Welcome international healthcare professionals

This site is no longer supported and will not be updated with new content. You are welcome to browse and download all content already included in the site. Please note you will have to register your email address to access the site.

You are here

Three clinical-stage tumor targeting antibodies reveal differential expression of oncofetal fibronectin and tenascin-C isoforms in human lymphoma

Leukemia Research, 12, 33, pages 1718 - 1722


The antibody-based targeted delivery of bioactive molecules to tumor vasculature is an attractive avenue to concentrate therapeutic agents at cancer sites, while sparing normal organs. L19, F8 and F16 are clinical-stage human monoclonal antibodies, which selectively recognize splice isoforms of fibronectin and tenascin-C in the modified extracellular matrix of neoplastic lesions. Here, we report the first comparative immunohistochemical analysis of L19, F8 and F16 in human Hodgkin and non-Hodgkin lymphomas. F16 was found to strongly stain the majority of lymphomas but also specimens of nonspecific lymphadenitis. By contrast, L19 exhibited a better discrimination between tumoral and inflammatory processes, yet at the expense of a weaker staining of the majority of lymphoma specimens investigated. The staining patterns observed for F8 were intermediate between the ones observed for L19 and F16. This study provides a rationale basis for the clinical investigation of therapeutic derivatives of the three antibodies in lymphoma patients.

Keywords: Extra domain A of fibronectin, Extra domain B of fibronectin, Extra domain A1 of tenascin-C, Lymphoma, Therapeutic antibodies, Vascular targeting.

1. Introduction

One avenue towards the development of more selective anti-cancer drugs consists in the targeted delivery of bioactive molecules (drugs, cytokines, procoagulant factors, photosensitizers, radionuclides, etc.) to the tumor environment by means of binding molecules (e.g., human antibodies) specific to tumor-associated markers. In this context, the targeted delivery of therapeutic agents to newly formed blood vessels (“vascular targeting”) is particularly attractive, because of the dependence of tumors on new blood vessels to sustain growth and invasion, and because of the inherent accessibility of neovascular structures to intravenously injected therapeutic agents [1] .

In collaboration with Philogen SpA, our laboratory has developed human monoclonal antibodies, capable of selective targeting of neovascular structures in solid tumors and in certain other angiogenesis-related diseases. The most promising antibodies reported so far in the literature include L19 (specific to the alternatively spliced EDB domain of fibronectin) [2] , F8 (specific to the EDA domain of fibronectin) [3] and F16 (specific to domain A1 of tenascin-C) [4] . Therapeutic derivatives of these antibodies, which are currently investigated in Phase I and Phase II clinical trials, include fusions to cytokines (e.g., IL-2, IL-12, IL-10, TNF) and iodine-131 conjugates for radioimmunotherapy.

While L19, F8 and F16 were originally developed for vascular tumor targeting strategies aimed at the selective pharmacodelivery of therapeutic agents to solid tumor lesions [1] , we have recently reported that the L19 antibody can selectively stain neovascular structures in human lymphomas [5] and [6]. Indeed, L19-131I has been shown to selectively localize on neoplastic lesions in Hodgkin lymphoma, leading to objective responses in heavily pre-treated patients [6] . Furthermore, the immunocytokine L19-IL2 (a fusion between L19 in scFv format and the pro-inflammatory cytokine IL-2) was shown to potentiate the therapeutic action of the anti-CD20 antibody Rituximab, leading to complete eradications in three models of human Non-Hodgkin lymphoma (NHL) implanted in immunocompromised mice [5] .

A comparative analysis of expression patterns for the extra domains EDB and EDA of fibronectin and A1 of tenascin-C has not been reported so far, but would be clinically relevant, considering the fact that L19, F8 and F16 are currently investigated in clinical trials and could potentially be considered for vascular targeting applications in patients with lymphoma. On one hand, the exquisite radiosensitivity of hematological malignancies makes these cancer types particularly suitable for therapeutic approaches based on the targeted delivery of radionuclides. On the other hand, the antibody-based targeted delivery of cytokines may synergize not only with therapeutic IgG products [5] , but also with certain chemotherapeutic agents. In this communication, we report the results of a comparative immunohistochemical analysis of L19, F8 and F16 in 63 frozen specimens of human Hodgkin and Non-Hodgkin lymphoma and chronic lymphadenitis.

2. Materials and methods

2.1. Tissues

The study group comprised 63 patients with Non-Hodgkin lymphomas, Hodgkin lymphomas, or chronic nonspecific lymphadenitis. All cases were retrieved from consultation files of the Institute of Pathology, University Hospital Schleswig-Holstein, Kiel, Germany. Histopathological diagnosis was made on the basis of conventional histological and immunohistological examination according to the criteria of the World Health Organization (WHO) classification. Non-Hodgkin lymphoma entities were divided into indolent and aggressive lymphomas according to the Physician Data Query Modification of the Revised European American Lymphoma Classification of Lymphoproliferative Diseases.

2.2. Immunohistochemical staining

The L19 antibody, specific to the extra domain B (EDB) of fibronectin, the F8 antibody, specific to the extra domain A (EDA) of fibronectin, and the F16 antibody, specific to the extra domain A1 of tenascin-C, have been described before [2], [3], and [4]. All antibodies were isolated from ETH2 antibody libraries, affinity matured, had dissociation constants in the low nanomolar range and exhibited kinetic dissociation constants koff towards the respective antigens <10−2 s−1 in real-time interaction analysis experiments on a BIAcore 3000 instrument (GE Healthcare, Otelfingen, Switzerland) [2], [3], and [4]. For immunohistochemistry, biotinylated antibodies were used in small immunoprotein format (SIP) under identical conditions (2 μg/ml for all three antibodies). Aliquots were prepared from a single batch of antibodies, frozen and used only once to ensure reproducibility of immunohistochemical stainings.

Frozen tissue samples were stored at −80 °C. Sections of 6 μm thickness were fixed in chilled acetone, rehydrated in TBS buffer (50 mM Tris, 100 mM NaCl, 0.001% Aprotinin, pH 7.4) and blocked with 20% fetal calf serum in TBS. Primary antibodies were added onto the sections in a final concentration of 2 μg/mL in 3% bovine serum albumin (BSA)/TBS solution and incubated for 1 h. After washing in TBS, bound antibodies were detected with streptavidin-biotinylated alkaline phosphatase complex (Biospa, Milan, Italy) in TBS 3% BSA + 2 mM MgCl2. The Fast Red substrate (Sigma) was used for detection of phosphatase activity. Sections were counterstained with Gill's hematoxylin No.2 (Sigma) and mounted with Glycergel mounting medium (Dako, Glostrup, Denmark).

2.3. Semiquantitative evaluation of EDB-fibronectin, EDA-fibronectin and tenascin-C (domain A1) expression

Immunostainings of the extra domains B and A of fibronectin and of the domain A1 of tenascin-C, as revealed by the antibodies L19, F8 and F16, respectively, were simultaneously assessed by two independent experienced investigators using light microscopy (Axiovert S100, Zeiss, Feldbach, Switzerland). Both investigators were blinded to the histopathological diagnosis when performing the evaluation of protein expression. Staining was scored using the four different levels “no staining”, “weak”, “moderate” and “strong” (0–3), according to a “training” table with examples of each staining level as reference ( Supplementary Fig. 1 ) [7] . Discrepantly scored slides (21 out of 252 evaluations [8.3%]) were re-evaluated together to reach a consensus score. Furthermore, the pattern of antigen expression was evaluated and scored as “vascular” (antigen expression predominantly restricted to blood vessels) or “stromal” (diffuse antigen expression in the stroma in addition to vascular staining). Controls which were performed in each immunohistochemical staining procedure by omitting the primary antibodies were consistently negative (data not shown).

3. Results

The expression of alternatively spliced isoforms of fibronectin and tenascin-C was analyzed by immunohistochemistry in 63 freshly frozen specimens, including indolent and aggressive Non-Hodgkin lymphoma (NHL), Hodgkin lymphoma (HL) and nonspecific lymphadenitis (LA) cases. Pathological characteristics are summarized in Table 1 . The three antibodies L19, F8 and F16 have comparably high affinities towards their cognate antigen and were used at identical concentrations, as previously reported for the immunohistochemical analysis of solid tumors [7] .

Table 1 Pathological characteristics.

Subtype Number
Indolent NHL 14
 Follicular lymphoma, grade 1 + 2 7
 Chronic lymphocytic leukemia 7
Aggressive NHL 25
 Follicular lymphoma, grade 3 1
 Diffuse large B cell lymphoma 8
 Burkitt lymphoma 8
 Mantle cell lymphoma 8
Hodgkin lymphoma 16
 Classical 8
 Nodular sclerosis 2
 Mixed cellularity 5
 Lymphocyte depletion 1
 Lymphocyte rich 0
 Nodular lymphocyte-predominant 8
Chronic unspecific lymphadenitis 8

Fig. 1 presents a representative collection of immunohistochemical results, outlining staining patterns and intensities observed with the three antibodies in the various types of specimens (three illustrative patients for each lymphoma subtype are presented). The same antibody could exhibit a strictly vascular staining in some tumors and a diffuse stromal reaction in others. On the other hand, within the same tumor, one antibody could show a predominantly vascular pattern, while others (in most cases F16) displayed a diffuse stromal reactivity in addition to vascular staining. Individual tumor cells did not show a preferential staining, even though it is known that tumor cells contribute to the biosynthesis of alternatively spliced fibronectin and tenascin-C isoforms, in addition to endothelial cells and fibroblasts of the tumor stroma.


Fig. 1 Expression of EDB-fibronectin, EDA-fibronectin and the tenascin-C splicing domain A1, detected with the antibodies L19, F8 and F16, respectively, in Hodgkin and Non-Hodgkin lymphomas. Three representative panels of immunostainings with L19, F8 and F16 on sections derived from the same specimen are presented for each lymphoma subtype. Note the different staining patterns observed with the three antibodies, ranging from strictly vascular in some cases to diffuse stromal in others. DLBCL, diffuse large B cell lymphoma; BL, Burkitt lymphoma; MCL, mantle cell lymphoma; FL, follicular lymphoma; CLL, chronic lymphocytic leukemia; cHL, classical Hodgkin lymphoma; NLPHL, nodular lymphocyte-predominant Hodgkin lymphoma. Scale bar, 100 μm.

Fig. 2 summarizes the results of the immunohistochemical analysis for the different antibodies. F16 strongly stained (levels 2–3) all HL and LA and virtually all NHL specimens (95%). By contrast, L19 did not exhibit any strong staining in LA specimens (6/8 being completely negative in the analysis). However, only 28% and 19% of NHL and HL specimens were strongly stained (levels 2–3) with this antibody. F8 displayed a staining performance intermediate between L19 and F16 (levels 2–3 in 64% and 69% of NHL and HL cases, respectively) ( Fig. 2 A). In general, aggressive NHLs were stronger stained with all three antibodies than indolent subtypes ( Fig. 2 B). Interestingly, F16 efficiently reacted with both indolent and aggressive NHL subtypes with high staining scores in the majority of the cases (86% and 100%), while F8 provided an excellent molecular discrimination between these two classes (levels 2–3: 14% vs. 92%) ( Fig. 2 B). When analyzing individual NHL and HL entities ( Fig. 2 C), F16 strongly reacted with all investigated specimens among the different lymphoma types, exception made for follicular lymphomas (FL) where a strong staining was observed in 6/8 samples (75%). L19 preferentially displayed level 2–3 staining towards diffuse large B cell lymphoma (DLBCL) and Burkitt lymphoma (BL) (50% and 75%, respectively). Again, F8 exhibited an intermediate behavior between the other two antibodies F16 and L19 ( Fig. 2 C).


Fig. 2 Analysis of immunostaining results obtained with the monoclonal antibodies L19, F8, and F16. For each antibody, values indicate percentage of cases with the indicated expression scores. The figure presents a comparison of Hodgkin lymphoma (HL), Non-Hodgkin lymphoma (NHL) and nonspecific lymphadenitis (LA) tissues (A), a comparison of indolent and aggressive NHL subtypes (B), the results for the individual lymphoma entities (C), and a comparison of antibody staining patterns in different lymphoma subtypes (D). DLBCL, diffuse large B cell lymphoma; BL, Burkitt lymphoma; MCL, mantle cell lymphoma; FL, follicular lymphoma; CLL, chronic lymphocytic leukemia; cHL, classical Hodgkin lymphoma; NLPHL, nodular lymphocyte-predominant Hodgkin lymphoma. Staining intensities: 0 = no staining, 1 = weak, 2 = moderate, 3 = strong (see also Section 2 ).

Finally, we analyzed the pattern of antigen expression. In general, F16 displayed a more diffuse staining over wide areas of the lymphoma sections (diffuse “stromal” vs. predominantly “vascular” expression pattern in 80% vs. 20% of the cases, respectively). In contrast, L19 exhibited a staining pattern that was restricted to vascular structures in the majority of the lymphoma cases (81%). Overall, F8 reacted equally frequently in a “vascular” or “stromal” fashion (50% and 50%, respectively). The tendency of F16 and F8 towards a stromal reaction was especially evident in aggressive NHLs and HLs (F16: “stromal” pattern in indolent NHLs, aggressive NHLs and HLs in 57%, 84% and 94%, respectively; F8: 27%, 52% and 63%, respectively; L19: 22%, 20% and 15%, respectively) ( Fig. 2 D).

4. Discussion

Alternatively spliced isoforms of fibronectin and tenascin-C have long been recognized as well-characterized markers of angiogenesis and good-quality targets for the antibody-based delivery of bioactive agents to solid tumors [1], [2], [3], and [4]. In the healthy adult, these oncofetal isoforms are virtually undetectable in normal organs (exception made for the endometrium in the proliferative phase, the placenta and some vessels in the ovaries), but are often strongly expressed in conditions of tissue remodeling, often with a prominent vascular pattern of staining [1], [2], [3], and [4]. Until now, a direct comparative immunohistochemical evaluation of oncofetal fibronectin and tenascin-C expression in human lymphomas has not been reported.

The three clinical-stage monoclonal antibodies L19 (specific to EDB fibronectin), F8 (specific to EDA fibronectin) and F16 (specific to the domain A1 of tenascin-C) displayed a clear ability to stain the stroma of human Hodgkin and Non-Hodgkin lymphomas, with striking differences both in terms of staining patterns and staining intensities. In general, F16 exhibited the strongest immunohistochemical reactions with a diffuse stromal pattern of staining in lymphoma tissue, yet at the expense of not discriminating between cancer lesions and inflammatory conditions. This is in line with our observations that spliced isoforms of tenascin-C are expressed during tissue remodeling events in other chronic inflammatory conditions, such as rheumatoid arthritis or artherosclerosis (Kathrin Schwager and Marta Pedretti, manuscripts submitted). L19 was the most specific immunohistochemical reagent towards cancer-associated angiogenesis exhibiting a predominantly vascular expression pattern, but did not intensely stain all lymphoma specimens. The performance of F8 was intermediate between the ones of L19 and F16, yielding an excellent molecular discrimination between indolent and aggressive NHL subtypes ( Fig. 2 ).

All three antibodies could potentially be used in vivo for vascular targeted pharmacodelivery applications. Imaging and radioimmunotherapy clinical trials are on-going with L19 and F16, while clinical studies with radioiodinated derivatives of F8 are scheduled to begin by the end of 2009. A comparative analysis of immunoscintigraphic results obtained with the three radioiodinated antibodies will confirm whether the differential staining patterns observed in lymphoma patients are reflected in the antibody performance in nuclear medicine investigations and help to resolve an elementary question in field of antibody-based tumor targeting, i.e. whether the absolute amount of antigen expression at the tumor site correlates with in vivo targeting efficacy. Indeed, it is entirely conceivable that stromal antigens expressed in long distance to the nearest blood vessel do not efficiently contribute to the overall targeting performance. The immunohistochemical results reported in this study suggest that comparative dosimetry data, derived from immunoscintigraphic [6] or immuno-PET [8] investigations could provide a rational basis for the choice of the antibody to be administered to individual lymphoma patients.

Conflict of interest

Dario Neri is shareholder of Philogen ( www.philogen.com ), a company that has licensed the L19, F8 and F16 antibodies from ETH Zürich.


Financial contributions from the Swiss National Science Foundation, the ETH Zürich, the European Union (ADAMANT and IMMUNO-PDT Projects), the Swiss Cancer League, the SwissBridge Foundation, the Stammbach Foundation are gratefully acknowledged. C.S. receives a bursary from the Deutsche Krebshilfe.

Contributions. C.S. designed and performed experiments, analyzed the results and wrote the manuscript; A.W. performed experiments; M.P. analyzed results; M.S. and W.K. provided tissue specimens, contributed to experimental design and reviewed the manuscript; D.N. proposed, designed and supervised the project, wrote and revised the manuscript.

Appendix A. Supplementary data



  • [1] C. Schliemann, D. Neri. Antibody-based targeting of the tumor vasculature. Biochim Biophys Acta. 2007;1776:175-192 Crossref.
  • [2] A. Pini, F. Viti, A. Santucci, B. Carnemolla, L. Zardi, P. Neri, et al. Design and use of a phage display library. Human antibodies with subnanomolar affinity against a marker of angiogenesis eluted from a two-dimensional gel. J Biol Chem. 1998;273:21769-21776 Crossref.
  • [3] A. Villa, E. Trachsel, M. Kaspar, C. Schliemann, R. Sommavilla, J.N. Rybak, et al. A high-affinity human monoclonal antibody specific to the alternatively spliced EDA domain of fibronectin efficiently targets tumor neo-vasculature in vivo. Int J Cancer. 2008;122:2405-2413 Crossref.
  • [4] S.S. Brack, M. Silacci, M. Birchler, D. Neri. Tumor-targeting properties of novel antibodies specific to the large isoform of tenascin-C. Clin Cancer Res. 2006;12:3200-3208 Crossref.
  • [5] C. Schliemann, A. Palumbo, K. Zuberbuhler, A. Villa, M. Kaspar, E. Trachsel, et al. Complete eradication of human B-cell lymphoma xenografts using rituximab in combination with the immunocytokine L19-IL2. Blood. 2009;113:2275-2283 Crossref.
  • [6] S. Sauer, P.A. Erba, M. Petrini, A. Menrad, L. Giovannoni, C. Grana, et al. Expression of the oncofetal ED-B-containing fibronectin isoform in hematologic tumors enables ED-B-targeted 131I-L19SIP radioimmunotherapy in Hodgkin lymphoma patients. Blood. 2009;113:2265-2274 Crossref.
  • [7] M. Pedretti, A. Soltermann, S. Arni, W. Weder, D. Neri, S. Hillinger. Comparative immunohistochemistry of L19 and F16 in non-small cell lung cancer and mesothelioma: two human antibodies investigated in clinical trials in patients with cancer. Lung Cancer. 2009;64:28-33 Crossref.
  • [8] B.M. Tijink, L.R. Perk, M. Budde, M. Stigter-van Walsum, G.W. Visser, R.W. Kloet, et al. (124)I-L19-SIP for immuno-PET imaging of tumour vasculature and guidance of (131)I-L19-SIP radioimmunotherapy. Eur J Nucl Med Mol Imag. 2009;36:1235-1244 Crossref.


a Institute of Pharmaceutical Sciences, ETH Zürich, Wolfgang-Pauli-Strasse 10, 8093 Zürich, Switzerland

b Institute of Pathology, Section Hematopathology and Lymph Node Registry, University Hospital Schleswig-Holstein, Arnold-Heller-Strasse 3, 24105 Kiel, Germany

lowast Corresponding author at: Institute of Pharmaceutical Sciences, Department of Chemistry and Applied Biosciences, ETH Zürich, Wolfgang-Pauli-Strasse 10, 8093 Zürich, Switzerland. Tel.: +41 44 6337401; fax: +41 44 6331358.