The dominant mechanism responsible for restenosis after angioplasty is believed to be the activation of medial smooth muscle cells (SMCs), leading to their proliferation, migration to the subintima, and further proliferation. To develop novel strategies that might inhibit or prevent restenosis, we previously used a chimeric toxin composed of transforming growth factor-at (which targets the epidermal growth factor receptor) and mutated Pseudomonas exotoxin to preferentially recognize and kill rapidly proliferating, versus quiescent, vascular SMCs. We have recently cloned and expressed a recombinant gene encoding Pseudomonas exotoxin with a mutated (nonfunctional) cell recognition domain fused with the ligand acidic fibroblast growth factor, termed aFGF-PE664GIUKDEL; thus, this recombinant toxin targets the fibroblast growth factor receptor. In the present study, we evaluated the relative effects of this chimeric toxin on quiescent versus rapidly proliferating vascular SMCs and also determined whether aFGFPE664G!UKDEL exerted different effects on SMCs versus endothelial cells. Rapidly proliferating SMCs (grown in 10% fetal bovine serum) were very sensitive to the cytotoxic effects of aFGF-PE664GIuKDEL, whereas cytotoxicity was significantly less when the SMCs were in a quiescent state (grown in medium supplemented with 0.5% fetal bovine serum). The chimeric toxin was also significantly less cytotoxic against endothelial cells. Competition studies using excess acidic fibroblast growth factor indicated that the cytotoxic effects are specifically mediated by the fibroblast growth factor receptor. Thus, the present studies suggest a potentially expanded role of recombinant toxin therapy in restenosis: multiple receptors can be targeted, and cytotoxic effects, at least in vitro, can be preferentially directed to rapidly proliferating vascular SMCs, with relative sparing of vascular endothelial cells. It will next be necessary to test this strategy for inhibiting restenosis in an in vivo model of vascular injury and SMC proliferation. (Circulation Research 1992;71:640-645)

Purpose: The immunotoxin BR96 sFv-PE40 is an effective antitumor agent against human breast and lung carcinoma xenografts in rodents. This study was designed to (a) determine the frequency with which canine carcinoma cells express Lewisy (Ley) antigen, thereby identifying canine carcinoma types suitable for the clinical evaluation of BR96 sFv-PE40, and (b) determine the safety and efficacy of BR96 sFv-PE40 in a canine model of spontaneously occurring cancers for investigation of targeted therapy. Experimental Design: Carcinoma tissue samples were obtained from client-owned dogs presented for medical care. The tissues were assessed for Ley antigen expression using immunohistochemical methods. Dogs with tumors expressing Ley antigen were offered enrollment in a clinical trial to receive twice-weekly infusions of 4 to 12 mg/m2 BR96 sFv-PE40. Clinical toxicity and response data were assessed at each treatment. Results: Twenty-two of 61 carcinomas evaluated were positive for Ley expression, including mammary, prostate, lung, and rectal carcinomas, and 12 dogs were enrolled in the clinical trial. The primary side effect was transient emesis. Partial responses or disease stabilization were noted in dogs with inflammatory mammary, bronchogenic, rectal, and tonsillar carcinoma. At least nine of the dogs developed antibodies to the immunotoxin after two to five infusions. Conclusions: Although development of anti-BR96 sFv-PE40 antibodies limited the long-term effectiveness of this immunotoxin in dogs, rapid clinical responses in several aggressive canine carcinomas suggest the immunotoxin has utility for treatment of certain naturally occurring tumors and that its clinical evaluation for treatment of similar human carcinomas is warranted.

The anti-CD20 antibody rituximab is useful in the treatment of certain B-cell malignancies, most notably non-Hodgkin’s lymphoma. Its efficacy has been increased when used in combination with chemotherapy, yet anti-CD20 monoclonal antibodies (mAbs) directly conjugated with drugs such as doxorubicin (Dox) have failed to deliver drug or to demonstrate antitumor activity. We have produced anti-CD20 antibody-drug conjugates that possess potent antitumor activity by using the anti-mitotic agent, monomethyl auristatin E (MMAE), linked via the lysosomally cleavable dipeptide, valine-citrulline (vc). Two anti-CD20 conjugates, rituximab-vcMMAE and 1F5-vcMMAE, were selectively cytotoxic against CD20+ B-lymphoma cell lines, with IC50 values ranging from 50 ng/mL to 1 μg/mL. Unlike rituximab, which showed diffuse surface localization, rituximab-vcMMAE capped and was internalized within 4 hours after binding to CD20+ B cells. Internalization of rituximab-vcMMAE was followed by rapid G2-M phase arrest and onset of apoptosis. Anti-CD20 antibody-drug conjugates prepared with Dox were internalized and localized as with rituximab-vcMMAE, yet these were not effective for drug delivery (IC50 > 50 μg/mL). Consistent with in vitro activity, rituximab-vcMMAE showed antitumor efficacy in xenograft models of CD20-positive lymphoma at doses where rituximab or rituximab-Dox conjugates were ineffective. These data indicate that anti-CD20–based antibody-drug conjugates are effective antitumor agents when prepared with a stable, enzyme-cleavable peptide linkage to highly potent cytotoxic agents such as MMAE.

Objectives: We previously presented preliminary data on pilot biodistributions of 64Cu‐labeled monoclonal antibodies (mAbs).1 The purpose of the present studies was to obtain comprehensive biodistributions using a two‐antibody model for direct comparison of an internalizing and a non‐internalizing mAb in the same animal model of cancer. These studies will allow evaluation of the efficacies of the two 64Cu‐labeled mAbs for radioimmunotherapy (RIT) and are part of a larger, ongoing project examining the unusual tumoricidal properties of internalized 64Cu.Methods: DOTA (1,4,7,10‐tetraazacyclododecane‐1,4,7,10‐tetraacetic acid) was conjugated to an internalizing antibody, cBR96, and a non‐internalizing antibody, cT84.66. Biodistributions of the 64Cu‐labeled conjugates in nude mice bearing LS174T human colorectal carcinoma xenografts were obtained at time points from 15 min to 48 h. Mouse tumor dosimetry was calculated using a Monte Carlo N‐Particle Transport Code.Results: The 64Cu‐DOTA‐cBR96 conjugate demonstrated rapid tumor uptake, reaching 20.2% injected dose per gram of tissue (% ID/g) at 3 h and peaking at 35.4% ID/g by 24 h. Tumor accumulation of 64Cu‐DOTA‐cT84.66 was more gradual, 8.19% ID/g at 3 h (p = 0.002), and reached 43.8% ID/g by 24 h, but maximum uptake was not statistically different from 64Cu‐DOTA‐cBR96 (p = 0.05). Liver, spleen, and kidney uptakes were statistically similar at 24 h for both conjugates (p = 0.05). Tumor absorbed radiation doses were estimated to be 1128 rad/mCi for 64Cu‐DOTA‐cBR96 and 1409 rad/mCi for 64Cu‐DOTA‐cT84.66.Conclusion: The biodistributions and tumor dosimetry of these two radiolabeled antibodies were sufficiently similar for direct comparison of the therapeutic efficacies of internalizing versus non‐internalizing 64Cu RIT agents in the same animal model of cancer. RIT studies using this two‐antibody model are currently in progress to test the hypothesis that internalization of 64Cu is necessary for enhanced cytotoxic properties at low tumor absorbed doses.

We describe the in vitro and in vivo properties of monoclonal antibody (mAb)-drug conjugates consisting of the potent synthetic dolastatin 10 analogs auristatin E (AE) and monomethylauristatin E (MMAE), linked to the chimeric mAbs cBR96 (specific to Lewis Y on carcinomas) and cAC10 (specific to CD30 on hematological malignancies). The linkers used for conjugate formation included an acid-labile hydrazone and protease-sensitive dipeptides, leading to uniformly substituted conjugates that efficiently released active drug in the lysosomes of antigen-positive (Ag+) tumor cells. The peptide-linked mAb-valine-citrulline-MMAE and mAb-phenylalanine-lysine-MMAE conjugates were much more stable in buffers and plasma than the conjugates of mAb and the hydrazone of 5-benzoylvaleric acid-AE ester (AEVB). As a result, the mAb-Val-Cit-MMAE conjugates exhibited greater in vitro specificity and lower in vivo toxicity than corresponding hydrazone conjugates. In vivo studies demonstrated that the peptide-linked conjugates induced regressions and cures of established tumor xenografts with therapeutic indices as high as 60-fold. These conjugates illustrate the importance of linker technology, drug potency and conjugation methodology in developing safe and efficacious mAb-drug conjugates for cancer therapy.

The chimeric monoclonal antibody cAC10, directed against CD30, induces growth arrest of CD30+ cell lines in vitro and has pronounced antitumor activity in severe combined immunodeficiency (SCID) mouse xenograft models of Hodgkin disease. We have significantly enhanced these activities by conjugating to cAC10 the cytotoxic agent monomethyl auristatin E (MMAE) to create the antibody-drug conjugate cAC10-vcMMAE. MMAE, a derivative of the cytotoxic tubulin modifier auristatin E, was covalently coupled to cAC10 through a valine-citrulline peptide linker. The drug was stably attached to the antibody, showing only a 2% release of MMAE following 10-day incubation in human plasma, but it was readily cleaved by lysosomal proteases after receptor-mediated internalization. Release of MMAE into the cytosol induced G2/M-phase growth arrest and cell death through the induction of apoptosis. In vitro, cAC10-vcMMAE was highly potent and selective against CD30+ tumor lines (IC50 less than 10 ng/mL) but was more than 300-fold less active on antigen-negative cells. In SCID mouse xenograft models of anaplastic large cell lymphoma or Hodgkin disease, cAC10-vcMMAE was efficacious at doses as low as 1 mg/kg. Mice treated at 30 mg/kg cAC10-vcMMAE showed no signs of toxicity. These data indicate that cAC10-vcMMAE may be a highly effective and selective therapy for the treatment of CD30+ neoplasias.

PURPOSE:Our purpose in the study was to establish the maximum tolerated dose and toxicity profile of SGN-10 (or BR96 sFv-PE40), a single-chain immunotoxin. SGN-10 is composed of the fused gene products encoding the translocating and ADP-ribosylating domains of Pseudomonas exotoxin (PE40) and the variable heavy (V(H)) and variable light (V(L)) regions of BR96 monoclonal antibody. This antibody is specific for a Lewis(Y) (Le(Y))-related carbohydrate antigen expressed on multiple carcinomas.EXPERIMENTAL DESIGN:A total of 46 patients with Le(Y)-positive metastatic carcinoma were enrolled in a Phase I dose-escalation study in cohorts of three to six patients who received SGN-10 at doses ranging from 0.024 to 0.962 mg/m(2), administered on days 1, 4, 8, and 11, followed by 2 weeks of rest and a second cycle of therapy. Pharmacokinetics and human antibody response to SGN-10 were also determined.RESULTS:The maximum tolerated dose of SGN-10 was 0.641 mg/m(2) with gastrointestinal dose-limiting toxicity. Pharmacokinetic studies performed in eight patients at the 0.641-mg/m(2) dose revealed a t([1/2]) of 2.5 +/- 0.3 h and a C(max) of 389 +/- 112 ng/ml. Pharmacodynamic analyses demonstrated a rapid clearance of the drug by day 11 associated with an antitoxin human antitoxin antibody (HATA) response in most patients. Signs consistent with a modest vascular leak syndrome, specifically, transient hypoalbuminemia, were observed in patients treated with doses of > or =0.384 mg/m(2). No complete or partial tumor responses were observed at an 8-week evaluation, although 31% of patients had stable disease.CONCLUSIONS:The maximal tolerated dose of SGN-10 given twice weekly for 2 weeks is 0.641 mg/m(2) with gastrointestinal dose-limiting toxicity. The immunogenicity of the toxin moiety limits the ability of SGN-10 to circulate by day 11 of therapy. Studies are ongoing to evaluate strategies to ameliorate toxicities and to inhibit the development of the anti-SGN-10 immune response.

CD40 is present on both normal and neoplastic B-lineage cells. CD40 stimulation of normal B cells has been shown to promote normal growth and differentiation, whereas aggressive histology B lymphomas are growth inhibited. The inhibition of neoplastic B-cell growth is believed to occur via activation-induced cell death in which stimuli that typically promote the growth of normal cells prevent the growth of their neoplastic counterparts. We show here that CD40 stimulation using either a soluble recombinant human CD40 ligand (srhCD40L) or anti-CD40 monoclonal antibody resulted in apoptosis of human Burkitt lymphoma cell lines. Additional studies examining the mechanism of CD40-mediated death revealed an increase in bax messenger RNA with a subsequent increase in Bax protein in the mitochondria of the treated cells. In vitro exposure of the cells to bax antisense oligonucleotides resulted in a significant decline in Bax protein levels and partial protection from CD40-mediated death, indicating that induction of Bax was at least one mechanism underlying this inhibitory effect of CD40 stimulation on lymphomas. When immunodeficient mice bearing Burkitt lymphoma were treated with srhCD40L, significant increases in survival were observed indicating a direct antitumor effect as a result of CD40 stimulation in vivo. Overall, these results demonstrate that CD40 ligation of aggressive histology B-lymphoma cells results in inhibition both in vitro and in vivo and thus may be of potential clinical use in their treatment.

The leukocyte activation marker CD30 is highly expressed on the Reed Sternberg cells of Hodgkin's disease (HD). On normal tissues, CD30 has a restricted expression profile limited to activated T cells, activated B cells, and activated natural killer cells. This expression profile makes CD30 an ideal target for monoclonal antibody (mAb)-based therapies of Hodgkin's disease. CD30 mAbs have been shown to be effective in in vitro and in vivo models of hematologic malignancies such as anaplastic large cell lymphoma, yet these mAb have not been efficacious in HD models. We have found that a mAb against CD30, AC10, was able to inhibit the growth of HD cell lines in vitro. To generate a more clinically relevant molecule, the variable regions from AC10 were cloned into an expression construct containing the human gamma1 heavy chain and kappa light chain constant regions. The resulting chimeric antibody, designated SGN-30, retained the binding and in vitro growth-inhibitory activities of the parental antibody. Treatment of HD cell lines with SGN-30 in vitro resulted in growth arrest in the G(1) phase of the cell cycle and DNA fragmentation consistent with apoptosis in the HD line L540cy. Severe combined immunodeficient mouse xenograft models of disseminated HD treated with SGN-30 produced significant increases in survival. Similarly, xenograft models of localized HD demonstrated dose-dependent reduction in tumor mass in response to SGN-30 therapy. SGN-30 is being developed for the treatment of patients who have HD that is refractory to initial treatment or who have relapsed and have limited therapeutic options.

The chimeric monoclonal antibody cBR96 conjugated to doxorubicin (cBR96-Dox) is selectively internalized by a wide variety of human carcinomas expressing an extended form of Lewis Y antigen (Le(y)). Endocytosis is followed by cleavage and release of free doxorubicin from the endocytic vesicles and subsequent cytotoxicity. Combination studies with standard anti-cancer agents, undertaken to further increase the potency of this targeted therapy, identified significant synergistic anti-tumor activity of cBR96-Dox and either of the taxanes paclitaxel or docetaxel. Treatment with cBR96-Dox 24 hr prior to paclitaxel resulted in a steady increase in the percentage of G(2) tumor cells and corresponding increase in sensitivity to taxanes. Cell cycle analysis indicated the cBR96-delivered doxorubicin was most effective against S-phase cells, yet cells exposed to even subtoxic levels progressed to and arrested in G(2), at a point of high sensitivity to the anti-tubulin agent paclitaxel. The synergy obtained by staged combination of cBR96-Dox and paclitaxel in vitro was reflected in significant anti-tumor efficacy in vivo against xenograft models of human lung and breast tumors that could not be achieved by either agent alone. The staged combination elicited significant or complete regressions of established human Le(y)-positive tumor xenografts using significantly reduced drug levels. Taken together, these data demonstrate a mechanistic approach to the selective elimination of Le(y)-positive tumors by using targeted doxorubicin followed by taxane treatment.