Innovations in understanding the biology of cervical cancer

Authors

  • Judith K. Wolf M.D.,

    Corresponding author
    1. Department of Gynecologic Oncology, The University of Texas M. D. Anderson Cancer Center, Houston, Texas
    • Department of Gynecologic Oncology, The University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Boulevard, Unit 440, Houston, TX 77030
    Search for more papers by this author
  • Eduardo L. Franco Ph.D., M.P.H.,

    1. Department of Epidemiology and Biostatistics, Faculty of Medicine, McGill University, Montreal, Quebec, Canada
    2. Department of Oncology, Faculty of Medicine, McGill University, Montreal, Quebec, Canada
    Search for more papers by this author
  • Jeffery M. Arbeit M.D.,

    1. Cancer Genetics Program, Comprehensive Cancer Center, University of California–San Francisco, San Francisco, California
    Search for more papers by this author
  • Kenneth R. Shroyer M.D., Ph.D.,

    1. Department of Pathology, School of Medicine, University of Colorado Health Sciences Center, Denver, Colorado
    Search for more papers by this author
  • Tzyy-Choou Wu M.D., Ph.D.,

    1. Department of Pathology, The Johns Hopkins Medical Institutions, Baltimore, Maryland
    2. Department of Oncology, The Johns Hopkins Medical Institutions, Baltimore, Maryland
    3. Department of Obstetrics and Gynecology, The Johns Hopkins Medical Institutions, Baltimore, Maryland
    4. Department of Molecular Microbiology and Immunology, The Johns Hopkins Medical Institutions, Baltimore, Maryland
    Search for more papers by this author
  • Carolyn D. Runowicz M.D.,

    1. St. Luke's–Roosevelt Hospital Center, Women's Health Service Line of Continuum Heath Partners, Inc., New York, New York
    Search for more papers by this author
  • Guillermo Tortolero-Luna M.D., Ph.D.,

    1. Department of Gynecologic Oncology, The University of Texas M. D. Anderson Cancer Center, Houston, Texas
    Search for more papers by this author
  • Rolando Herrero M.D.,

    1. Proyecto Epidemiologico Guanacaste, Santa Ana, Costa Rica
    Search for more papers by this author
  • Christopher P. Crum M.D.

    1. Department of Pathology, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts
    Search for more papers by this author

Abstract

Revelation of the connection between the human papillomavirus (HPV) and cervical neoplasia and invasive cervical cancer is prompting new investigations to expand that understanding and promote vaccines, gene therapy, and other interventions. At the Second International Conference on Cervical Cancer (Houston, TX, April 11–14, 2002), laboratory and clinical researchers reported advances in new studies meant to increase understanding of the natural history of HPV and cervical intraepithelial neoplasia, to evaluate new cervical cancer screening techniques, and to promote new therapies. Using K14-HPV type 16 transgenic mice, researchers are investigating the effects of estrogen on cervical cancer carcinogenesis, and results are lending support to epidemiological theories showing a difference in HPV infection rates and the development of cervical lesions in women using oral contraceptives. Other work involves investigating genes that are up-regulated by HPV infection and the role of the p53 homologue, p63, in cervical neoplasia evolution. Telomerase also is under investigation as a biomarker in high-risk populations. Gene therapy that replaced p53 in cervical cancer cell lines in vitro and a nude mouse model inhibited cell and tumor growth, confirming previous findings in squamous epithelial carcinomas of the head and neck. Furthermore, research in intracellular targeting of antigens to subcellular locations shows promise for treating cervical cancer preclinically. Identification of molecular changes in cervical cancer and knowledge about the importance of HPV infection in cervical cancer can lead to new therapies to treat existing cervical cancer and, in the long term, prevent the disease. Cancer 2003;98(9 Suppl):2064–2069. © 2003 American Cancer Society.

With the relatively recent understanding of the causal relation between the human papillomavirus (HPV) and cervical neoplasia, a new paradigm of research in the prevention, detection, and treatment of cervical intraepithelial neoplasia, as well as invasive cervical cancer, was begun.1–3 Research now spans the spectrum from understanding the epidemiology of HPV infection, including its nat

ural history, to understanding the molecular biology of cervical cancer. Since the conclusion of the Second Annual International Conference on Cervical Cancer, Koutsky et al.4 reported the use of an HPV type 16 (HPV16) vaccine that successfully reduced the incidence of HPV16 infection and HPV16-related intraepithelial neoplasia, and others may someday describe methods for correcting alterations of genes in cervical cancer and reversing its malignant process.

To understand the research in all of these areas, an understanding of HPV is imperative.5 The HPVs are a family of DNA viruses with over 150 genotypes. More than 40 of these genotypes infect the anogenital tract, causing a variety of abnormalities ranging from genital warts to invasive cancer. Certain types are considered more carcinogenic in humans. HPV-16 and HPV-18 probably are the most carcinogenic types; HPV-31, HPV-33, HPV-35, HPV-39, HPV-45, HPV-51, HPV-52, HPV-56, HPV-59, HPV-66, and HPV-68 also are included in the group of carcinogenic HPV types. HPV-6 and HPV-11, which commonly are associated with genital warts, are not believed to be carcinogenic. Although HPV certainly has many actions in the infected cell, two of the more pertinent (and certainly more studied) in cancer are its effects on the cellular genes p53 and retinoblastoma (RB). Both of these genes are known as tumor-suppressor genes. The p53 gene is present in all cells and is known as the housekeeping gene or cell cycle regulator. One of the important functions of p53 is to recognize when DNA damage has occurred in a cell and arrest the growth of that cell in the G1 period of the cell cycle to allow for DNA repair or, if repair is not possible, to lead that cell into cell-mediated death or suicide, called apoptosis. The p53 gene performs these functions by activating a number of downstream genes. Mutation or deletion of p53 is one of the most common genetic abnormalities in cancers of all types. In cervical cancer, p53 mutations are less common. Instead, the HPV early gene, E6, inactivates p53 in these cells by causing its degradation through the ubiquitin system. RB, also a tumor suppressor gene, was first recognized as being mutated or deleted in retinoblastomas. Its function is to allow or disallow progression through the cell cycle, depending on its state of phosphorylation and interaction with downstream genes. HPV inactivates RB by having its early gene, E7, bind to the RB gene and block these interactions. Both of these events, the promotion of p53 degradation by E6 and the blocking of RB gene function by E7, allow for unregulated growth of cells, particularly cells with DNA abnormalities.

Approximately 40% of sexually active adults are infected with HPV.5 Risk factors for HPV infection are number of recent sexual partners (unprotected intercourse) and young age. Understanding who gets infected, how long the infection lasts, and the natural history of HPV infection are the first, most important steps in learning about the process that leads to cervical cancer. With HPV testing increasingly considered a primary screening tool, either alone or as an adjunct to cytology, it becomes essential to characterize the circumstances and variables that mediate acquisition and clearance of infections with different HPV types.6, 7 Large proportions of women who have HPV and are screened will not have lesions and will be told that they harbor an infection by a potentially oncogenic virus that is transmitted by sexual activity. This will lead to considerable uncertainty and may have consequential psychological impact on patient quality of life. The existing knowledge base concerning the epidemiology of HPV is insufficient in that respect, and physicians are ill prepared to answer all of the questions that their patients may have regarding the significance of a positive HPV finding or of the circumstances leading to it, such as interpartner transmissibility.

At the Second International Conference on Cervical Cancer, both Eduardo L. Franco, Ph.D., M.P.H., of McGill University in Montreal, and Rolando Herrero, M.D., of Proyecto Epidemiologico Guanacaste in Santa Ana, Costa Rica, presented results from epidemiologic studies evaluating the natural history of HPV infections in different populations.8, 9 The Ludwig-McGill cohort10 involved 2528 women who were followed prospectively for evidence of HPV infection using polymerase chain reaction (PCR) detection. The group was comprised of a high-risk population of women from Sao Paulo, Brazil, who attended a maternal and child health clinic. Results from the group revealed a 15% prevalence of HPV infection with 1.3% incidence of new infections per month. Only 35% of women who were positive at enrollment remained infected at 12 months, and the monthly clearance rate was greater for nononcogenic HPV types than for oncogenic HPV types. Franco also presented the results from another HPV infection study of a group of McGill University students.11 The baseline prevalence of HPV infection in that group was 29%, with 21.8% high-risk types and 14.8% low-risk types. In that group, persistence of HPV infection was similar in nononcogenic types and oncogenic types, with a median retention of infection of 16.6 months for high-risk types and 14.7 months for low-risk types. In these studies, it was found that young age and the numbers of recent sexual partners were related to HPV infection. Use of oral contraceptives was related to infection with high-risk or oncogenic HPV types.

Dr. Herrero has focused on another group of women at risk for HPV infection to investigate the natural history of HPV infection and cervical neoplasia and cofactors for progression to high-grade squamous intraepithelial lesions (HGSIL) and to evaluate new cervical cancer screening techniques.9, 12, 13 The Guanacaste project included a random sample of 10,000 women who underwent intensive screening, colposcopic referral, treatment, and follow-up for HGSIL outcome. Worsening Pap test abnormalities were associated with increased HPV detection. Among women with Pap tests with normal results, 11.2% were HPV-positive; among those with low-grade squamous intraepithelial lesions, 72.9% were HPV-positive; and among those with HGSIL or cancer, more than 88% were HPV-positive. High-risk HPV types, specifically HPV-16, were more common in HGSIL or cancer. Risk factors found to be associated with HGSIL or cancer among HPV-positive women were high parity, current smoking, and long-term use of oral contraceptives. Among certain subgroups of women, risk factors included markers of cervical or vaginal inflammation and specific human lymphocyte antigen types.

The results of these studies led to more questions and directions for future investigations. Future epidemiologic studies should incorporate the designs and methods that are appropriate to answer questions regarding the impact of HPV infection from a multidisciplinary standpoint (i.e., covering the putative behavioral, environmental, and host susceptibility variables that mediate the risk of infection). Also needed is an understanding of the duration of the protection conferred by a negative HPV test. The published literature on HPV testing is based on cross-sectional assessments of the relation between positivity and lesions but provides reassuringly high negative predictive values.13–15 However, no studies provide information on the long-term risk of women for HPV who have negative HPV results at their first screening. Such studies are needed to orient future policy decisions concerning screening intervals and the need for ancillary tests to improve prognostic performance. The effects of age and hormone use suggested by some studies also require further investigation. Another area that is understudied is the importance of HPV infection in males. Finally, further study is needed for the development of vaccines against HPV using intermediate biomarkers.

In Costa Rica, Dr. Herrero is beginning a vaccine study of at-risk women. The National Cancer Institute is sponsoring an HPV-16 vaccine trial to evaluate the efficacy of two HPV-16 vaccines to protect women against persistent HPV infection and the development of HPV-positive squamous intraepithelial lesions and also to define the rates of specific side effects associated with vaccination. Secondary objectives are to evaluate the duration of protection in relation to local and systemic antibody levels, to evaluate protection against infection or lesions with other HPV types, and to investigate the distribution of other HPV types in vaccinated women. They will initially test HPV16 L1 virus-like particles (VLPs) as vaccines, and L1-L2-E2-E7 VLPs will be introduced in a later phase of the trial. The L1-L2-E2-E7 includes a viral capsid plus mutated early HPV genes. These vaccines are highly antigenic and effective in animal models, and the L1 VLPs have proven highly immunogenic and safe in early trials.16, 17 The study will be a double-blind, randomized Phase III trial of 22,000 women ages 18–25 years. They will receive L1, L1-L2-E2-E7, or saline placebo. Vaccine will be given at baseline, at 1 month, and at 6 months. Women will then be followed yearly with interviews, liquid-based cytology, cervical secretion collection, and blood and HPV testing. The results from this important trial, which are expected in 6–7 years, are awaited eagerly.

As investigators begin to understand the epidemiology of HPV infection and define its relation to cervical cancer, parallel research aimed at understanding the molecular mechanisms of cervical cancer development continues. Jeffrey M. Arbeit, M.D., of the University of California–San Francisco, who uses murine models to investigate cervical cancer prevention and carcinogenesis, presented important studies using a transgenic model to study the effects of HPV and estrogen on cervical cancer development.18–20 His research group's model, K14-HPV16 transgenic mice, is an expression of the entire early region of HPV16 to the basal keratinocytes of squamous epithelium using the human keratin-14 expression cassette. Eighty percent of these mice develop spontaneous skin cancers by age 8–12 months, but no cervical carcinogenesis. To induce cervical carcinogenesis, female transgenic mice age 1 month are treated with continuous-release pellets containing 17β-estradiol at 2-month intervals for up to 6 months. Transgenic mice develop lesions like those of cervical intraepithelial neoplasia 1 (CIN-1) after 1 month, CIN-3-like lesions by 3 months, and carcinoma in situ or invasive cancers by 4 months of treatment. Invasive cancers develop in the vulva, vagina, or outer cervix and at the cervical transformation zone, which, in the mouse, is at the upper cervical canal adjacent to the insertion of each uterine horn. Decremental estrogen titration identified a lower estrogen dose that is more selective for transformation zone carcinogenesis. Further studies found that E7 is necessary and sufficient for estrogen-induced neoplasia. Using mice containing the E7 open reading frame (K14-E7 mice), investigators found that cervical squamous epithelial proliferation was 2–3 times greater compared with mice that carried only the E6 gene or with nontransgenic mice. The correlation between estrogen and cervical cancer development also varied, depending on the strain of mouse used. Some strains were more susceptible to carcinogenesis than others. These studies lend support to epidemiological studies that showed a difference in HPV infection rate and development of cervical lesions in women using oral contraceptives, as found by both Franco and Herrero.

Models of this type in the study of the development of cervical cancer can be particularly useful in evaluating chemopreventive agents. Early studies of αdifluromethylornithine using this model found a decrease of 25% in the incidence of cervical cancer. This model also can be used to identify candidate molecules that contribute to the progression or spread of cervical cancer.

Christopher Crum, M.D., of Harvard Medical School, has taken a different approach to studying the molecular events in cervical carcinogenesis by evaluating the genes that are up-regulated by HPV infection and the target epithelium.21 It has been shown by immunohistochemistry that genes like MN, CEA, Ki67, cyclin E, and p16 are up-regulated in HPV-infected or neoplastic epithelium. Gene arrays also have uncovered a wealth of new candidates that currently are being tested. Ultimately, the concepts addressed by these technologies will include: 1) more specific markers to determine which abnormal Pap tests merit immediate attention, 2) markers for resolving the biology of problematic histologic abnormalities, and 3) solution-based screening techniques to replace the Pap smear. The first two concepts are being addressed currently with HPV testing and immunohistochemistry and are designed to reduce the number of women subjected to either colposcopy or surgical ablation.

In addition, Dr. Crum is evaluating the importance of p63, a newly identified gene that is a homologue of p53, in the development of the cervical transformation zone.22 The p63 gene is associated strongly with basal and reserve cells, and it is believed that p63 is of critical importance in the development of the transformation zone. By using this biomarker, in some cases, Dr. Crum has been able to identify unique subsets of basal or reserve cell populations in cervical neoplasms.23 The goal is better understanding of cell types infected and their role in the evolution of cervical neoplasia. The underlying theme of Dr. Crum's work is the importance of developing biomarkers of cervical cancer.

Telomerase is another potential biomarker of cervical cancer, and it is being studied by Kenneth R. Shroyer, M.D., Ph.D., of the University of Colorado Health Sciences Center.17, 24, 25 Telomerase is a ribonucleoprotein DNA polymerase containing an RNA component, designated hTR, and a single catalytic protein subunit, human telomerase reverse transcriptase (hTERT), which directs the synthesis of telomeric DNA repeats. Telomerase contributes to cellular immortalization as a fundamental component of the processes of malignant transformation and may play a central role in HPV-mediated transformation of the cervical mucosa. Expression of hTERT and hTR is confined to the lower levels of the normal cervical mucosa but extends into the upper layers of the lesion epithelium in most cases of cervical dysplasia. This biomarker has undergone some clinical testing. Studies at the University of Colorado Health Sciences Center have focused on the evaluation of telomerase and HPV in a large series of colposcopy patients with a recent history of abnormal squamous cells of unknown significance or another cytologic abnormality of higher grade. Telomerase expression was detected by the telomeric repeat-amplification protocol assay, with amplification products evaluated by polyacrylamide gel electrophoresis and phosphorimager analysis. HPV DNA was detected by L1 consensus PCR. Dr. Shroyer presented preliminary data from this high-risk population. Telomerase expression reflected underlying high-grade dysplasia on cervical biopsy, but it did not correlate well with detection of high-risk HPV types. Ongoing studies will help determine the role of this biomarker as a diagnostic adjunct in high-risk populations.

The overall objective of the work of Drs. Arbeit, Crum, and Shroyer is the identification of and testing of new biomarkers important in the development of cervical cancer. Identification of molecular changes in cervical cancer and knowledge about the importance of HPV infection and cervical cancer may lead to new therapies to treat existing cervical cancer and, in the long term, prevent the disease.

One potential molecular therapy for cervical cancer was presented by Judith K. Wolf, M.D., of The University of Texas M. D. Anderson Cancer Center. The HPV E6 gene, as discussed above, interrupts the function of the p53 gene, allowing genetically abnormal cells to proliferate indefinitely. To evaluate the effect of replacing p53 in cervical cancer cells with HPV infection, her research group used an E1-deleted, replication-deficient adenovirus to deliver human p53 cDNA under a cytomegalovirus promoter (adenovirus-mediated p53 [Adp53]) to cervical cancer cell lines in vitro and in a nude mouse model. Cell line studies included those with HPV16 or HPV18 infections and an intrinsic p53 mutation with no HPV infection. Dr. Wolf and her colleagues found that, in both the in vitro model and the mouse model, treatment with Adp53 inhibited cell growth and tumor growth. They also showed that p53 was overexpressed and that infected cells underwent G1 arrest. Using the Rhesus monkey cervix as a model similar to human cervical epithelium, they found that the adenovirus could be delivered best by direct injection into the cervical epithelium. These findings are similar to what has been found in squamous epithelial carcinomas of the head and neck. In clinical trials, direct injection of Adp53 for patients with head and neck malignancies has proven to be safe, and an international Phase III study of Adp53 with chemotherapy currently is ongoing in patients with head and neck cancer. With regard to the future of gene therapy in patients with cervical cancer, Dr. Wolf and her colleagues believe that the preclinical evidence supports clinical trials of Adp53 in cervical cancer. Developing better delivery systems for gene therapy and placing genes under specific promoters, so that expression of the transgene occurs only in the targeted cell, also are areas of active investigation. In addition, with the identification of new biomarkers specific for cervical cancer, gene therapy potentially may target one of these markers specifically.

The development of therapeutic as well as preventional vaccines is another area of importance in the field of new treatments for cervical neoplasia. Naked DNA vaccines have emerged as attractive approaches for vaccine development. However, their potency is limited by their inability to amplify and spread in vivo as some replicating viral vaccine vectors do. Tzyy-Choou Wu, M.D., Ph.D., of the Johns Hopkins Medical Institutions, has been investigating development of intracellular targeting and intercellular spreading strategies as an option for enhancement of DNA vaccine potency,26–30 and he described how intracellular targeting directs antigens to different subcellular locations to enhance antigen processing and presentation. For example, the sorting signal of lysosome-associated membrane protein type 1 (LAMP-1) can be linked to HPV16-E7 antigen to redirect HPV16-E7 antigen from a cytoplasmic protein to the endosomal/lysomal compartments, resulting in enhancement of antigen processing and presentation to E7-specific, CD4-positive and CD8-positive T cells.31 Meanwhile, intercellular spreading facilitates the distribution of antigens to neighboring cells by taking advantage of unique intercellular transport properties, thus allowing for an increase in the amount of antigen presented to effector cells. This strategy has been exemplified using herpes simplex virus type 1 VP22 protein linked to E7 antigen, which leads to an increase in the number of antigen-presenting cells that present antigen to effector cells and enhancement of E7-specific, CD8-positive T-cell immune responses.32 These intracellular targeting, intercellular spreading, and combinatorial antiangiogenesis strategies have shown promise for the treatment of a preclinical model of cervical cancer and may be applied to other cancer systems. The encouraging preclinical data have led to several clinical trials to test these hypotheses. The presentations of both Dr. Wolf and Dr. Wu suggest promising translation of preclinical therapies to patients with cervical neoplasia.

In summary, innovations in understanding the biology of cervical cancer include the study of the epidemiology of HPV infection, the development of new biomarkers through investigation of the underlying molecular changes associated with cervical cancer, and the translation of new therapies into treatment and prevention strategies to combat cervical neoplasia. Humankind may be on the threshold of what has been called, subsequent to the conference, ‘the beginning of the end for cervical cancer’.33 With the rapid development and translation of molecular technology, in the not-so-distant future, we may see patients with cervical cancer treated with targeted vaccines, such as the one described by Koutsky et al.,4 or molecular therapies. Even better, girls may one day receive their HPV vaccination along with other childhood vaccinations and enjoy protection against cervical neoplasia altogether.

Ancillary