By John L. Marshall, In my practice as an oncologist specializing in gastrointestinal tract cancers, a recent week was fairly typical. I saw 50 patients, ranging in age from 32 to 87, equally divided between men and women.

Though a couple of them have inherited a gene that may have caused their GI cancers, I have no explanation for why most developed their disease. It is as if they were simply struck by lightning.

My patients seek state-of-the-art therapy, access to clinical trials and new treatments, all of which we provide at our institution. Almost all of them have insurance, and most have some form of prescription drug coverage; their access to care is virtually limitless. We employ the latest diagnostic tests, targeted chemotherapy, minimally invasive surgical techniques and incredibly precise radiation. Yet, despite the many recent advances in detection and treatment, of the 50 patients, 40 of them are likely to lose the fight.

In the extensive debate over health-care reform, we have heard little discussion about the enormous cost of cancer care. (Some of the only voices to broach the subject are those fearing “death panels.”) But at this moment, when a significant shift in the health system in this country is possible, we must ask some difficult questions: Does it make sense to support cancer care at the current levels in the United States? Who should determine the value of care?

At the moment, there is a giant disconnect between patients, the cost of care and the clinical benefit of the treatment — a disconnect that has caused us to lose perspective. When it comes to cancer care, we’re not getting what we pay for.

Cancer medicine is often regarded as an area of significant progress and clinical research, so we should be able to tell without much difficulty what kinds of treatment are valuable and what kinds aren’t. But given that 80 percent of my patients will die of their cancer, it’s clear that we have not found an “optimum” therapy.

In 1971, President Richard Nixon declared war on cancer, with the charge to “cure” the disease. Since then, billions of dollars have been spent on research, yet we have made only minor real progress. The most common approach to treatment involves exposing large populations of patients to highly toxic poisons in the hopes that the treatment will kill the cancer cells and not the patient.

This strategy has succeeded with several types of less-common cancers, curing some patients with leukemia, lymphoma, testicular cancer and most childhood cancers. But it has not worked for more common forms of the disease, including breast, prostate, lung, colon, pancreas, stomach and ovarian cancers. These cancers represent an enormous public health problem, consuming the majority of our cancer-specific health-care costs and research dollars.

In many ways, we have quit trying to win the war on these diseases. Few cancer clinical trials are designed to “cure” patients. They are commonly aimed at detecting small differences between the treatments being compared: an extension of average survival from 5 months to 6 months, for example. These trials typically cost millions of dollars (often including taxpayer support), take years to complete and can involve thousands of patients. It is this kind of care that many Americans are afraid they will lose access to as a result of health-care reform.

Globally, cancer care is a medical luxury. With some diseases such as colon cancer, the treatments alone cost more than $15,000 a month, yet on average add only a few months to survival. Most poor countries do not support any cancer care; most developed countries highly restrict it because of its cost and limited effectiveness. The United States is the only place on Earth with relatively unfettered access to cancer care, including the latest medicines, sophisticated scans and high-tech radiation, all of which are very expensive. But despite their more limited access, cancer patients in other high-income nations may live longer and with a higher quality of life than patients in this country.

If the Obama administration were to call on oncologists to help choose which cancer treatments would be included in a public option plan and which would fail to meet some “value” standard, we would have no guidelines to follow. The Food and Drug Administration uses safety and efficacy as standards for drug approval, but neither of those considers the magnitude of benefit or cost. I frequently ask my students and peers if there is a cancer drug today that they would pay for out of pocket if they had to. We all have patients who lack insurance but have some financial means, making this an exercise that is played out regularly in our clinics.

After a long pause, someone invariably will say “Gleevec,” a pill that treats a terrible disease called chronic myeloid leukemia. Gleevec, a brand name for imatinib, costs up to several thousand dollars a month, but it makes this leukemia “go away” for a long time with very few side effects — a true magic bullet. Very few cancer drugs can be described as having this kind of value. “Thank God for my insurance” is a common line from patients.

In cancer medicine, fewer than 5 percent of all patients in the United States enter clinical trials. That means more than 95 percent are treated with the “standard of care” — a legal term denoting a minimum level of care for an ill or injured person.

How did we end up here? The answer is simple: Cancer patients are scared for their lives and will accept what is offered, and we oncologists want to offer improved outcomes and recommend the best treatments we can. Insurance will pay for these treatments. A portion of fees collected by cancer doctors and hospitals is based on how much chemotherapy we administer. So the more drugs we give, the more radiation we give, the more we collect from health insurance. The incentive system makes it less lucrative to talk to patients — to counsel them, to help with their decision-making — than to treat them, regardless of the value of the treatment.

A major focus of health-care reform is for doctors to practice evidence-based medicine: to offer therapies that have been proven to help patients live longer or, at least, live better. The problem in cancer medicine is that we have very little evidence to support what we are doing. Because so few of our patients enter clinical trials, we have no way of tracking their outcomes collectively, and we learn almost nothing from them. Our understanding of cancer therapies comes from the 5 percent of patients who enroll in trials, a tiny database when we consider the highly variable nature of the disease. In fact, most of our evidence has not come from clinical trials performed in the United States but from nations where patients have little access to advanced care except through participation in such trials.

In this country, the highest hurdle we must leap is our patients’ expectations. Cancer patients facing death want treatment; they want hope that they will be cured, even if they have been told that they cannot be cured. They will try toxic treatments over and over, hoping to extend their lives. We physicians are co-conspirators. Of course, we also want to believe that the next treatment will help more than the last, even though we know that is rarely the case. What if we had to pay for all this out of our pockets? Would we pay that much for some possible hope?

I believe we can invest more in actual hope. To do so, we must further explore the genetic makeup of patients and their cancers. We can no longer diagnose cancers using only a microscope. We must profile them at a molecular level to determine precise treatments, instead of using our current trial-and-error approach.

To assess a patient’s specific genetic problem, we must understand all the possible permutations and patterns. This will come only from a comprehensive clinical database — a high priority of the administration’s reform plans. For example, we know there are at least four different types of breast cancer; they look exactly the same under a microscope but are very different diseases. The repeated biopsies and blood tests that are needed, none of which is covered by most health insurance plans, will become critical to finding our answers.

The future of cancer care will rely on personalized medicine. This requires a significant change to our medical system, which is built around one size-fits-all treatment and seemingly unrestricted access to care. The system answers our emotional needs and provides some hope for a cure, but moves us forward only a few yards at a time.

Oncologists are optimists, and I am proud to be among them. I truly believe we can cure cancer. I care greatly for my patients and am doing everything in my power to improve and lengthen their lives. When I offer a clinical trial to a patient, I am hopeful that it will be better than the standard treatment. I am optimistic that health-care reform will not simply provide everyone with insurance that will cover the “standard of care” but will also force us to determine the true value of treatments.

There is nothing that focuses the mind more than a cancer diagnosis. I can assure you that all of my patients have very clear vision and have shifted their priorities. We as a country must do the same on their behalf. Health-care reform — and a better understanding of what we’re paying for when we treat these terrible diseases — will help focus our approach to caring for cancer patients and finding a cure.

John L. Marshall is the director of the Otto J. Ruesch Center for the Cure of Gastrointestinal Cancers at the Lombardi Comprehensive Cancer Center at Georgetown University. He will be online to chat with readers on Monday at noon. Submit your questions and comments before or during the discussion.