Recurrent respiratory papillomatosis (RRP) is a mucosal disease of the airway, caused primarily by human papillomaviruses (HPVs) -6 and -11. RRP is characterized by growth of multiple papillomas that recur following surgical removal. Disease can last for the life of the patient or can be interspersed by short or long periods of remission. Recurrence rates vary between patients, with the worst cases requiring surgery as frequently as every three weeks to maintain an airway. Fortunately, the prevalence of the disease is low, estimated to be approximately 3/100,000 in the United States. The papillomas are primarily located in the larynx, but approximately 17 percent of patients will have tracheal disease and 5 percent will have papillomas of the bronchus and lung parenchyma. We are conducting a number of studies to better understand RRP, and to develop better therapies for the disease.
Defects in control of growth and differentiation
We have found that papilloma cells are characterized by a proliferative rate that is no higher than uninfected cells. Rather, they have a defect in terminal differentiation that results in an accumulation of cells that have left the cell cycle but have minimal expression of normal differentiation-specific keratins and filaggrin. The cells are also relatively resistant to apoptosis. Studies are continuing to determine the molecular mechanism(s) of these phenotypes.
Altered signal transduction and expression of COX-2
We have found that papilloma cells and respiratory papilloma tissues express high levels of the EGF receptor (EGFR), and constitutive activation of the EGFR causes the defect in differentiation. Papilloma cells also have alterations in several signaling pathway intermediates linked to the EGFR. These include activation of PI 3-kinase, overexpression of Rac1, activation of Pak1 and Pak2, activation of p50 NF-κB, and activation of p38 MAP kinase.
Activation of the signal transduction intermediates result in constitutive expression of the enzyme COX-2 and its product PGE2. Inhibition of COX-2 or some of the upstream intermediates in papilloma cells causes slowed proliferation and increases spontaneous apoptosis. We are currently asking whether COX-2 and PGE2 contribute to the defect in differentiation, and whether PGE2 contributes to constitutive activation of the signaling intermediates through a positive feed-back loop.
Clinical trial of celecoxib for treatment of RRP
Celecoxib is a selective COX-2 inhibitor. Based on the finding that celecoxib inhibited papilloma cell growth and enhanced apoptosis, and a literature showing that these potent anti-inflammatory drugs were effective in certain cancers, we conducted a preliminary clinical treatment study on three patients with severe RRP. The results were striking – the papillomas stopped recurring during the one year of treatment and improvement was maintained after the treatment was discontinued.
We are now conducting a multi-centered double-blind clinical trial testing the benefits of celecoxib as an adjunct to surgery.
Persistence of latent HPV infection
Patients with RRP (or in remission) have widespread latent HPV infection in clinically normal tissues of the larynx, trachea and bronchi, with life-long persistence of viral DNA but essentially no expression of viral RNA and no clinical or histologic evidence of disease. We propose that recurrent papillomas are due to repeated activation of this latent infection, rather than re-infection or “seeding” of the virus. The mechanism of activation is not known, but trauma or irritation facilitate or enhance activation in model systems. We are currently asking whether induction of COX-2 during transient inflammation or irritation mediates activation of latent HPV DNA.
Altered host immune response to HPV
We are collaborating with Dr. Vincent Bonagura and his laboratory in their studies to determine why patients with RRP don’t generate an effective immune response to prevent recurrent disease. We are trying to understand how the ubiquitous human papillomavirus can remain latent, or dormant, in most people but cause respiratory papillomas in a few. This question has broad implications beyond respiratory papillomatosis, since latent HPV infections are present in the airway, skin and genital tract of many people and may be a source of subsequent skin warts, genital warts, and cervical and airway cancers.
We have found that patients with RRP have a bias toward a TH2-like response to HPV proteins, have elevated levels of Treg cells in the papillomas, and show altered expression of a number of innate and adaptive immune response genes. These altered responses may be, at least in part, genetic. Patients have a skewed distribution of HLA Class II alleles, and may lack some activating Kir gene alleles. They may also express higher levels of COX-2 in their airway tissues, which could bias the local immune response toward expression of TH2-like cytokines and chemokines, reduce clearance of active infection, and make the patients more likely to have the disease.
Tumor angiogenesis and inhibition
Papilloma tissues are highly vascularized, and express VEGF. In this respect, they are similar to many cancers. The VEGF may be induced by PGE2 expressed in the papillomas. We are collaborating with Dr. Sam Soffer in his studies of angiogenesis and the molecular basis for the use of drugs to inhibit angiogenesis. Current studies focus on use of rapamycin in a sarcoma model. This drug, at low doses, suppresses tumor growth. It may act by inducing apoptosis of vascular endothelial cells in the tumor and possibly by inhibiting expression of COX-2 by the sarcoma cells.
The principal investigator for infectious disease research is Bettie Steinberg, PhD.