a previous study on PDL1 expression in THRBCL [20], we could show that the majority of THRBCL does not express PDL1 on tumor cells but it is abundantly expressed by the surrounding T-cells and histiocytes. We confirmed these results by a double immunostaining for BCL6 and PDL1, while the previous study had used a double staining for PAX5 and PDL1. Using both PDL1 clones we obtained identical results as shown in Figure 2H. In contrast to us, the former study [20] applied another antibody for PDL1 staining (clone 015). Another unexpected finding of ours was that primary testicular lymphomas investigated in our study

ACCEPTED MANUSCRIPT did not express PDL1 in contrast to a study of PDL1 expression in primary central nervous system lymphomas [23]. This difference is of interest, since both entities are summarized as

T

lymphomas arising in immuno-privileged sites. However, the mentioned study [23] also

RI P

utilized another antibody (clone 5H1) and thus might not be directly comparable with our study. Importantly, serious doubts on the validity of results assessed by older anti-PDL1

SC

antibody closes have recently been risen [24]. These authors also stress the importance of

MA NU

rigorous validation studies [25], which were also part of our study set-up. In PMBCL, the genetic basis for PDL1 over-expression was one more time confirmed by our results since gains of the PDL1 locus strongly correlated with PDL1 positivity. Several studies have already shown that aberrations of the PDL1 locus are a specific feature of PMBCL [26]

ED

and that the protein expression of PDL1 is correlated with gains of 9p24.1 [27]. Besides

PT

adding another in situ evidence for potentials of PDL1 checkpoint modulation in this entity, the reproduction of known results on our collective supports the validity of our data. Such a

CE

correlation between PDL1 expression and gains of 9p24.1 could not be detected in our cHL

AC

study cases, even if the subgroup of nodular sclerosis cHL, which have been described to have more commonly 9p24.1 amplifications [27], was analyzed separately (data not shown). One possible explanation for this is that cHL also use other mechanisms than PDL1 gains to activate the PD1-PDL1 pathway, such as e.g. EBV infection, which could increase signal transducer and activator of transcription (STAT) signaling. This signaling pathway is also activated by 9p24.1 amplifications which include the Janus kinase 2 (JAK2) locus as well [28]. Besides EBV activation, JAK2-modulated activation of PDL1 can also be achieved by other alterations at the genetic level [29] as well as by microRNA interferences [30]. Another major player activating the PD1-PDL1 pathway in cHL is activator protein 1 (AP-1), which acts as a promoter of PDL1, mainly via its components c-Jun and JunB [28].

ACCEPTED MANUSCRIPT Immunomodulatory drugs such as lenalidomide, are already in use for relapsed DLBCL [31] and PD1 blockade with nivolumab in relapsed or refractory cHL showed remarkable results

T

[16]. Nowakowski et al. could recently show that addition of lenalidomide to the R-CHOP

RI P

(rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisone) chemotherapy regimen counterbalances the negative prognostic impact of ABC-DLBCL [32]. The significant

SC

PDL1 expression in ABC-DLBCL as observed in our study, suggests that the PD1-PDL1

MA NU

pathway might be a mechanism by which ABC-DLBCL silence the immune system and explain their high sensitivity towards immunomodulation [21]. Importantly, first clinical trials using a targeted therapy approach by blocking the PD1-PDL1 pathway in DLBCL are currently being undertaken with results to be awaited [33]. Our and other’s results suggest a potential for a

ED

more tailored approach with a particular expected sensitivity of cases with activated

PT

“immune escape” like ABC-DLBCL, PMBCL and cHL. The observed correlation between PD1 positive TIL and PDL1 expression on tumor cells in

CE

general also supports the validity of our study. This correlation has also been shown in

AC

studies on solid tumors [34, 35]. One of these studies on breast cancer showed that PDL1 expression is more frequent in the aggressive basal-like subtype of breast cancer, which might be more amenable to immunomodulatory treatment than other breast cancer subtypes [36]. In in vitro experiments investigating autoimmune diseases, it has been demonstrated that blocking the PD1-PDL1 pathway helps increasing the number of FoxP3 positive T-cells [37]. This is reflected in our finding of an inverse correlation of the number FoxP3 positive TIL and PDL1 expression e.g. in cHL. Considering prognostic importance, there was a trend towards worse outcome in EBV negative cHL expressing PDL1, yet this did not improve the predictive model based on CD68

ACCEPTED MANUSCRIPT positive tumor macrophages in combination with granzyme B, FoxP3 and PD1 positive TIL in cHL proposed earlier by our and other groups [38, 39].

T

In conclusion, we investigated the largest cohort so far of Hodgkin and B-cell lymphomas for

RI P

expression of PDL1. We could show that PDL1 assessment can be of some assistance in the differential diagnosis of cHL and THRBCL and its expression favors the diagnosis of DLBCL

SC

over other aggressive B-cell lymphomas. The consistency of our results is validated by

MA NU

correlations of 9p24.1 amplifications and PDL1 expression in PMBCL as well as with regard to correlations between PDL1 expression on lymphoma cells and the presence or absence of respective T-cell subtypes and macrophages. PDL1 expression in ABC-DLBCL could be additional evidence pointing to a potential for employing immunomodulatory approaches in

PT

ACKNOWLEDGEMENTS

ED

this more treatment resistant entity.

CE

The authors would like to thank Mrs. Petra Hirschmann for performing the immunohistochemical stainings.

AC

Thomas Menter is supported by the Nuovo-Soldati Cancer Research Foundation, Vaduz, Liechtenstein.

COMPETING INTERESTS The authors declare to have no competing interests.

ACCEPTED MANUSCRIPT Tables Table 1: staining results of PDL1

N

RI P

13 41 253 17 50 10 6

SC

MA NU

Burkitt lymphoma Chronic lymphocytic leukemia Diffuse large B-cell lymphoma Transformed low grade lymphomas Follicular lymphoma(FL)G1-G2 G3 Lymphoplasmacytic lymphoma

+/evaluable (%)

T

Entity

0/7 (0%) 1/37 (3%) 80/260 (31%) 2 (all FL)/15 (13%) 3/50 (6%) 1/9 (11%) 0/6 (0%)

35 73 59

0/35 (0%) 1/54 (2%) 5/46 (11%)

Primary mediastinal B-cell lymphoma T-cell and histiocyte rich B-cell lymphoma

51 11

12/33 (36%) 1/11 (9%)

Nodular sclerosis Hodgkin lymphoma

152

87/134 (65%)

Mixed cellularity Hodgkin lymphoma

80

60/74 (81%)

14 8 11 15

9/10 (90%) 4/6 (67%) 6/9 (67%) 7/13 (54%)

PT

ED

Mantle cell lymphoma Marginal zone lymphoma low grade Marginal zone lymphoma high grade

AC

CE

Lymphocyte rich Hodgkin lymphoma Lymphocyte depleted Hodgkin lymphoma Unclassifiable classical Hodgkin lymphoma Nodular lymphocyte predominant Hodgkin lymphoma

ACCEPTED MANUSCRIPT Table 2: Details of antibodies used in this study and results from previous studies

CD20

Ventana Roche 760-4241 Dako IR604

Prediluted

CC1 mild

GCET1

Abcam ab68889

PD1

Ventana Roche 760-4448 Ventana Roche 800-2842

PAX5

EBER

Reference for the article this antibody has been used in

DAB

DAB

CC1 mild

DAB

CC1 24 min

DAB

CC1 mild

DAB

CC1 40 min

DAB

[7] [39]

1:400 overnight 4°C Prediluted incubationn

Citrate buffer pH6,98°C, 30 min CC1 24 min microwave

DAB

[40]

DAB

[8]

Prediluted

ISH Protease 3 4 min

DAB

[41]

Prediluted 1:50

CE

FOXP3

Ventana 7904508 Ventana Roche 790-4420 Abcam ab20034

Prediluted A/B Block Prediluted

DAB

T

CC1 32 min

Detection

SC

1:50

AC

CyclinD1

CC1 24 min

MA NU

BCL6

1:50

ED

PDL1

Retrieval

Cell signaling 13684, clone E1L3N Ventana Roche, clone SP142

PT

PDL1

Dilution

RI P

Source

ACCEPTED MANUSCRIPT Figure Legends Figure1: Comparison of two commercially available PDL1 antibodies. A. Non-small cell lung

T

cancer lymph node metastasis stained with the E1L3N clone; original magnification 200x. B.

RI P

Same case stained with the SP142 clone; identical magnification. Note the better signal-tonoise ratio as well as higher amount of positively staining macrophages on image A. C.

SC

Positively staining sinus-lining cells and macrophages in a reactive lymph node; original

MA NU

magnification 360x.

Figure 2: PDL1 expression in various lymphoma entities A. Reed-Sternberg cells of classical Hodgkin lymphoma; original magnification 400x; B. L & H

ED

cells of nodular lymphocyte-predominant Hodgkin lymphoma and some smaller background

PT

macrophages; original magnification 400x. C. Diffuse large B-cell lymphoma; original magnification 240x.D.Primary mediastinal B-cell lymphoma; original magnification 240x.E.

CE

Marginal zone lymphoma; note positively stained dots showing the complexes of PDL1 and

AC

PD1 in an area consisting almost exclusively of B-cells; original magnification 360x. F. Marginal zone lymphoma with a high grade component; note unequivocal expression on blasts; original magnification 360x. G. Follicular lymphoma grade 3; original magnification 360x. H. Double staining of a T-cell and histiocyte rich B-cell lymphoma showing tumor cells highlighted by a nuclear BCL6 staining (brown) and a membranous PDL1 staining (red) on tumor-infiltrating macrophages; original magnification 400x (left: anti-PDL1 antibody clone SP142; right: anti-PDL1 antibody clone E1L3N).

Figure 3: Event-free survival respecting PDL1 expression of Epstein-Barr virus (EBV)negative classical Hodgkin lymphoma patients

ACCEPTED MANUSCRIPT References

7.

8. 9.

10. 11.

12.

13. 14. 15.

16. 17. 18.

19.

20.

T

RI P

SC

6.

MA NU

5.

ED

4.

PT

.

CE

2.

Keir ME, Butte MJ, Freeman GJ, Sharpe AH. PD-1 and its ligands in tolerance and immunity. Annu Rev Immunol 2008;26:677–704. Jin H, Ahmed R, Okazaki T. Role of PD-1 in regulating T-cell immunity. Curr Top Microbiol Immunol 2011;350:17–37. Fife BT, Pauken KE. The role of the PD-1 pathway in autoimmunity and peripheral tolerance. Ann N Y Acad Sci 2011;1217:45–59. Nishimura H, Nose M, Hiai H, Minato N, Honjo T. Development of lupus-like autoimmune diseases by disruption of the PD-1 gene encoding an ITIM motif-carrying immunoreceptor. Immunity 1999;11:141–51. Weber J. Immune checkpoint proteins: a new therapeutic paradigm for cancer--preclinical background: CTLA-4 and PD-1 blockade. Semin Oncol 2010;37:430–9. Muenst S, Soysal SD, Tzankov A, Hoeller S. The PD-1/PD-L1 pathway: biological background and clinical relevance of an emerging treatment target in immunotherapy. Expert Opin Ther Targets 2015;19:201–11. Muenst S, Hoeller S, Dirnhofer S, Tzankov A. Increased programmed death-1+ tumor-infiltrating lymphocytes in classical Hodgkin lymphoma substantiate reduced overall survival. Hum Pathol 2009;40:1715–22. Muenst S, Hoeller S, Willi N, Dirnhofera S, Tzankov A. Diagnostic and prognostic utility of PD-1 in B cell lymphomas. Dis Markers 2010;29:47–53. Rosenwald A, Wright G, Leroy K, Yu X, Gaulard P, Gascoyne RD, et al. Molecular diagnosis of primary mediastinal B cell lymphoma identifies a clinically favorable subgroup of diffuse large B cell lymphoma related to Hodgkin lymphoma. J Exp Med 2003;198:851–62. Hutchinson CB, Wang E. Primary mediastinal (thymic) large B-cell lymphoma: a short review with brief discussion of mediastinal gray zone lymphoma. Arch Pathol Lab Med 2011;135:394–8. Yamamoto W, Nakamura N, Tomita N, Ishii Y, Takasaki H, Hashimoto C, et al. Clinicopathological analysis of mediastinal large B-cell lymphoma and classical Hodgkin lymphoma of the mediastinum. Leuk Lymphoma 2013;54:967–72. Blank C, Gajewski TF, Mackensen A. Interaction of PD-L1 on tumor cells with PD-1 on tumorspecific T cells as a mechanism of immune evasion: implications for tumor immunotherapy. Cancer Immunol Immunother 2005;54:307–14. Pico de Coaña Y, Choudhury A, Kiessling R. Checkpoint blockade for cancer therapy: revitalizing a suppressed immune system. Trends Mol Med 2015;21:482–91. Robert C, Long GV, Brady B, Dutriaux C, Maio M, Mortier L, et al. Nivolumab in previously untreated melanoma without BRAF mutation. N Engl J Med 2015;372:320–30. Rizvi NA, Mazières J, Planchard D, Stinchcombe TE, Dy GK, Antonia SJ, et al. Activity and safety of nivolumab, an anti-PD-1 immune checkpoint inhibitor, for patients with advanced, refractory squamous non-small-cell lung cancer (CheckMate 063): a phase 2, single-arm trial. Lancet Oncol 2015;16:257–65. Ansell SM, Lesokhin AM, Borrello I, Halwani A, Scott EC, Gutierrez M, et al. PD-1 blockade with nivolumab in relapsed or refractory Hodgkin's lymphoma. N Engl J Med 2015;372:311–9. Fankhauser CD, Curioni-Fontecedro A, Allmann V, Beyer J, Tischler V, Sulser T, et al. Frequent PDL1 expression in testicular germ cell tumors. Br J Cancer 2015;113:411–3. Meier C, Hoeller S, Bourgau C, Hirschmann P, Schwaller J, Went P, et al. Recurrent numerical aberrations of JAK2 and deregulation of the JAK2-STAT cascade in lymphomas. Mod Pathol 2009;22:476–87. Sterlacci W, Savic S, Schmid T, Oberaigner W, Auberger J, Fiegl M, Tzankov A. Tissue-sparing application of the newly proposed IASLC/ATS/ERS classification of adenocarcinoma of the lung shows practical diagnostic and prognostic impact. Am J Clin Pathol 2012;137:946–56. Chen BJ, Chapuy B, Ouyang J, Sun HH, Roemer MGM, Xu ML, et al. PD-L1 expression is characteristic of a subset of aggressive B-cell lymphomas and virus-associated malignancies. Clin Cancer Res 2013;19:3462–73.

AC

1.

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA NU

SC

RI P

T

21. Andorsky DJ, Yamada RE, Said J, Pinkus GS, Betting DJ, Timmerman JM. Programmed death ligand 1 is expressed by non-hodgkin lymphomas and inhibits the activity of tumor-associated T cells. Clin Cancer Res 2011;17:4232–44. 22. Traverse-Glehen A, Pittaluga S, Gaulard P, Sorbara L, Alonso MA, Raffeld M, Jaffe ES. Mediastinal gray zone lymphoma: the missing link between classic Hodgkin's lymphoma and mediastinal large B-cell lymphoma. Am J Surg Pathol 2005;29:1411–21. 23. Berghoff AS, Ricken G, Widhalm G, Rajky O, Hainfellner JA, Birner P, et al. PD1 (CD279) and PD-L1 (CD274, B7H1) expression in primary central nervous system lymphomas (PCNSL). Clin Neuropathol 2014;33:42–9. 24. Rimm D, Schalper K, Pusztai L. Unvalidated antibodies and misleading results. Breast Cancer Res Treat 2014;147:457–8. 25. Velcheti V, Schalper KA, Carvajal DE, Anagnostou VK, Syrigos KN, Sznol M, et al. Programmed death ligand-1 expression in non-small cell lung cancer. Lab Invest 2014;94:107–16. 26. Joos S, Otaño-Joos MI, Ziegler S, Brüderlein S, Du Manoir S, Bentz M, et al. Primary mediastinal (thymic) B-cell lymphoma is characterized by gains of chromosomal material including 9p and amplification of the REL gene. Blood 1996;87:1571–8. 27. Green MR, Monti S, Rodig SJ, Juszczynski P, Currie T, O'Donnell E, et al. Integrative analysis reveals selective 9p24.1 amplification, increased PD-1 ligand expression, and further induction via JAK2 in nodular sclerosing Hodgkin lymphoma and primary mediastinal large B-cell lymphoma. Blood 2010;116:3268–77. 28. Green MR, Rodig S, Juszczynski P, Ouyang J, Sinha P, O'Donnell E, et al. Constitutive AP-1 activity and EBV infection induce PD-L1 in Hodgkin lymphomas and posttransplant lymphoproliferative disorders: implications for targeted therapy. Clin Cancer Res 2012;18:1611–8. 29. Weniger MA, Melzner I, Menz CK, Wegener S, Bucur AJ, Dorsch K, et al. Mutations of the tumor suppressor gene SOCS-1 in classical Hodgkin lymphoma are frequent and associated with nuclear phospho-STAT5 accumulation. Oncogene 2006;25:2679–84. 30. Navarro A, Diaz T, Martinez A, Gaya A, Pons A, Gel B, et al. Regulation of JAK2 by miR-135a: prognostic impact in classic Hodgkin lymphoma. Blood 2009;114:2945–51. 31. Wiernik PH, Lossos IS, Tuscano JM, Justice G, Vose JM, Cole CE, et al. Lenalidomide monotherapy in relapsed or refractory aggressive non-Hodgkin's lymphoma. J Clin Oncol 2008;26:4952–7. 32. Nowakowski GS, LaPlant B, Macon WR, Reeder CB, Foran JM, Nelson GD, et al. Lenalidomide combined with R-CHOP overcomes negative prognostic impact of non-germinal center B-cell phenotype in newly diagnosed diffuse large B-Cell lymphoma: a phase II study. J Clin Oncol 2015;33:251–7. 33. Cheah CY, Fowler NH, Neelapu SS. Targeting the programmed death-1/programmed deathligand 1 axis in lymphoma. Curr Opin Oncol 2015;27:384–91. 34. Bertucci F, Finetti P, Mamessier E, Pantaleo MA, Astolfi A, Ostrowski J, Birnbaum D. PDL1 expression is an independent prognostic factor in localized GIST. Oncoimmunology 2015;4:e1002729. 35. Sabatier R, Finetti P, Mamessier E, Adelaide J, Chaffanet M, Ali HR, et al. Prognostic and predictive value of PDL1 expression in breast cancer. Oncotarget 2015;6:5449–64. 36. Ghebeh H, Barhoush E, Tulbah A, Elkum N, Al-Tweigeri T, Dermime S. FOXP3+ Tregs and B7H1+/PD-1+ T lymphocytes co-infiltrate the tumor tissues of high-risk breast cancer patients: Implication for immunotherapy. BMC Cancer 2008;8:57. 37. Singh RP, La Cava A, Hahn BH. pConsensus peptide induces tolerogenic CD8+ T cells in lupusprone (NZB x NZW)F1 mice by differentially regulating Foxp3 and PD1 molecules. J Immunol 2008;180:2069–80. 38. 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. 39. Tzankov A, Matter MS, Dirnhofer S. Refined prognostic role of CD68-positive tumor macrophages in the context of the cellular micromilieu of classical Hodgkin lymphoma. Pathobiology 2010;77:301–8.

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA NU

SC

RI P

T

40. Menter T, Gasser A, Juskevicius D, Dirnhofer S, Tzankov A. Diagnostic Utility of the Germinal Center-associated Markers GCET1, HGAL, and LMO2 in Hematolymphoid Neoplasms. Appl Immunohistochem Mol Morphol 2015;23:491–8. 41. Krugmann J, Tzankov A, Gschwendtner A, Fischhofer M, Greil R, Fend F, Dirnhofer S. Longer failure-free survival interval of Epstein-Barr virus-associated classical Hodgkin's lymphoma: a single-institution study. Mod Pathol. 2003 Jun;16(6):566-73.

AC

CE

PT

ED

MA NU

SC

RI P

T

ACCEPTED MANUSCRIPT

Figure 1

AC

CE

PT

ED

MA NU

SC

RI P

T

ACCEPTED MANUSCRIPT

Figure 2

SC

RI P

T

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA NU

Figure 3