Growth Factor Receptor Blockade Moving From Laboratory to Clinic

PARIS, France--Overexpression of growth factor receptors correlateswith a poor prognosis in many malignancies, including breast,bladder, and lung cancer, and new evidence suggests that growthfactor receptors may be an important target for anticancer therapy,John Mendelsohn, MD, said at the Fifth International Congresson Anti-Cancer Chemotherapy.

The rationale for this novel therapeutic approach, said Dr. Mendelsohn,of Memorial Sloan-Kettering Cancer Center, also stems from theobservations that most cells require growth factors to proliferate,and that about one third of proto-oncogenes code for growth factors,their receptors, or biochemical steps in the signal transductionpathways activated by the receptors.

Moreover, he noted, transfection of growth factors into nonmalignantcells can cause these cells to behave in a malignant fashion.Thus, if the actions of growth factors on their receptors canbe interrupted therapeutically, it may be possible to block cellularproliferation.

Growth factors may induce cellular proliferation, either throughautocrine stimulation or through stimulation of adjacent cells,Dr. Mendelsohn told the conference audience. For example, he said,platelet-derived growth factor (PDGF) produced by breast cancercells stimulates fibroblast PDGF receptors, causing the fibroblaststo produce IGF (insulin-like growth factor), which then in turngoes back and autostimulates the breast cancer cells.

To illustrate autostimulation of cells, he described how the portionof the epidermal growth factor (EGF) receptor that exists outsidethe cell membrane can be stimulated by EGF or by transforminggrowth factor-alpha (TGF-alpha), both of which are often producedby the same cell. The stimulated receptor then activates its intrinsictyrosine kinase, leading to protein phosphorylation and, eventually,to cellular proliferation.

Dr. Mendelsohn and his colleagues have produced monoclonal antibodiesthat block the binding of EGF to its receptor, and thereby inhibitEGF- and TGF-alpha-induced activation of tyrosine kinase. "Thisis not immunotherapy--this is a drug that will block the enzymetyrosine kinase," he stressed, pointing out that even antibodyfragments that lack immune activity can still inhibit cellulargrowth.

Although very early administration of these monoclonal antibodiescan completely inhibit human tumor growth in nude mice, well-establishedtumors are more resistant to cure. However, Dr. Mendelsohn said,in combination with such cytotoxic agents as platinum, doxorubicin,or paclitaxel (Taxol), EGF receptor blockade can eliminate establishedtumors in xenograft models.

Thus, blocking the signal transduction pathway for the EGF receptoraugments three different kinds of chemotherapy, he said. (Seebelow for a possible explanation of this synergism.)

First Clinical Study

The first clinical study of a murine monoclonal antibody againstEGF revealed that a 30 nmol/L concentration of antibody (enoughto saturate the EGF receptors) produced no toxicity in patientswith squamous cell lung cancer.

Current trials are using multiple doses of chimerized human EGFantibody in combination with cisplatin (Platinol), doxorubicin,or paclitaxel, to determine whether the effects seen in cancercell lines can be duplicated in patients.

"It's also possible to think about using combinations ofdifferent antireceptor antibodies," Dr. Mendelsohn said."We have preliminary evidence that antibodies against theHER-2/neu receptor combined with antibodies against the EGF receptorhave an additive effect against xenografts, and we also are exploringantibodies against the transferrin receptor," he added.

HER-2/neu Receptor Blockade

Dr. Mendelsohn reported that receptor blockade with an antibodyagainst HER-2/neu has been tested in a Genentech-sponsored multi-institutionalphase II trial involving 43 heavily pretreated women with stageIV breast cancer whose tumors had demonstrated expression of HER-2/neu.

After a loading dose of 250 mg, followed by 10 weekly 100 mg doses,the patients were evaluated for responses. Disease stabilizationor partial responses were observed in a substantial number ofpatients, and a complete response was seen in one patient, whohas remained disease-free for a year on continuing antibody therapy.

"This study provides proof in concept that an antireceptoragent is a reasonable anticancer agent and can produce completeresponses in heavily pretreated patients," Dr. Mendelsohnsaid. A parallel study using this antibody in combination withcisplatin yielded higher response rates than are generally seenwith cisplatin in heavily pretreated women, he said.

Dual Hit Checkpoint Model May Explain Synergism Between Chemoand EGF Blockade

To explain the synergism between epidermal growth factor (EGF)receptor blockade and chemotherapy seen in animal studies (seestory above), Dr. Mendelsohn suggested a dual hit checkpoint model.

He reminded the audience that there are checkpoints in the cellcycle, where cells that have damaged DNA or that have been deprivedof essential nutrients or growth factors will stop to repair themselvesbefore going on to divide.

For example, he said, chemotherapy-induced damage activates checkpointarrests in the late G1 or G2 phases of the cell cycle, while growthfactor deprivation activates checkpoint arrests in the late G1phase.

"Treatments that activate these checkpoints through two unrelatedmechanisms may produce death in tumor cells," Dr. Mendelsohnsuggested.

He said that tumor cells receiving two different checkpoint signalsthat they cannot obey may self destruct through programmed celldeath (apoptosis). In contrast, normal cells, which obey checkpointsand "go into hibernation," may be less susceptible todeath from dual insults.

"If this logic is correct, it would suggest that modulationof any signal transduction pathway might synergize with cytotoxicchemotherapy," he proposed. "So, instead of increasingthe dose of chemotherapy, let's concurrently attack the cell'scheckpoints through a signal transduction pathway."