Ovarian cancer is an avaricious tumor, and its domain is nothing less than the entire abdominal cavity. It can extend from the deepest part of the pelvis up to the diaphragm and to the right and left of the colon and everything in between. It can appear after a few weeks of the mildest symptoms, and by then it has already declared open season on the body of a woman. It is fiendishly difficult to treat and unrelenting in its destructive ambition. It is a modern day scourge, casting a narrow and selective net, forever changing the lives of its victims.
The initial symptoms of ovarian cancer are vague and frequently present as gastrointestinal disorders. The ovary is the only organ in the body that has its functioning cells facing the interior of the abdomen, so long before a tumor actually forms, cells detach and implant on the undersurface of the diaphragm, the capsule of the liver, and most important, on the surface of the bowel.
Hundreds of nodules accumulate on the serosal surfaces of the large and small bowel, impeding the smooth flow of intestinal contents, and cause, along with the production of ascites, the cramping distention of the abdomen, which is the hallmark of ovarian cancer. The presentation is actually an intermittent, partial small bowel obstruction and represents a stage 3 cancer at the time of diagnosis. The finding of a stage 1 cancer is usually a serendipitous event – the surgeon is operating for some other reason, and a small nodule on an ovary is discovered.
Despite what you may have been told, there is no way to screen for ovarian cancer. CA-125 is a protein that has been around since 1981 and is merely a test for inflammation. It is by no means specific for this cancer. In fact, the CA-125 blood test is negative in 20 percent of patients with advanced cancer.
My lecture on ovarian cancer contains a slide with the heading, Is It Possible To Screen For Ovarian Cancer? The remainder of the slide contains the word no in 41 languages. Many laboratories are involved in the discovery of a diagnostic test for ovarian cancer that would approach 100 percent accuracy. Such a test would be one of modern oncology’s Holy Grails!
The Ovarian Cancer Institute was founded in 1999, and its work is centered in the McDonald Laboratory in the Department of Biology at The Georgia Institute of Technology. For the past 15 years, we have been investigating the genetic and molecular structure of ovarian cancer in the hope that a highly accurate diagnostic test might one day emerge.
Tissue and serum samples are immediately flash-frozen in my operating room at Northside Hospital and transported to the lab at Georgia Tech. They are stored in the minus 80 degree Celsius freezer, rendering them “eternal.” These specimens can be as accurately studied 100 years from now as they would be on the day they were collected. We now have one of the largest serum, tissue and data banks for ovarian cancer in the world.
This work is unfortunately expensive. Several years ago the Institute paid $250,000 for a laser capture dissection microscope. This device allows us to outline precisely the tissue we wish to analyze and then detach it from the specimen. The DNA from this tissue is extracted, thus allowing for precision analysis. The DNA is not contaminated by stroma or connective tissue but represents the epithelium of the ovarian cancer. The DNA is then transported to a microarray analyzer. This unit allows us to identify genes that are aberrantly expressed in ovarian cancer tissue.
Years ago, if researchers were interested in studying the genetic morphology of a cancer, they would have to proceed one gene at a time. Today, the microarray analyzer prints out the entire genetic composition of a cancer in quadruplicate on a microchip the size of a thumbnail.
The Ovarian Cancer Institute is very fortunate to be located at Georgia Tech, where there are so many departments working in areas related to ovarian cancer research, including bioengineering, bio-informatics and nanotechnology. There are many ways in which basic science research may eventually impact the way in which patients with cancer of the ovary are treated, but for now, the Ovarian Cancer Institute is focusing on three areas.
1) The Diagnostic Test
If only it were as simple as it is with cancer of the cervix. A pap smear is positive, a biopsy directed with the colposcope shows a CINIII lesion and a LEEP conization is done in the office under local anesthesia completing treatment and preserving the uterus.
The pap smear, unfortunately, is useless in the diagnosis of cancer of the ovary. A positive pap smear has led me to the diagnosis of this cancer only three times in my career. A diagnostic test for ovarian cancer must approach 100 percent accuracy, otherwise cancers will be missed or women will undergo unnecessary surgery.
Our initial attempts at the discovery of a diagnostic test at the Ovarian Cancer Institute involved the study of proteins. These are large and cumbersome structures that produced inaccurate results. We eventually started using mass spectrophotometric analysis of metabolites found in our serum samples. This instrument is amazingly accurate in separating out peaks in similar metabolites from the many samples studied.
We published our results several years ago and reported a nearly 100 percent accuracy. The only time that we found a positive result in a patient with a benign tumor was in someone whose mother and grandmother had died of ovarian cancer. The justifiable criticism of this paper concerned the fact that there were so few samples from stage 1 cancers. No one wants a diagnostic test that is positive only in advanced disease. We then purchased 90 serum samples from patients with stage 1 disease and found that our test picked up every one of them. The data analysis is complete, and we are about to publish our results. It should be noted that the test will be run on a single drop of serum and cost only a few dollars.
2) Targeted Gene Therapy
Very little has changed since Sidney Farber ushered in the modern age of chemotherapy in the mid-1940s. Newer drugs have been developed, dosage has changed as have routes of administration. One thing, however, has remained constant – the pineal gland gets as much of the drug as does the nucleus of the cancer cell.
This is most unfortunate since chemotherapy is a poison, and the dose and the interval between treatments is directly related to the body’s ability to withstand repetitive poisoning. It would be wonderful to be able to deliver the chemotherapy drug to the cancer cell and only to the cancer cell. This would allow the use of a dosage unthinkable today.
Modern genetic profiling identifies specific genes disrupted in a cancer. It is estimated that only 10 percent of mutated genes in a cancer are druggable at the protein level, which is the level at which drug therapy is currently focused.
Targeted gene function at the RNA level is preferred because all malfunctioning genes in a cancer can be targeted at the RNA level. The problem resides in the inability to deliver these RNA-inhibiting drugs directly to the cancer cell. It is important to remember that DNA is the same in every cell, but RNA codes for specific function. In collaboration with the nanotechnologists at Georgia Tech, we are developing a new class of nanoparticle delivery vehicles for this purpose. These technologies are being tested in animals and, if successful, will lead to phase 1 trials in humans.
3) Personalized Medicine
Carboplatin and Paclitaxel are chemotherapy drugs that are used as first-line therapy for ovarian cancer around the world. However, because of a significant incidence of platinum resistance, there are patients who fail this regimen. Few things are more disconcerting to a gynecologic oncologist then to spend six hours in the operating room removing the last remnant of ovarian cancer only to watch it return after several cycles of chemotherapy.
The choice of a chemotherapy regimen is sometimes the roll of the dice – a prediction based merely on experience and not science. Several companies are involved in choosing the right drug for the precise genetic aberration in a particular person’s particular cancer. It is time to stop approaching cancer based on the organ or origin; rather, we must choose treatment based on individual molecular structure. We are learning that the molecular structure of an ovarian cancer may have more in common with the molecular structure of certain pancreatic cancers than it does with other ovarian cancers.
Personalized medicine in oncology simply refers to treatment based on the structural idiosyncrasies of an individual’s cancer. Once this is nailed down, we should remember that the initial regimen might not be the proper treatment should the cancer return. This recommendation is based on the work of the Ovarian Cancer Institute published a few years ago. We used the microarray analyzer to compare the structure of the primary ovarian cancer with that of the recurrent cancer expecting them to be identical. To our surprise they were frequently quite disparate.
In collaboration with the College of Computer Science at Georgia Tech, we are developing computational algorithms that can accurately predict drug responsiveness of patients based on genomic/gene expression profiles. This approach uses learning algorithms, which are much more accurate than current methods employed by commercial firms such as Foundation Medicine etc.
This development is being coupled with genomic studies (DNA/RNA sequencing analyses) on ovarian cancer primary, metastatic and recurrent tumors all collected from the same cohort of patients. Further studies aim to validate these predictions in current patients by establishing primary cell lines from patient tumor samples. By submitting the patient sample to genomic profiling, we will be able to predict drug responsiveness and hopefully delete chance from the equation.
In reflecting over a 40-year career devoted to the care of women with ovarian cancer, I find myself consumed with the sheer barbarity of it all. A sharp knife opens the abdomen from the pubic symphysis to the xiphoid process to remove cancerous cells, there’s six rounds of chemotherapy, a recurrence and then more surgery and chemotherapy, etc.
I would like to envision the next generation of oncologists sending a newly diagnosed patient to an interventional radiologist to have some cells sucked through a skinny needle passed into the tumor under CAT scan guidance. These cells would be easily grown in the cell culture laboratory, then a geneticist would create the exact antidote to the nuclear protein in the cancer cell.
This material would then be injected into the patient at 10 in the morning, killing every cancer cell without harming a hair on her head and seeing to it that she is not late for her 2 p.m. tennis match. I would call this designer therapy – Giorgio Armani constructs the cancer treatment. Maybe one day we will see this idea come to fruition.