It won’t hurt to try? Or will it?

There are many gray areas in veterinary cancer care. Rarely am I certain that a particular treatment option or surgical strategy or chemotherapy protocol is “the absolute best” plan of action for any given patient.

My uncertainty stems not from a lack of knowledge or experience; it arises from a dearth of evidence based information to guide my decision making process.

Practicing evidence based medicine means I would conscientiously explore only the current best proof in making decisions about the care of my patients. This requires scouring research summaries and scrutinizing details contained within the reports to define the applicability of such work to the specific pet presented to me in the exam room.

As an example, evidence based tells me that the optimal treatment plan for a dog diagnosed with multicentric lymphoma is a multidrug chemotherapy protocol administered over a six month period. This combines the lowest chance of side effects with the longest anticipated survival time. Similarly, research tells me the patient’s prognosis without treatment is only 2-3 months.

These statistics are based off of data accrued during studies designed specifically to look at the outcome of many dogs diagnosed with lymphoma treated in a similar fashion, allowing conclusions to be drawn that are applicable to a wider subset of patients.

The contrary of evidence based medicine is incorporating the idea that “anything that could help, and doesn’t hurt” is a valid option for a patient’s treatment regimen. This approach relies not on factual information but on “soft findings,” such as personal experience, anecdotes, or even ambiguous best guesses.

There are several flaws with this latter approach to practicing medicine, namely the assumption of a failure to cause harm. Even when there is a lack of a positive response to therapy, this doesn’t imply an absence of a potentially negative outcome.

Owners frequently approach me with questions about untested remedies they’ve read about on the internet or that were suggested by a caring friend, relative, breeder, therapist, etc. While some of these purportedly “harmless” options are likely to be truly harmless, my concern is that the negative effects of others are potentially vastly underestimated.

For example, owners inquiring about feeding their dogs Gatorade when they are feeling ill are unlikely to harm their pets by doing so. I inform them that the small volume of fluid they are able to feed to their pet orally will not provide enough glucose (sugar) and electrolytes to reverse acute dehydration, but as long as there’s no artificial xylitol sweetener in the product, the chance of causing harm is minimal. I can’t think of a specific study proving my assumption, but I’m comfortable with my conclusion nonetheless.

The bigger problems are those seemingly innocuous therapies where evidence based information in scarce but questionable enough to raise concern for a detrimental effect. Consider the supposed benefits of antioxidant supplements for dogs and cats.

Research supports the concept that antioxidants are able to protect cells from free-radical damage — in test tubes and living animals. However, opposing research has shown that antioxidants can potentially increase risk for disease (e.g., cancer), as well as counteract the beneficial effects of treatments such as chemotherapy.

It’s surprisingly difficult for a doctor to know how to keep the evidence-based medicine in check and ensure that the optimal standard of care is offered for their patients. I may not always be able to use research based information to make decisions about my patients’ care, but I also am wary of accepting an option simply because “it couldn’t hurt.”

I spend a lot of time researching options, hitting walls, and being frustrated in the lack of confirmative data to guide the decision making process. This process allows me to maintain the greatest responsibility I have to my patients: to “first, do no harm.”


My vet did all these tests and we still don’t know anything…

Diagnostic tests are essential to my daily activities as a veterinary oncologist. For example:

I require a complete blood count (CBC) test before every chemotherapy treatment.

I analyze results from fine needle aspirates and biopsies in order to formulate therapeutic plans.

I use radiographs (x-rays) to look for metastasis (spread) of cancer to internal organs.

I request ultrasounds to compare tumor size before and after therapy to ensure success.

Every test I order requires interpretation. The expectation is that I will always know precisely how to do so. The reality is I typically do. But sometimes I struggle to decipher the precise “next best step.”

Results typically exist either on a quantitative (yes or no) or qualitative (sliding scale) basis. Most owners assume I’ll present them with the former. Their dog’s CBC will either be good or bad. The aspirate will show cancer or a benign growth. The radiographs will depict metastases or be clear. The ultrasound will measure growth or shrinkage.

Unfortunately, with few exceptions, nearly all results possess some degree of intrinsic qualitative characteristics.

The patient’s platelet count on their CBC may be considered adequate for administering chemotherapy, but if the numerical value is 50% lower than it was the week prior, I’ll pause to ask “why?” before ordering their drug.

Aspirates can show cancer but still not provide enough information to give me the exact tissue of origin, precluding a specific treatment plan.

Radiographs can suggest spread of cancer, but the pattern could also result from pneumonia or asthma, offering three completely different diagnoses and prognoses.

The ultrasound might reveal a change in the appearance, but not size, of a tumor, leading to the possibility that the cancer is not as well controlled as the measurements imply.

Ambiguous outcomes are, at minimum, frustrating for both veterinarians and owners. More often, if owners are unaware of the possibility of an indeterminate result, they could over interpret the equivocal diagnostics, inappropriately assuming an incorrect positive (or negative) conclusion.

The utmost unfortunate situation occurs when owners, unprepared for the possibility of inconclusive results, leave the clinic focused on how they’ve spent a great deal of money on tests they feel showed “nothing.”

Through personal experience, I’ve learned the importance of explaining anticipated uncertainties prior to an owner committing to any given test. The most important warning I can offer to an owner is, “absence of evidence is not evidence of absence.”

Consider the usefulness of thoracic radiographs (chest x-rays) for predicting metastasis in a dog diagnosed with appendicular osteosarcoma (a form of primary bone cancer).

Evidence based information based on studies with hundreds of dogs tells me that 1) greater than 90% of dogs with osteosarcoma will have negative thoracic radiographs at the time of diagnosis, and 2) within 4-5 months following amputation of the tumor-containing limb, 90% of those same dogs will develop radiographically detectable tumors in their lungs.

We conclude that the metastatic tumors were present when the first set of x-rays was taken, despite the report indicating the scans were clean. Clearly, the absence of evidence on the first set of x-rays is not absolute evidence of the absence of tumors for the majority of dogs.

To make a medically appropriate choice for their dogs with osteosarcoma, owners need to be aware of the predictive value of the first set of radiographs, and that the lack of initial spread of disease doesn’t preclude future metastasis. This also underscores the importance of repeating x-rays at specific time points following surgery.

Medical analyses are a necessary part of the treatment plan for my patients. They are an essential part of monitoring and ensuring pets are healthy enough to withstand further procedures and therapeutics.

I rely on my experience and intuition to fill the gap when results are confusing or inexact. Those same attributes allow me to predict the possibility of an uncertain answer and to talk about those possibilities with owners before reports are entered into their pet’s chart.

Owners should also feel comfortable enough to ask their veterinarian about the expected possible outcomes of recommended tests, including the positive, negative, and “in between” results.

This will ensure that expectations are clear on both sides, so that we can each contribute to the optimal treatment plan for the pet.

When the remedy is the poison…

There’s a specific routine we follow for each pet arriving for a chemotherapy appointment. Owners arrive and are greeted by a technician, who will ask several questions about how their pet is doing and if any complications from a previous treatment arose.

If all is “status quo,” the patient will be taken to our treatment area, where their vital parameters (temperature, heart rate, respiratory rate, and body weight) will be recorded and the required blood samples will be drawn and run in our laboratory.

I then perform a full physical exam and make sure there are no contraindications to treatment (i.e., health related reasons to withhold treatment).

The oncology technician will retrieve the lab results, examining the printout for any sign that the blood machines are having a meltdown, and if necessary, make blood smears for me to interpret in conjunction with the automated results.

I review the results, then write out the prescription for the chemotherapy drug, including all associated calculations, determining the amount of drug in both milligrams and millilitres where applicable, and reiterating the route of administration (e.g., intravenous, subcutaneous, orally). Every calculation is then double checked by the technician responsible for administering the dosage.

The patient’s body weight, drug, dosage, and amount, as well as results of their lab-work, are manually entered on their “chemotherapy flowsheet,” a tangible record of all prior treatments.

Current dosages are back-checked to that patient’s previous dosages, where applicable. For example, we cross-reference their current weight to be sure it is within their previous weights, that it was recorded in the correct units (kilograms versus pounds), and that the dose of chemotherapy is similar to what it was at a previous visit.

This painstaking attention to detail may seem ridiculously tedious. Why is the process of administering a medication so involved—especially when that patient has received the same drug numerous times before? What is the point behind the orderly procession of events we prescribe?

The answer lies in what is known as the narrow therapeutic index of chemotherapy drugs.

Therapeutic index refers to a comparison of the amount of a drug necessary to cause a beneficial effect and the amount causing toxicity.

Paracelsus, a 16th century philosopher, stated, “All things are poison and nothing is without poison; only the dose makes a thing not a poison.” This is frequently paraphrased to, “the dose makes the poison” (Latin: sola dosis facit venenum), an excellent summary of the basis of therapeutic index.

Every prescription medication has a therapeutic index. A dose below the lowest margin of this index will result in a lack of effectiveness. A dose above the highest margin can lead to side effects. In the most extreme cases, the side effects can equal death. Dosages within the therapeutic index will be effective for treating the condition in question, but will remain non-toxic for the patient’s healthy cells.

Some prescriptions have a wide therapeutic index, and veterinarians have a good deal of “wiggle room” in what can be dispensed based on a given patient’s size.

For example, the same exact dosage of an antibiotic can be equally therapeutic for a 30lb dog as for a 50lb dog. Similarly, a 50lb dog can be prescribed 2-3 tablets of a particular pain medication to be given every 8-12 hours. The wide therapeutic index of those drugs allows for such variations.

Chemotherapy drugs, on the other hand, have little to no safety margin and a very narrow therapeutic index. This means the dosage of a chemotherapy drug necessary to cause an anti-cancer effect is very similar to that which causes adverse effects.

Therefore a slight error in calculation leading to even a minuscule overdose of drug can lead to catastrophic effects for that patient. In those cases, the patient’s healthy tissues will be exposed to levels of drug that can be at best moderately damaging or permanently affected, and at worst cause a fatal reaction.

We might be able to cure more cancers in pets if we could give them higher dosages of chemotherapy, but we would also bring those animals to the brink of death before any potential success. This is neither an ethically or financially feasible option in veterinary medicine. We also would have a much higher death rate from treatment, losing large numbers of patients to complications from treatment rather than disease.

I’d be remiss if I didn’t acknowledge that at least part of my anxiety about dosing chemotherapy arises from my Type A personality. I’m known for calculating and re-calculating doses several times before giving a thumbs up on the prescription (and even continuing to recheck calculations as the drug is being given). My paranoia stems from knowing all the things that can go wrong when the therapeutic index is breached. However, it’s certainly fueled by a tiny bit of compulsion as well, as I tend to be more obsessive about such details than my colleagues.

With proper and meticulous attention to detail, I’m ensuring that the therapeutic index of chemotherapy drugs I prescribe isn’t breached and errors are avoided.

Although it’s certainly monotonous to perform so many extra steps for every appointment, the process is integral to guaranteeing my patients are treated with the same standard of care I would expect for myself.