cancers

Article Screening a Broad Range of Solid and Haematological Tumour Types for CD70 Expression Using a Uniform IHC Methodology as Potential Patient Stratification Method

Tal Flieswasser 1,2, Valérie Camara-Clayette 3, Alina Danu 4, Jacques Bosq 5, Vincent Ribrag 6,7, Piotr Zabrocki 8, Luc Van Rompaey 8, Hans de Haard 8,*, Karen Zwaenepoel 1,2, Evelien Smits 1,9 , Patrick Pauwels 1,2 and Julie Jacobs 1,2

1 Center for Oncological Research (CORE), University of Antwerp, 2610 Wilrijk, Belgium; tal.fl[email protected] (T.F.); [email protected] (K.Z.); [email protected] (E.S.); [email protected] (P.P.); [email protected] (J.J.) 2 Department of Pathology, Antwerp University Hospital, 2650 Edegem, Belgium 3 Translational Hematology Unit, AMMICa INSERMUS23/CNRS UMR3655, Gustave Roussy Cancer Campus, 94805 Villejuif, France; [email protected] 4 Hematology Department, Gustave Roussy Cancer Campus, 94805 Villejuif, France; [email protected] 5 Pathology Laboratory, Gustave Roussy Cancer Campus, 94805 Villejuif, France; [email protected] 6 Département des Innovations Thérapeutiques et Essais Précoces (DITEP), Université Paris-Saclay, Gustave Roussy Cancer Campus, 94805 Villejuif, France; [email protected] 7 INSERM U1170, 94805 Villejuif, France 8 Argenx, 9052 Zwijnaarde, Belgium; [email protected] (P.Z.); [email protected] (L.V.R.) 9 Center for Cell Therapy and Regenerative Medicine, Antwerp University Hospital, 2650 Edegem, Belgium * Correspondence: [email protected]

 Received: 20 September 2019; Accepted: 18 October 2019; Published: 22 October 2019 

Abstract: The constitutive expression of CD70 has been described in various haematological and solid tumour types. In addition, the co-expression of its receptor in tumours has been demonstrated, mediating tumour cell proliferation. Although CD70 expression is a prerequisite to enrol patients in solid tumour clinical trials using anti-CD70 immunotherapy, there is currently no standardised test to evaluate CD70 expression. These differences in immunohistochemistry (IHC) protocols make it challenging to compare the expression levels that were obtained in different studies, pointing out the need for one uniform methodology. In this retrospective study, over 600 tumour samples from different solid and haematological malignancies were analysed while using one validated IHC method. CD70 and CD27 expression was demonstrated in a broad range of tumour types. In solid tumours, 43% demonstrated CD70 positivity with the highest degree in (79.5%). Kaposi sarcoma showed no CD70 expression on the tumour cells. In samples, 58% demonstrated CD70 positivity. Moreover, the co-expression of CD70 and CD27 was observed in 39% of lymphoma samples. These findings highlight the need to further explore anti-CD70 therapies in a broad range of CD70 expressing tumour types and in doing so, implementing one standardised protocol to define CD70 overexpression to use it as a diagnostic tool.

Keywords: CD70; CD27; immunohistochemistry; biomarker;

Cancers 2019, 11, 1611; doi:10.3390/cancers11101611 www.mdpi.com/journal/cancers Cancers 2019, 11, 1611 2 of 13

1. Introduction Numerous successful clinical trials have led to a great number of FDA approvals for PD-1/PD- blockade therapy in a spectrum of different tumour types: melanoma, non-small cell lung cancer, urothelial carcinoma, classic Hodgkin lymphoma, renal cell carcinoma, Merkel cell carcinoma, hepatocellular carcinoma, gastric carcinomas, and head and neck squamous cell carcinoma. Although the blockage of the PD-1/PD-L1 axis has elicited durable clinical responses in many patients, only 10–40% show clinical response to monotherapy, pointing out the possibility that alternative pathways may have significant impact on the antitumor immune activity in patients with various tumour types [1,2]. In this regard, CD70 is a ligand of the tumour necrosis factor superfamily, of which its expression is tightly regulated under physiological conditions: it is only transiently expressed on antigen-activated T and B cells, mature dendritic cells, and a unique population of antigen presenting cells, which is exclusively localised in the gut lamina propria [3–5]. Its unique receptor, CD27, acts in a costimulatory pathway, leading to proliferation, differentiation, and survival signals. Depending on the amount, duration, and timing of CD70 expression, CD27 signalling might either lead to improved T cell function or T cell dysfunction [3,6]. The overexpression of CD70 on tumour cells has been described in multiple haematological malignancies and carcinomas [7–9]. This overexpression facilitates immune suppression through CD27 signalling in the tumour microenvironment by enhanced survival of regulatory T cells, induction of T cell apoptosis, and skewing T cells towards a T cell exhausted phenotype [10–13]. In addition, many of the haematological malignancies also co-express CD27 on the tumour cells, which enhances tumour cell proliferation and survival [14,15]. For myeloid malignancies, CD70 was demonstrated to play a critical role in leukemic stem cell maintenance and proliferation [14,16]. Consequently, anti-CD70 immunotherapy is being evaluated in clinical trials and it has been associated with objective clinical responses against various CD70 positive malignancies [17]. Thereof, 92% response rates could be seen in acute myeloid leukaemia patients while using CD70-targeting antibodies in a Phase 1 trial [18]. Different methods to detect CD70 expression in tissues have been described, such as real-time PCR and western blot [19]. To date, immunohistochemistry (IHC) remains the most frequently used method for determining CD70 expression due to its simple and rapid detection of protein expression levels and distribution. Although CD70 positivity of the tumour cells is a prerequisite to enter solid tumour clinical trials while using anti-CD70 immunotherapy, there is currently no uniform IHC assay for evaluating CD70 expression. Consequently, differences in the use of antibody clones and in thresholds for positivity lead to discrepancies in the percentage of CD70 positive cases within identical tumour types, pointing out the need for one uniform and validated methodology [20]. In this retrospective study, we assessed the degree of CD70/CD27 expression across various tumour types while using one standardised and validated IHC assay.

2. Results

2.1. CD70 Staining Patterns in Solid Tumour Types CD70 expression was assessed in 25 different solid tumour types (N = 496), which ranged from two to 101 samples per tumour type (Figure1). The staining of CD70 in various tumour types was localised either on the membrane or in the cytoplasm. Intensity of CD70 expression was categorised as no, weak, moderate, or strong staining (Figure2). CD70 expression was divided into three groups; no/low (0–10%), moderate (11–50%) and strong expression (>50%), based on the percentage of CD70 positive cells (of any intensity) (Figure1). The cut-off value of 10% was used to determine CD70 positivity, which is in accordance with the cut-off used in clinical trials of anti-CD70 therapy [17]. CD70 positivity (>10%) was observed in all solid tumour types, with the exception of Kaposi sarcoma (N = 7) (Figure1). CD70 positivity was significantly most prevalent in the samples of renal cell carcinoma (79.5%), as compared to all other tumour types (p < 0.0001). In addition, the majority of samples from oesophageal carcinoma Cancers 2019, 11, 1611 3 of 13 Cancers 2019, 11, x 3 of 13

N( =66.7%, 12) andN =mesothelioma12) and mesothelioma (62.5%, N = (62.5%, 8) showedN = CD708) showed positivity. CD70 CD70 positivity. expression CD70 in expression more than in 50% more ofthan the tumour 50% of thecells tumour was mainly cells wasobserved mainly in observedrenal cell incarcinoma renal cell (61 carcinoma.5%, N = (61.5%,39), gastricN = carcinoma39), gastric (50%carcinoma, N = 6), (50%, and adenoidN = 6), and cystic adenoid carcinoma cystic (60% carcinoma, N = 5). (60%, Contrarily,N = 5). CD70 Contrarily, negative CD70 tumour negative cells tumour (0– 10%)cells were (0–10%) mostly were observed mostly observedin prostate in carcinoma prostate carcinoma (75%, N (75%,= 12), NLangerhans= 12), Langerhans cell histiocytosis cell histiocytosis (80%, N( =80%, 5), KaposiN = 5), sarcoma Kaposi sarcoma(100%, N (100%, = 7), andN colorectal= 7), and carcinoma colorectal carcinoma(80%, N = 75 (80%,), wherebyN = 75), a sig wherebynificant a lowersignificant expression lower of expression CD70 was of observed CD70 was in observedthe latterin two the group latters two as compared groups as comparedto all other to tumour all other typestumour (p = types0.012 and (p = p0.012 < 0.0 and001,p respectively< 0.0001, respectively).).

FigureFigure 1. 1.ProportionProportion of ofCD70 CD70 staining staining among among solid solid tumour tumour types. types. CD70 CD70 staining staining was was categori categorisedsed in 3 in different3 different groups groups based based on onthe thepercentage percentage of tumour of tumour cells cells staining staining positive: positive: expression expression levels levels rangin rangingg fromfrom 0 to 0 to 10% 10% (empty (empty bars bars);); between between 10 10 to to50% 50% (pattern (patterneded bars bars);); and and,, expression expression levels levels above above 50% 50% (solid(solid bars bars).). * p *

Cancers 2019, 11, x 4 of 13 tissueCancers 2019biopsies,, 11, 1611 such as lung carcinoma (85.7%, N = 7 vs 40%, N = 80) (p = 0.048) and pancreatic4 of 13 carcinoma (87.5%, N = 8 vs 26.1%, N = 23) (p = 0.004). We observed different CD70 staining intensities amongst the tumour types. Ovarian carcinoma andIn several endometrial tumour carcinoma types, CD70 mainly positivity showed was weak higher positive in metastatic staining. specimens Moderate as comparedstaining was to primarymostly observedtissue biopsies, in mesothelioma such as lung, while carcinoma most patients (85.7%, withN = renal7 vs 40%,cell carcinomaN = 80) ( pexhibited= 0.048) strong and pancreatic staining whencarcinoma compared (87.5%, to Nother= 8 vssolid 26.1%, tumourN = types23) (p (Supplementary= 0.004). Figure S1).

FigureFigure 2. MicrographsMicrographs of of CD70 CD70 immunohistochemical staining stainingss across across various various solid solid tumour types showingshowing range range of of staining staining percentages. percentages (A.) > (A50%) >50% CD70 CD70 staining staining on tumour on tumour cells in renal cells cell in carcinoma;renal cell carcinoma(B) 10–50%; staining (B) 10–50% in melanoma; staining in (C melanoma) <10% staining; (C) in<10 breast% staining cancer; in and, breast (D) CD70 cancer staining; and, (D in) controlCD70 stainingtissue (tonsil). in control Magnitude tissue (tonsil). 200 . Magnitude 200×. × 2.2. CD70We observed and CD27 di Proteinfferent Expression CD70 staining in Haematological intensities amongst Malignancies the tumour types. Ovarian carcinoma and endometrial carcinoma mainly showed weak positive staining. Moderate staining was mostly Lymphoma samples (N = 130) were analysed for CD70 and CD27 expression and they were observed in mesothelioma, while most patients with renal cell carcinoma exhibited strong staining categorised in five histological subgroups according to their defined pathology: mantle cell when compared to other solid tumour types (Supplementary Figure S1). lymphoma (MCL, N = 65), peripheral T cell lymphoma (PTCL, N = 50), NK/T-cell lymphoma (NKTCL, N2.2. = 4), CD70 anaplastic and CD27 large Protein cell lymphoma Expression in(ALCL, Haematological N = 3), and Malignancies diffuse large B cell lymphoma (DLBCL, N = 8). Lymphoma samples displayed different CD70 expression levels, as shown in Figure 3. Of the 50 PTCLLymphoma samples, four samples had unknown (N = 130) CD70 were levels analysed due forto tissue CD70 that and was CD27 not expression representative. and theyOf the were 50 PTCL,categorised three inALCL five histologicaland eight DLBCL subgroups samples, according seven to, theirone, and defined three pathology: patient(s) mantle had unknown cell lymphoma CD27 levels,(MCL, respectively.N = 65), peripheral T cell lymphoma (PTCL, N = 50), NK/T-cell lymphoma (NKTCL, N = 4), anaplasticCD70 positive large cell tumour lymphoma cells (>10%) (ALCL, wereN = 3),observed and di ffinuse all largelymphoma B cell types, lymphoma with the (DLBCL, exceptionN = of8). NKTCLLymphoma patients’ samples samples displayed, which dididfferent not show CD70 any expression expression levels, of CD70 as shown(Figure in4). FigureCD70 3expression. Of the was50 PTCL detectable samples, in 59% four of had PTCL. unknown Twenty CD70-five levelspatients due with to tissuePTCL that(54%) was had not CD70 representative. expression Oflevels the between50 PTCL, 10 three–50% ALCL and two and patients eight DLBCL (4%) samples,had very seven,high levels one, andof CD70 three (>50%). patient(s) In hadaddition, unknown seven CD27 out oflevels, eight respectively. DLBCL samples (88%) and two out of three ALCL samples showed CD70 positivity. ACD70 larger positive cohort tumour of 65 MCL cells patients’ (>10%) weresamples observed were analysed in all lymphoma for CD70 types, expression with the. An exception overview of ofNKTCL the patient patients’ characteristics samples, which can be did found not showin Supplementary any expression Table of CD70 S1. Thirty (Figure-seven4). CD70 tumour expression samples (57%)was detectable were collected in 59% before of PTCL. treatment Twenty-five and 28 patients(43%) at withrelapse. PTCL 52 (54%)samples had (80%) CD70 were expression from patients levels atbetween stage III 10–50% or IV andof the two disease. patients Most (4%) of had the very samples high levels were ofcollected CD70 (> from50%). lymph In addition, nodes seven (57%) out and of gastrointestinaleight DLBCL samples tract (18%). (88%) and CD70 two positivity out of three was ALCL found samples in 37 MCLshowed samples CD70 positivity. (57%), of which 26

Cancers 2019, 11, 1611 5 of 13

Cancers 2019, 11, x 5 of 13 A larger cohort of 65 MCL patients’ samples were analysed for CD70 expression. An overview of samplesthe patient (70%) characteristics displayed CD70 can be expression found in Supplementary on >50% of the Table tumour S1. Thirty-seven cells (Figure tumour 4, Table samples 1). We (57%) found no significantwereCancers collected 2019 , difference11, x before treatment in CD70 and expressio 28 (43%)n at between relapse. 52the samples samples (80%) at werediagnosis frompatients or relapse. at stage5 ofAlthough 13 III a slightor IV enrichment of the disease. in MostCD70 of expression the samples in were lymph collected nodes from could lymph be nodesfound (57%) in the and samples gastrointestinal of relapsed tract (18%). CD70 positivity was found in 37 MCL samples (57%), of which 26 samples (70%) displayed patientssamples as opposed (70%) displayed to the samplesCD70 expression that were on >50%taken of at the diagnosis tumour cells (56% (Figure vs 43%, 4, Table respectively) 1). We found (P>0.05) CD70no significant expression difference on >50% in ofCD70 the tumourexpressio cellsn between (Figure 4the, Table samples1). We at founddiagnosis no significantor relapse. diAlthoughfference (Table 1). ina slight CD70 enrichment expression betweenin CD70 expression the samples in atlymph diagnosis nodes or could relapse. be found Although in the a samples slight enrichment of relapsed in CD70patients expression as opposed in lymph to the nodessamples could that be were found taken in the at diagnosis samples of (56% relapsed vs 43%, patients respectively) as opposed (P>0.05) to the samples(Table 1). that were taken at diagnosis (56% vs 43%, respectively) (P>0.05) (Table1).

Figure 3. Representative micrographs of CD70 immunohistochemical staining on samples of patients withFigure Figurehaematological 3.3. RepresentativeRepresentative malignancies. micrographs (A) ofCD70 CD70 staining immunohistochemical immunohistochemical in a mantle cell staining staining lymphoma on on samples samples; (B) cutaneous of of patients patients T-cell lymphomawithwith haematological haematological; and, (C) diffuse malignancies.malignancies. large B cell (A) CD70 lymphoma staining is in in shown a a mantle mantle. Magnitude cell cell lymphoma lymphoma; 100; ×(B. ( B) )cutaneous cutaneous T T-cell-cell lymphoma;lymphoma; and,and, ((CC)) didiffuseffuse largelarge BB cellcell lymphomalymphoma is shown.shown. Magnitude Magnitude 100 100×. . ×

Figure 4. Proportion of CD70 staining among various lymphoma subsets. CD70 staining was Figure 4. Proportion of CD70 staining among various lymphoma subsets. CD70 staining was categorised in three different groups: CD70 positive tumour cells ranging from 0 to 10% (white bars); Figurecategorised 4. Proportion in three di offf erent CD70 groups: staining CD70 among positive various tumour cellslymphoma ranging fromsubsets. 0 to 10% CD70 (white staining bars); was between 10% to 50% (patterned bars); and expression levels above 50% (blue bars). ALCL, anaplastic categoribetweensed in 10% three to 50% different (patterned groups: bars); CD70 and expression positive tumour levels above cells 50%ranging (blue from bars). 0 ALCL, to 10% anaplastic (white bars); large cell lymphoma; DLBCL, diffuse large B cell lymphoma; MCL, mantle cell lymphoma; NKTCL, betweenlarge 10 cell% lymphoma; to 50% (pattern DLBCL,ed dibarsffuse); and large expression B cell lymphoma; levels above MCL, mantle50% (blue cell lymphoma;bars). ALCL, NKTCL, anaplastic NK/T cell lymphoma; PTCL, peripheral T-cell lymphoma. largeNK cell/T celllymphoma; lymphoma; DLBCL, PTCL, peripheraldiffuse large T-cell B cell lymphoma. lymphoma; MCL, mantle cell lymphoma; NKTCL, NK/T cell lymphoma; PTCL, peripheral T-cell lymphoma. CD27Table expression1. CD70 and CD27 was assessed expression in analysis all lymphoma in mantle samplescell lymphoma (Figure (MCL5).) All samples. of the subsets showed CD27 positivity (>10% CD27 positive tumourDiagnosis cells) (FigureRelapse5). CD27All Tumour positivity Samples was observed in Table 1. CD70 andCD70/CD27 CD27 expression analysis in mantle cell lymphoma (MCL) samples. 23 patients with PTCL (54%), one out of twoN = 37 ALCL patients,N = 28 one out of fourN = patients65 with NKTCL, and one out of fiveCD70 patients+, n (%) with DLBCL.Diagnosis20 (54) 17Relapse (61) All 37Tumour (57) Samples CD70/CD27 For MCL,Staining CD27 %, expression median (range) was detectable60%N (10 = –3790%) in 41 cases60% (10N (63%), –=90%) 28 24 at diagnosis60% (10–N90%) = (65%), 65 and a slight drop in the expression≥CD70 50% staining,+, atn (%) relapse n (%) was observed2013 (65)(54)(17 samples1317 (76) (61%), (61) p > 0.05).26 Co-expression(70)37 (57) of CD70 CD27+, n (%) 24 (65) 17 (61) 41(63) and CD27Staining was observed %, median in (range) 19 PTCL 60% (46%, (10N–90%)= 41), one60% ALCL(10–90%) (N = 2),60% and (10 one–90%) DLBCL (N = 5) Staining %, median (range) 60% (30–90%) 70% (10–90%) 70% (10–90%) patient(s) (Figure6). ≥ 50%≥ 50% staining, staining, n (%)n (%) 1316 (67)(65) 1313 (76) (76) 29 (71)26 (70) + CD70CD27+, ,CD27 n (%)+, n (%) 2415 (41)(65) 101 (36)7 (61 ) 25 (38)41 (63) Staining %, median (range) 60% (30–90%) 70% (10–90%) 70% (10–90%) CD27 expression was assessed in all lymphoma samples (Figure 5). All of the subsets showed ≥ 50% staining, n (%) 16 (67) 13 (76) 29 (71) CD27 positivity (>10% CD27 positive tumour cells) (Figure 5). CD27 positivity was observed in 23 CD70+, CD27+, n (%) 15 (41) 10 (36) 25 (38)

CD27 expression was assessed in all lymphoma samples (Figure 5). All of the subsets showed CD27 positivity (>10% CD27 positive tumour cells) (Figure 5). CD27 positivity was observed in 23

Cancers 2019, 11, 1611 6 of 13

Table 1. CD70 and CD27 expression analysis in mantle cell lymphoma (MCL) samples.

Diagnosis Relapse All Tumour Samples CD70/CD27 N = 37 N = 28 N = 65 Cancers 2019, 11, x 6 of 13 CD70+, n (%) 20 (54) 17 (61) 37 (57) patients withStaining PTCL %, (54%), median one (range) out of two60% ALCL (10–90%) patients,60% one(10–90%) out of four patient 60%s with (10–90%) NKTCL, and one out of five 50%patient staining,s withn (%)DLBCL. 13 (65) 13 (76) 26 (70) ≥ CD27+, n (%) 24 (65) 17 (61) 41(63) Cancers 2019Staining, 11, x %, median (range) 60% (30–90%) 70% (10–90%) 70% (10–90%) 6 of 13 50% staining, n (%) 16 (67) 13 (76) 29 (71) patients with≥ PTCL (54%), one out of two ALCL patients, one out of four patients with NKTCL, and + + one out of fiveCD70 patient, CD27s with, n (%)DLBCL. 15 (41) 10 (36) 25 (38)

Figure 5. Proportion of CD27 staining among various lymphoma subsets. CD27 staining was categorised in three different groups: CD27 positive tumour cells ranging from 0 to 10% (white bars); between 10 to 50% (patterned bars); and expression levels above 50% (red bars). ALCL, anaplastic large cell lymphoma; DLBCL, diffuse large B cell lymphoma; MCL, mantle cell lymphoma; NKTCL, NK/T cell lymphoma; PTCL, peripheral T-cell lymphoma. FigureFigure 5. 5.ProportionProportion of of CD27 CD27 staining staining among among various various lymphoma lymphoma subsets. subsets. CD27 CD27 staining staining was was For MCL, CD27 expression was detectable in 41 cases (63%), 24 at diagnosis (65%), and a slight categoricategorisedsed in inthree three different different groups: groups: CD27 CD27 positive positive tumour tumour cells cells ranging ranging from from 0 to 0 to10% 10% (white (white bar bars);s); drop in the expression at relapse was observed (17 samples (61%), p > 0.05). Co-expression of CD70 betweenbetween 10 10to to50% 50% (pattern (patterneded bars bars);); and and expression expression levels levels above above 50% 50% (red (red bars bars).). ALCL, ALCL, anaplastic anaplastic and CD27 was observed in 19 PTCL (46%, N = 41), one ALCL (N = 2), and one DLBCL (N = 5) patient(s) largelarge cell cell lymphoma; lymphoma; DLBCL, DLBCL, diffuse diffuse large large B cell B cell lymphoma; lymphoma; MCL, MCL, mantle mantle cell cell lymphoma; lymphoma; NKTCL, NKTCL, (FigureNK/TNK 6). /cellT cell lymphoma lymphoma;; PTCL, PTCL, peripheral peripheral T- T-cellcell lymphoma. lymphoma.

For MCL, CD27 expression was detectable in 41 cases (63%), 24 at diagnosis (65%), and a slight drop in the expression at relapse was observed (17 samples (61%), p > 0.05). Co-expression of CD70 and CD27 was observed in 19 PTCL (46%, N = 41), one ALCL (N = 2), and one DLBCL (N = 5) patient(s) (Figure 6).

FigureFigure 6. Overview 6. Overview of CD70 of CD70 and and CD27 CD27 expression expression in in117 117 lymphoma lymphoma samples samples.. Each Each symbol symbol represents represents a a differentdifferent patient’spatient’s sample sample and and bars bar represents represent the the mean. mean ALCL,. ALCL, anaplastic anaplastic large large cell lymphoma; cell lymphoma; DLBCL, DLBCL,diffuse diffuse large Blarge cell B lymphoma; cell lymphoma; MCL, MCL, mantle mantle cell lymphoma; cell lymphoma; NKTCL, NKTCL, NK/ TNK/ cellT lymphomacell lymphoma PTCL, PTCL,peripheral peripheral T-cell T- lymphoma.cell lymphoma. TS, tumourTS, tumour sample sample

TwentyTwenty-five-five MCL MCL samples samples (38%) were were positive positive for for both both CD70 CD70 and and CD27, CD27, whereof whereof 60% (15 60% samples) (15 Figure 6. Overview of CD70 and CD27 expression in 117 lymphoma samples. Each symbol represents samples)showed show doubleed positivitydouble positivity within within50% of ≥ tumour 50% of cells. tumour The majoritycells. The of majority double positive of double samples positive (60%) a different patient’s sample and≥ bars represent the mean. ALCL, anaplastic large cell lymphoma; sampleswas collected (60%) was from collected patients from at diagnosispatients at ( pdiagnosis> 0.05). ( Twenty-eightp > 0.05). Twenty MCL-eight tumour MCL samplestumour samples (43%) had DLBCL, diffuse large B cell lymphoma; MCL, mantle cell lymphoma; NKTCL, NK/T cell lymphoma a very low CD70 expression (< 10%). Thereof, 12 samples also displayed a very low level of CD27 (43%)PTCL, had a peripheral very low T CD70-cell lymphoma. expression TS (<, tumour 10%). Thereof,sample 12 samples also displayed a very low level of (TableCD271 ).(Table Figure 1).7 showsFigure representative 7 shows representative images of CD70 images and of CD27 CD70 expression and CD27 in MCLexpression samples. in MCL Cyclin samplesTwenty. Cyclin-five D1 MCL (Figure samples 7F), CD3 (38%), and were CD20 positive staining fors ( bothSupplementary CD70 and F CD27,igure Swhereof2) were 60%used (15as controlsamples) marker showsed to doubleensure specificpositivity CD70/CD27 within ≥ 50% staining of tumour on the malignantcells. The majoritycells. of double positive samplesThe (6co0-%)expression was collected of CD70 from and patients CD27 at in diagnosis all haematological (p > 0.05). Twenty samples-eight showed MCL atumour weak positivesamples correlation(43%) had a(p very = 0.002 low, correlation CD70 expression coefficient (< 10%). (R) = Thereof, 0.3). 12 samples also displayed a very low level of CD27 (Table 1). Figure 7 shows representative images of CD70 and CD27 expression in MCL samples. Cyclin D1 (Figure 7F), CD3, and CD20 stainings (Supplementary Figure S2) were used as control markers to ensure specific CD70/CD27 staining on the malignant cells. The co-expression of CD70 and CD27 in all haematological samples showed a weak positive correlation (p = 0.002, correlation coefficient (R) = 0.3).

Cancers 2019, 11, 1611 7 of 13

D1 (Figure7F), CD3, and CD20 stainings (Supplementary Figure S2) were used as control markers to Cancersensure 2019 specific, 11, x CD70/CD27 staining on the malignant cells. 7 of 13

Figure 7. Micrographs of immunohistochemical stainings in one MCL patient sample. Sequential cuts Figure 7. Micrographs of immunohistochemical stainings in one MCL patient sample. Sequential cuts of one MCL patient sample showing staining of (A) HE; (B) CD70; (C) CD27; (D) PD-L1; (E) PD-1; of one MCL patient sample showing staining of (A) HE; (B) CD70; (C) CD27; (D) PD-L1; (E) PD-1; and, (F) nuclear staining of MCL cells for cyclin D1. Magnitude 40 . and, (F) nuclear staining of MCL cells for cyclin D1. Magnitude 40×. The co-expression of CD70 and CD27 in all haematological samples showed a weak positive 2.3. Programmed Cell Death Protein 1 (PD-1) and Its Ligand (PD-L1) Expression in MCL Patient Cohort correlation (p = 0.002, correlation coefficient (R) = 0.3). In addition to CD70/CD27 staining, the expression of PD-L1 and PD-1 was analysed in the large MCL2.3. Programmed cohort to estimate Cell Death the Protein expression 1 (PD-1) of other and Its relevant Ligand (PD-L1) immun Expressionotherapeutic in MCLtargets Patient (Figure Cohort 7D,E). PD-L1In was addition undetectable to CD70 in/CD27 the majority staining, of the MCL expression samples of. Twelve PD-L1 tumour and PD-1 samples was analysed (18%) had in thebetween large 5MCL–25% cohort PD-L1 to positive estimate cells the and expression they were of otherconsidered relevant as immunotherapeuticPD-L1 positive (PD-L1+) targets (Table (Figure 2). 7SevenD,E). PDPD-L1-L1- waspositive undetectable samples ( in58 the%) were majority newly of MCLdiagnosed samples. MCL Twelve patients tumour (19% samplesof all samples (18%) at had diagnosis) between and5–25% 5 (42 PD-L1%) relapsed/refractory positive cells and MCL they werepatients considered (18% of all as relapse PD-L1 positivesamples). (PD-L1 +) (Table2). Seven PD-L1-positive samples (58%) were newly diagnosed MCL patients (19% of all samples at diagnosis) and 5 (42%) relapsed/refractoryTable 2. PD MCL-1 and patients PD-L1 expression (18% of all analysis relapse in samples).MCL samples. Diagnosis Relapse All Tumour Samples PD-L1/PD-1 Table 2. PD-1 and PD-L1N = 37 expression analysisN in= 28 MCL samples. N = 65 PD-L1+, n (%) 7 (19) 5 (18) 12 (18) Diagnosis Relapse All Tumour Samples PD-L1/PD-1 Staining %, median (range) 10%N =(1037–25%) N10%= 28(5–25%) N =10%65 (5–25%) PD-1+, n (%) 14 (38) 9 (32) 23 (35) PD-L1+, n (%) 7 (19) 5 (18) 12 (18) Staining %, median (range) 10% (5–25%) 10% (5–50%) 10% (5–50%) PDStaining-1+, PD %,-L1 median+, n (%) (range) 10% (10–25%)5 (14) 10% (5–25%)3 (11) 10% (5–25%)8 (12) PD-1+, n (%) 14 (38) 9 (32) 23 (35) In total, 23 samples (35%) were positive for PD-1 with staining typically between 5–25% of cells per sample.Staining Thirty %,- medianeight samples (range) (58%)10% were (5–25%) double negative 10% (5–50%) for PD-L1 or PD 10%-1 (22 (5–50%) at diagnosis and 16 at relapse)PD-1 and+, PD-L1 only+ eight, n (%) samples (12%) 5 (14) were double positive 3 (11) for both PD-L1 8 (12)and PD-1. A weak positive correlation was observed between PD-L1 and PD-1 with (p = 0.05, R = 0.34). Interestingly, significantlyIn total, higher 23 samples expression (35%) werelevels positive of CD70 for and PD-1 CD27 with could staining be found typically as opposed between 5–25%to PD-L1 of cellsand perPD-1 sample. (p < 0.0001) Thirty-eight (Figure samples8). (58%) were double negative for PD-L1 or PD-1 (22 at diagnosis and 16 at relapse) and only eight samples (12%) were double positive for both PD-L1 and PD-1. A weak positive correlation was observed between PD-L1 and PD-1 with (p = 0.05, R = 0.34). Interestingly,

Cancers 2019, 11, 1611 8 of 13 significantly higher expression levels of CD70 and CD27 could be found as opposed to PD-L1 and CancersPD-1 ( 2019p < ,0.0001) 11, x (Figure8). 8 of 13

Figure 8. Expression levels of CD70, CD27, PD-1, and PD-L1 in MCL cohort. The expression of CD70 Figure 8. Expression levels of CD70, CD27, PD-1, and PD-L1 in MCL cohort. The expression of CD70 CD27, PD-1, and PD-L1 in a total of 65 MCL samples. Y axis shows percentage of positive cells for a CD27, PD-1, and PD-L1 in a total of 65 MCL samples. Y axis shows percentage of positive cells for a particular protein. Each symbol represents a different patient. ** p < 0.01, **** p < 0.0001. particular protein. Each symbol represents a different patient. ** p < 0.01, **** p < 0.0001. 3. Discussion 3. Discussion Our study demonstrated CD70 expression with varying intensities in all tumour types, excluding KaposiOur sarcoma study and demonstrated NKTCL. Of CD70note, for expression some tumour with types, varying the number intensities of samples in all was tumour too limited types, excludingto significantly Kaposi determine sarcoma CD70and NKTCL. positivity Of and note, predict for some the tumour benefit fromtypes, anti-CD70 the number therapy. of samples was too limitedIn addition, to significantly we only had determine limited data CD70 regarding positivity the clinicopathological and predict the benefit parameters from of anti patients,-CD70 whichtherapy. did not allow us to analyse links between CD70 expression and additional factors, such as previousIn addition, treatments wee onlyffects had on CD70limited expression data regarding or disease the stage.clinicopathological Nevertheless, weparameters were able of to patients, identify whichsolid tumours did not andallow us to analyse with links different between CD70 CD70 expression expression levels and and weadditional believe factors, that our such unified as previousmethod can treatments improve theeffects stratification on CD70 of expression patients for or immunotherapy. disease stage. Nevertheless, we were able to identifyCD70 solid overexpression tumours and on lymphomas tumour cells with is associated different withCD70 poor expression prognosis levels [21– 23and]. Co-expressionwe believe that of ourCD27 unified has also method been described can improve in several the stratification tumours of hematopoieticof patients for lineage, immunotherapy. implicating a possible role for the CD70-CD27CD70 overexpression axis in the regulation on tumour of cells tumour is associated cell proliferation with poor and prognosis survival [24 [21,25–23]]. CD27. Co-expression expression ofis almostCD27 has exclusively also been restricteddescribed toin immuneseveral tumours cells in of solid hematopoietic tumour types, lineage, in contrast implicating to CD70-CD27 a possible roleco-expression for the CD70 on- tumourCD27 axis cells in inthe hematopoietic regulation of malignanciestumour cell proliferation [20]. Nonetheless, and survival persistent [24,25] signalling. CD27 expressionof CD27 can is occuralmost in exclusively the tumour restricted micro-environment to immune cells through in solid its tumour expression type ons, in tumour-infiltrating contrast to CD70- CD27lymphocytes co-expression (TILs), particularly on tumour on cells immune in hematopoietic suppressive Tregs malignancies [26]. Therefore, [20]. Nonetheless, CD70-CD27 persistentsignalling signallingis one of the of immuneCD27 can escape occur mechanismsin the tumour that micro takes-environment place in the through tumour its microenvironment expression on tumour and it- infiltratingfacilitates the evasion of cancer (TILs), cells particularly from the onhost’s immune immune suppressive system [10 Tregs,27]. [26]. Therefore, CD70- CD27Targeting signalling the isCD70 one/CD27 of the pathway immune has emerged escape mechanisms as a potential thatnovel takes immunotherapeutic place in the strategy,tumour whichmicroenvironment is already being and explored it facilitates in several the evasion early of phase cancer clinical cells trialsfrom the testing host’s CD70- immune and CD27-targetingsystem [10,27]. antibodiesTargeting in various the CD70/CD27 malignancies pathway [17,28 ,29 has], but, emerged to date, as no a potentialstandardised novel method immunotherapeutic allowing for a strategy,comparison which of CD70 is already expression being in explored different intumour several types early has phase been published.clinical trials testing CD70- and CD27The-targeting detection ant ofibodies patients’ in various CD70 expressionmalignancies status [17,28,29] should, but be performed, to date, no in standardised a consistent mannermethod allowingto correctly for stratify a comparison and monitor of CD70 patients expression for anti-CD70 in different immunotherapy: tumour types has both been the published. technical staining procedureThe detection as well asof thepatients’ interpretation CD70 expression of the test status must should be reproducible. be performed Although in a consistent IHC remains manner the tomost correctly frequently stratify used and method monitor todetermine patients for CD70 anti expression,-CD70 immunotherapy dissimilarities: both are seenthe technical in the percentage staining procedureof CD70 positive as well cases as the within interpretation identical tumourof the test types must due be to reproducible. e.g., different Although antibodies IHC or cut-o remainsff values the (summarisedmost frequently in Jacobs used et method al. [20]). These to determin discrepanciese CD70 highlight expression, the need dissimilarities for one uniform are and seen validated in the percentagemethodology of forCD70 stratifying positive tumour cases within types foridentical anti-CD70 tumour immunotherapy. types due to e.g. In this, different study, noantibodies equipment or cutfor- computer-assistedoff values (summari imagesed in classification Jacobs et al. was [20] used,). These since discrepancies this is not yet highlight available inthe the need majority for one of uniform and validated methodology for stratifying tumour types for anti-CD70 immunotherapy. In this study, no equipment for computer-assisted image classification was used, since this is not yet available in the majority of clinical pathology laboratories. Therefore, positivity was assessed by visual annotation of pathologists, to closely resemble clinical practice. Rizzardi et al. have already

Cancers 2019, 11, 1611 9 of 13 clinical pathology laboratories. Therefore, positivity was assessed by visual annotation of pathologists, to closely resemble clinical practice. Rizzardi et al. have already demonstrated the strong similarities between computer-based and visual annotation, ensuring the reproducibility of our results with computer-assisted image classification in the future [30]. The variation in cut-off values to define CD70 expression also leads to inconsistent interpretations regarding CD70 expression profiles within identical tumour types. In our study, we compared CD70 expression in 496 solid tumour specimens while using one uniform CD70 IHC method, which allowed for us to compare CD70 expression within one specific neoplasm type as well as among different tumour types. The highest number of CD70 positive cells was detected in renal cell carcinoma. CD70 expression in renal cell carcinoma has already been described in several studies, although with varying positivity for CD70. While Law CL et al. [31] reported CD70 expression in 30% to 68% of clear cell and 40% of papillary renal cell carcinoma (cut-off value of 25%), Junker K. et al. [32] showed that all clear cell samples (41 in total) and only one papillary sample (19 in total) were positive for CD70 (cut-off unknown). Our study demonstrated that 79.5% of all renal cell carcinoma patients (39 in total) were positive for CD70. In the case of brain carcinoma, samples were examined for CD70 expression, whereby Ryan MC et al. [33] demonstrated CD70 positivity in 10% of the samples (59 in total) using 1% as cut-off. Contrarily, Wischhusen J et al. [12] found that 42% of glioblastoma patients (12 in total) were positive for CD70 (cut-off unknown). Similarly, our study showed that 44.4% of all brain carcinoma patients (nine in total) were CD70 positive. Surprisingly, while Ryan MC et al. [33] reported that only 2% of breast cancer patients were positive for CD70 (N = 204) while using a cut-off of 1%, our study showed that 47.6% breast cancer patients were CD70 positive (cut-off of 10%). We tested both primary and metastatic solid tumour specimens to determine CD70 expression. In tumour types, such as renal cell carcinoma, hepatocellular carcinoma, and head & neck carcinoma, tissues from both primary and metastatic sites were similarly positive for CD70. Similar findings of the matched primary and metastatic tissue samples of patients with renal cell carcinoma were observed by Law CL et al. [31] and Adam PJ et al. [34]. Interestingly, we showed that the number of CD70 positive cells appeared to be even higher in metastatic tissue as compared to primary specimens in pancreatic carcinoma and lung carcinoma. Although we acknowledge that, for certain tumour types, a limited number of samples were available and thus more samples would be needed to draw more precise conclusions. Furthermore, we also determined CD70 expression in 130 lymphoma samples and detected CD70 positive cells in all tested lymphoma subtypes, except for the NKTCL samples. Though it should be noted that a limited number of samples of NKTCL were analysed. On the contrary, Yoshino et al. [35] showed that CD70 was expressed in 33% of nasal NKTCL tissues (total of 21 Japanese patients), although their cut-off value was not stated. Interestingly, more than half of MCL patients appeared to be CD70 positive in our study. To date, immune checkpoint inhibitors, such as PD-1 and PD-L1, are mostly used with promising results in different types of cancer, although limited data are available on immune checkpoint inhibitors in MCL. Based on the few available studies, PDL-1/PD-1 expression is usually not detectable by IHC analysis on MCL cells [36–38]. Nevertheless, using a cut-off of 5%, we could detect small numbers of PDL-1/PDL1 positive cells in 18% and 35% of MCL samples, respectively, and much lower numbers of double positive samples were observed. Furthermore, in a phase I study with nivolumab, four patients with MCL were treated, but none of the MCL patients achieved an objective response in this study [39]. This underscores the need to search for alternative approaches to efficiently target MCL. In this regard, our study demonstrated the expression of CD70 and CD27 in the majority of MCL samples and co-expression in 38% of patients, suggesting that the CD70/CD27 axis is a more attractive target for therapy in case of MCL patients. Therefore, targeting the CD70/CD27 axis could be of great relevance in treating cancer types that currently lack effective treatment strategies. In this regard, a breakthrough was recently achieved Cancers 2019, 11, 1611 10 of 13 for cusatuzumab (ARGX-110), an anti-CD70 targeting antibody, in combination with Azacitidine in a Phase 1/2 trial in acute myeloid leukaemia (AML). The response rates were observed in 11 out of 12 AML patients, including 10 complete remissions, and median response duration was 6.9 months as of the data cut-off date on July 16, 2018 [18]. As we demonstrate high expression levels of CD70/CD27 in other tumour types using our uniform IHC method, these results underline the potential of anti-CD70 immunotherapy not only in AML, but also in other haematological malignancies and solid tumour types where CD70 is overexpressed.

4. Materials and Methods

4.1. Patient Selection and Tissue Specimen 620 formalin fixed paraffin embedded (FFPE) specimens were collected from different tumour types, of which 496 samples from solid tumours (Figure1) and 130 samples from haematological malignancies (Figure6). The Ethics committee of Antwerp University Hospital and the Ethics committee of Institute Gustave Roussy approved this study (EC13/47/469).

4.2. CD70 IHC 5 µm-thick sections were exposed to heat-induced antigen retrieval by incubation in a high pH buffer for 20 min. at 97 ◦C (PT-Link) (DAKO, Glostrup, Denmark). Subsequently, endogenous peroxidase activity was quenched by incubating the slides in peroxidase blocking buffer (DAK0) for 5 min. CD70 IHC was performed as previously described [26]. Shortly, anti-CD70 antibody (R&D systems, Minneapolis, MN, Canada; Clone 301731, 1:40) was incubated at room temperature for 20 min., followed by mouse enhanced polymer-based linker (30 min., DAKO) and visualised via a DAKO autostainer Link 48 instrument while using the Envision FLEX+ detection (DAKO) according to the instructions of the manufacturer. Tonsil tissue was included in each staining run and used as positive and negative tissue control (Figure2D). Furthermore, biopsies were checked for internal positive control (e.g., lymphocytes) and negative control (e.g., epithelial cells, muscular layer, neuronal cells). Two independent observers as well as one pathologist performed scoring. Validation of the most optimal immunohistochemical CD70 staining protocol was performed at the University Hospital of Antwerp (UZA) by manual analysis. FFPE material of human spleen and tonsil, renal cell carcinoma, as well as cell line material in agar block with known high (MJ), intermediate (MM1-S), and low (SU-DHL-6) CD70 copy number were used for the comparison of multiple commercial antibodies available for CD70 staining. Based on validation parameters repeatability (intra-run precision), reproducibility (inter-run precision), accuracy, specificity (using normal tissue and cancer-related tissue), and sensitivity (using agar cell block of cell lines with different levels of CD70 expression), CD27Ligand, clone 301731 (R&D Systems) showed the most specific staining. Further information on the evaluation criteria can be found in supplementary methods.

4.3. CD27, PD-1 and PD-L1 CD27, PD-1 and PD-L1 stainings were performed on a BenchMark ULTRA Ventana (Roche Diagnostics, Indianapolis, IN, USA). For CD27, the samples were pre-incubated in Ultra CC1 for 64 minutes, incubated with anti-CD27 monoclonal antibodies for 92 min. at 37 ◦C (Thermo Scientific, Rockford, IL, USA; Clone 137B4, 1:20), and detected by the ultraView plus detection kit and Amplification kit (Roche diagnostics). For PD-1 and PD-L1 detection, the samples were pre-incubated in CC1 solution (Roche Diagnostics) for 32 min. and 64 min. in case of PD-1 and PD-L1 detection, respectively, and then incubated with anti-PD-1 (Abcam, Cambridge, UK, clone NAT105, 60 min., RT) and anti-PD-L1 ( Technology, Danvers, MA, USA, Clone E1L3N®, 60 min., 37 ◦C). Dilutions of primary antibodies were adjusted, depending on the type of fixator used (in case of PD-1 1:750 for formal solution and 1:350 for AFA solution; in case of PD-L1 1:2000 for formal solution and 1:1000 for AFA solution). Control tissue Cancers 2019, 11, 1611 11 of 13 was used to determine the correct antibody dilutions for each fixator. For detection, OptiView kit was used and, in addition, OptiView Amplification kit was used in the case of PD-L1 detection.

4.4. Statistical Analysis Differences in expression levels of CD70 were assessed by Mann–Whitney U tests. In each comparison, one tumour type was compared to all others. Positivity (primary, metastasis, diagnosis, relapse) was assessed by Chi-square or Fisher’s Exact test. Correlations between CD70, CD27, PD-1, and PD-L1 were investigated by Pearson correlation coefficients and McNemar tests. Differences in the expression levels of these four markers were studies by Wilcoxon Signed Rank tests. All analyses were performed using SPSS version 25 and significance was reached if p < 0.05.

5. Conclusions We have shown strong CD70 overexpression in different solid and haematological malignancies. Moreover, we could demonstrate higher relevance of the CD70/CD27 axis for potential therapeutic application in MCL in comparison to the PD-L1/PD-1 axis. These findings highlight the potential of anti-CD70 therapies in a broad range of tumour types and, in doing so, implementing one standardised protocol to define CD70 overexpression to use it as a diagnostic tool.

Supplementary Materials: The following are available online at http://www.mdpi.com/2072-6694/11/10/1611/s1, Figure S1: Range of CD70 staining intensities in solid tumours, Figure S2: Staining examples in 2 MCL cases, Table S1: Characteristics of MCL patient cohort, Supplementary Methods: Evaluation criteria for selection of CD27Ligand antibody. Author Contributions: Formal analysis, T.F. and J.J.; Investigation, V.C.-C., A.D., J.B., V.R., K.Z. and J.J.; Data analysis, J.B., P.Z., J.J.; Methodology, V.C.-C., A.D., J.B., V.R., K.Z., E.S. and P.P.; Project administration, P.Z.; Supervision, P.Z., L.V.R., H.d.H. and J.J.; Visualization, T.F. and J.J.; Writing—original draft, T.F.; Writing—review & editing, T.F., V.C.-C., A.D., J.B., V.R., P.Z., L.V.R., H.d.H., K.Z., E.S., P.P. and J.J. Funding: This research was funded by the agency for Flanders Innovation & Entrepreneurship under Grant number 120-128 and Kom op tegen Kanker. Conflicts of Interest: The authors declare no conflict of interest. L.V.R., H.d.H., hold ownership interest (including patents) in argenx. V.R. is a consultant/advisory board member for Infinity Pharmaceuticals, Bristol-Myers Squibb, Eisai, PharmaMar, Gilead Sciences, NanoString Technologies, Incyte, BMS, MSD, Roche/Genentech, Epizyme, Astra Zeneca and Servier.

References

1. Zou, W.; Wolchok, J.D.; Chen, L. PD-L1 (B7-H1) and PD-1 pathway blockade for cancer therapy: Mechanisms, response, biomarkers and combinations. Sci. Transl. Med. 2016, 8, 328rv4. [CrossRef] 2. Cottrell, T.R.; Taube, J.M. PD-L1 and Emerging Biomarkers in Immune Checkpoint Blockade Therapy. Cancer J. 2018, 24, 41–46. [CrossRef] 3. Nolte, M.A.; van Olffen, R.W.; van Gisbergen, K.P.J.M.; van Lier, R.A.W. Timing and tuning of CD27-CD70 interactions: The impact of signal strength in setting the balance between adaptive responses and immunopathology. Immunol. Rev. 2009, 229, 216–231. [CrossRef] 4. Laouar, A.; Haridas, V.; Vargas, D.; Zhinan, X.; Chaplin, D.; van Lier, R.A.W.; Manjunath, N. CD70+ antigen-presenting cells control the proliferation and differentiation of T cells in the intestinal mucosa. Nat. Immunol. 2005, 6, 698–706. [CrossRef][PubMed] 5. Wajant, H. Therapeutic targeting of CD70 and CD27. Expert Opin. Ther. Targets 2016, 20, 959–973. [CrossRef] [PubMed] 6. Tesselaar, K.; Arens, R.; van Schijndel, G.M.W.; Baars, P.A.; van der Valk, M.A.; Borst, J.; van Oers, M.H.J.; van Lier, R.A.W. Lethal T cell immunodeficiency induced by chronic costimulation via CD27-CD70 interactions. Nat. Immunol. 2003, 4, 49–54. [CrossRef][PubMed] 7. Boursalian, T.E.; McEarchern, J.A.; Law, C.-L.; Grewal, I.S. Targeting CD70 for human therapeutic use. Adv. Exp. Med. Biol. 2009, 647, 108–119. Cancers 2019, 11, 1611 12 of 13

8. De Meulenaere, A.; Vermassen, T.; Aspeslagh, S.; Zwaenepoel, K.; Deron, P.; Duprez, F.; Ferdinande, L.; Rottey, S. CD70 Expression and Its Correlation with Clinicopathological Variables in Squamous Cell Carcinoma of the Head and Neck. Pathobiology 2016, 83, 327–333. [CrossRef] 9. Pahl, J.H.; Santos, S.J.; Kuijjer, M.L.; Boerman, G.H.; Sand, L.G.; Szuhai, K.; Cleton-Jansen, A.; Egeler, R.M.; Bovee, J.V.; Schilham, M.W.; et al. Expression of the immune regulation antigen CD70 in osteosarcoma. Cancer Cell Int. 2015, 15, 31. [CrossRef] 10. Claus, C.; Riether, C.; Schurch, C.; Matter, M.S.; Hilmenyuk, T.; Ochsenbein, A.F. CD27 signaling increases the frequency of regulatory T cells and promotes tumor growth. Cancer Res. 2012, 72, 3664–3676. [CrossRef] 11. Diegmann, J.; Junker, K.; Loncarevic, I.F.; Michel, S.; Schimmel, B.; von Eggeling, F. Immune escape for renal cell carcinoma: CD70 mediates apoptosis in lymphocytes. Neoplasia 2006, 8, 933–938. [CrossRef][PubMed] 12. Wischhusen, J.; Jung, G.; Radovanovic, I.; Beier, C.; Steinbach, J.P.; Rimner, A.; Huang, H.; Schulz, J.B.; Ohgaki, H.; Aguzzi, A.; et al. Identification of CD70-mediated apoptosis of immune effector cells as a novel immune escape pathway of human glioblastoma. Cancer Res. 2002, 62, 2592–2599. [PubMed] 13. Yang, Z.-Z.; Grote, D.M.; Xiu, B.; Ziesmer, S.C.; Price-Troska, T.L.; Hodge, L.S.; Yates, D.M.; Novak, A.J.; Ansell, S.M. TGF-beta upregulates CD70 expression and induces exhaustion of effector memory T cells in B-cell non-Hodgkin’s lymphoma. Leukemia 2014, 28, 1872–1884. [CrossRef][PubMed] 14. Riether, C.; Schurch, C.M.; Buhrer, E.D.; Hinterbrandner, M.; Huguenin, A.-L.; Hoepner, S.; Zlobec, I.; Pabst, T.; Radpour, R.; Ochsenbein, A.F. CD70/CD27 signaling promotes blast stemness and is a viable therapeutic target in acute myeloid leukemia. J. Exp. Med. 2017, 214, 359–380. [CrossRef] 15. Lens, S.M.; Drillenburg, P.; den Drijver, B.F.; van Schijndel, G.; Pals, S.T.; van Lier, R.A.; van Oers, M.H. Aberrant expression and reverse signalling of CD70 on malignant B cells. Br. J. Haematol. 1999, 106, 491–503. [CrossRef] 16. Riether, C.; Schurch, C.M.; Flury, C.; Hinterbrandner, M.; Druck, L.; Huguenin, A.-L.; Baerlocher, G.M.; Radpour, R.; Ochsenbein, A.F. Tyrosine kinase inhibitor-induced CD70 expression mediates drug resistance in leukemia stem cells by activating Wnt signaling. Sci. Transl. Med. 2015, 7, 298ra119. [CrossRef] 17. Aftimos, P.; Rolfo, C.; Rottey, S.; Offner, F.; Bron, D.; Maerevoet, M.; Soria, J.; Moshir, M.; Dreier, T.; Rompaey, L.V.; et al. Phase I Dose-Escalation Study of the Anti-CD70 Antibody ARGX-110 in Advanced Malignancies. Clin. Cancer Res. 2017, 23, 6411–6420. [CrossRef] 18. Ochsenbein, A.F.; Riether, C.; Bacher, U.; Müller, R.; Höpner, S.; Banz, Y.; Hinterbrandner, M.; Bargetzi, M.; Manz, M.G.; Van Rompaey, L.; et al. Argx-110 Targeting CD70, in Combination with Azacitidine, Shows Favorable Safety Profile and Promising Anti-Leukemia Activity in Newly Diagnosed AML Patients in an Ongoing Phase 1/2 Clinical Trial. Blood 2018, 132, 2680. [CrossRef] 19. Diegmann, J.; Junker, K.; Gerstmayer, B.; Bosio, A.; Hindermann, W.; Rosenhahn, J.; von Eggeling, F. Identification of CD70 as a diagnostic biomarker for clear cell renal cell carcinoma by gene expression profiling, real-time RT-PCR and immunohistochemistry. Eur. J. Cancer 2005, 41, 1794–1801. [CrossRef] 20. Jacobs, J.; Deschoolmeester, V.; Zwaenepoel, K.; Rolfo, C.; Silence, K.; Rottey, S.; Lardon, F.; Smits, E.; Pauwels, P. CD70: An emerging target in cancer immunotherapy. Pharmacol. Ther. 2015, 155, 1–10. [CrossRef] 21. Bertrand, P.; Maingonnat, C.; Penther, D.; Guney, S.; Ruminy, P.; Picquenot, J.M.; Mareschal, S.; Alcantara, M.; Bouzelfen, A.; Dubois, S.; et al. The costimulatory molecule CD70 is regulated by distinct molecular mechanisms and is associated with overall survival in diffuse large B-cell lymphoma. Genes. Chromosomes Cancer 2013, 52, 764–774. [CrossRef][PubMed] 22. Jilaveanu, L.B.; Sznol, J.; Aziz, S.A.; Duchen, D.; Kluger, H.M.; Camp, R.L. CD70 expression patterns in renal cell carcinoma. Hum. Pathol. 2012, 43, 1394–1399. [CrossRef][PubMed] 23. Petrau, C.; Cornic, M.; Bertrand, P.; Maingonnat, C.; Marchand, V.; Picquenot, J.-M.; Jardin, F.; Clatot, F. CD70: A Potential Target in Breast Cancer? J. Cancer 2014, 5, 761–764. [CrossRef][PubMed] 24. Goto, N.; Tsurumi, H.; Takemura, M.; Kanemura, N.; Kasahara, S.; Hara, T.; Yasuda, I.; Shimizu, M.; Yamada, T.; Sawada, M.; et al. Serum soluble CD27 level is associated with outcome in patients with diffuse large B-cell lymphoma treated with rituximab, cyclophosphamide, doxorubicin, vincristine and prednisolone. Leuk. Lymphoma 2012, 53, 1494–1500. [CrossRef] 25. Nilsson, A.; de Milito, A.; Mowafi, F.; Winberg, G.; Bjork, O.; Wolpert, E.Z.; Chiodi, F. Expression of CD27-CD70 on early B cell progenitors in the bone marrow: Implication for diagnosis and therapy of childhood ALL. Exp. Hematol. 2005, 33, 1500–1507. [CrossRef] Cancers 2019, 11, 1611 13 of 13

26. Jacobs, J.; Zwaenepoel, K.; Rolfo, C.; Van den Bossche, J.; Deben, C.; Silence, K.; Hermans, C.; Smits, E.; Van Schil, P.; Lardon, F.; et al. Unlocking the potential of CD70 as a novel immunotherapeutic target for non-small cell lung cancer. Oncotarget 2015, 6, 13462–13475. [CrossRef] 27. Rolfo, C.; Sortino, G.; Smits, E.; Passiglia, F.; Bronte, G.; Castiglia, M.; Russo, A.; Santos, E.S.; Janssens, A.; Pauwels, P.; et al. Immunotherapy: Is a minor god yet in the pantheon of treatments for lung cancer? Expert Rev. Anticancer Ther. 2014, 14, 1173–1187. [CrossRef] 28. Riether, C.; Schurch, C.; Ochsenbein, A.F. Modulating CD27 signaling to treat cancer. Oncoimmunology 2012, 1, 1604–1606. [CrossRef] 29. Silence, K.; Dreier, T.; Moshir, M.; Ulrichts, P.; Gabriels, S.M.E.; Saunders, M.; Wajant, H.; Brouckaert, P.; Huyghe, L.; Hauwermeiren, T.V.; et al. ARGX-110, a highly potent antibody targeting CD70, eliminates tumors via both enhanced ADCC and immune checkpoint blockade. MAbs 2014, 6, 523–532. [CrossRef] 30. Rizzardi, A.E.; Johnson, A.T.; Vogel, R.I.; Pambuccian, S.E.; Henriksen, J.; Skubitz, A.P.; Metzger, G.J.; Schmechel, S.C. Quantitative comparison of immunohistochemical staining measured by digital image analysis versus pathologist visual scoring. Diagn. Pathol. 2012, 7, 42. [CrossRef] 31. Law, C.-L.; Gordon, K.A.; Toki, B.E.; Yamane, A.K.; Hering, M.A.; Cerveny, C.G.; Petroziello, J.M.; Ryan, M.C.; Smith, L.; Simon, R.; et al. activation antigen CD70 expressed by renal cell carcinoma is a potential therapeutic target for anti-CD70 antibody-drug conjugates. Cancer Res. 2006, 66, 2328–2337. [CrossRef][PubMed] 32. Junker, K.; Hindermann, W.; von Eggeling, F.; Diegmann, J.; Haessler, K.; Schubert, J. CD70: A new tumor specific biomarker for renal cell carcinoma. J. Urol. 2005, 173, 2150–2153. [CrossRef][PubMed] 33. Ryan, M.C.; Kostner, H.; Gordon, K.A.; Duniho, S.; Sutherland, M.K.; Yu, C.; Kim, K.M.; Nesterova, A.; Anderson, M.; McEarchern, J.A.; et al. Targeting pancreatic and ovarian carcinomas using the auristatin-based anti-CD70 antibody-drug conjugate SGN-75. Br. J. Cancer 2010, 103, 676–684. [CrossRef][PubMed] 34. Adam, P.J.; Terrett, J.A.; Steers, G.; Stockwin, L.; Loader, J.A.; Fletcher, G.C.; Lu, L.-S.; Leach, B.I.; Mason, S.; Stamps, A.C.; et al. CD70 (TNFSF7) is expressed at high prevalence in renal cell carcinomas and is rapidly internalised on antibody binding. Br. J. Cancer 2006, 95, 298–306. [CrossRef][PubMed] 35. Yoshino, K.; Kishibe, K.; Nagato, T.; Ueda, S.; Komabayashi, Y.; Takahara, M.; Harabuchi, Y. Expression of CD70 in nasal natural killer/T cell lymphoma cell lines and patients; its role for cell proliferation through binding to soluble CD27. Br. J. Haematol. 2013, 160, 331–342. [CrossRef] 36. Xerri, L.; Chetaille, B.; Serriari, N.; Attias, C.; Guillaume, Y.; Arnoulet, C.; Olive, D. Programmed death 1 is a marker of angioimmunoblastic T-cell lymphoma and B-cell small lymphocytic lymphoma/chronic lymphocytic leukemia. Hum. Pathol. 2008, 39, 1050–1058. [CrossRef] 37. Andorsky, D.J.; Yamada, R.E.; Said, J.; Pinkus, G.S.; Betting, D.J.; Timmerman, J.M. Programmed death ligand 1 is expressed by non-hodgkin lymphomas and inhibits the activity of tumor-associated T cells. Clin. Cancer Res. 2011, 17, 4232–4244. [CrossRef] 38. Menter, T.; Bodmer-Haecki, A.; Dirnhofer, S.; Tzankov, A. Evaluation of the diagnostic and prognostic value of PDL1 expression in Hodgkin and B-cell lymphomas. Hum. Pathol. 2016, 54, 17–24. [CrossRef] 39. Lesokhin, A.M.; Ansell, S.M.; Armand, P.; Scott, E.C.; Halwani, A.; Gutierrez, M.; Millenson, M.M.; Cohen, A.D.; Schuster, S.J.; Lebovic, D.; et al. Nivolumab in Patients With Relapsed or Refractory Hematologic Malignancy: Preliminary Results of a Phase Ib Study. J. Clin. Oncol. 2016, 34, 2698–2704. [CrossRef]

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).