Treat-to-Target Decisions and Dilemmas

Reflecting on recent progress in the treatment of inflammatory arthritis, I am especially gratified by the concept of treat to target (T2T). T2T places our specialty smack in the mainstream of modern medicine. After all, much of internal medicine is based on T2T, whether the target is blood pressure, glycosylated hemoglobin (hemoglobin A1C) levels, or the international normalized ratio (INR). As a scientist, I like numbers, and I welcome them to our specialty. The more the better. Interestingly, some important clinical measures, such as the pulse rate or respiratory rate, have never been established as therapeutic targets because the range is large, and circumstances dictate where in the range the sweet spot lies. Of course, age is never explicitly a target for efforts at prevention or treatment—beyond it reaching a number as large as possible—but we hope it is at least three score and ten.

As those of you who are regular readers of The Rheumatologist (TR) know, T2T has long been a theme of our publication, and we have been pleased to publish articles on the history of the investigators who established the metrics that are currently the foundation of T2T. Thus, TR has recounted the seminal contributions of investigators such as the late John Sharp, Piet van Riel, Josef Smolen, Jim Fries, and Ted Pincus, among others. Without the pioneering work of these inspired individuals, rheumatology would not have targets at which to aim. In chronicling the story of outcome measures, we have not forgotten our lab-based colleagues. We already have started a series on outstanding scientists like Eng Tan, who founded modern serology and made testing of antinuclear antibodies an essential part of rheumatologic evaluation.

Treat to target efforts could lead to better outcomes in the future.

For many years, I have been an enthusiast of the Disease Activity Score (DAS) and use it regularly when I see patients and teach house officers and fellows, constantly Googling “DAS calculator,” only to find a site established by a company with a biologic in its portfolio. The product was not a big seller, but the site remains available and is quite nifty. I like to punch in the numbers from my exam and see how close to target we currently are.

DAS Versus Patient-Reported Symptoms

Recently, along with a resident, I saw a patient whose DAS left me baffled. As is the practice in our teaching clinic, the house officer sees the patient first and presents the case to me for preliminary discussion. We then see the patient together, and I do my own history and physical exam. The resident that day was a bright young man, aspiring to a career peering though a colonoscope at colonic mucosa although, to his credit, he was interested in learning more rheumatology because of the association of inflammatory bowel disease with arthritis and the frequent use of tumor necrosis factor blockers for patients with active Crohn’s disease.

According to the resident, the patient, who we can call Mr. Jones, was 50 years of age and had had rheumatoid arthritis (RA) for about three years. Mr. Jones was on a stable regimen of methotrexate 15 mg per week and 5 mg of prednisone each day. The history suggested that the disease was not very active. The patient worked full time as a manager of an auto parts store and regularly golfed. When I asked the resident about the joint count, he said that the Mr. Jones had “some” tender and swollen joints.

I pressed the resident to be more specific about the actual number, indicating that, in medicine, we do not say that a patient has “some” hypertension or “some” anemia. Good care requires precise numbers, and rheumatology therefore must have its numbers. I then Googled the DAS calculator website to show the resident the picture of the “joint man,” with the essential 28 joints depicted by circles to make it clear where attention is needed.

“Let’s do a real joint count,” I said to the house officer, who, alas, needed a shave, “and then we can calculate a DAS.” Down the hall we went, sidestepping an old man with thick glasses and thin gray hair, shuffling along bent over a walker.

We greeted Mr. Jones, who was sitting comfortably, speaking to someone on his phone, saying quickly, “Gotta go, the doctors are here,” snapping the phone cover closed. Mr. Jones looked up and smiled, extending his hand to shake mine, clearly a sign that his disease was not too active because no one with tender joints wants to risk the pain associated with even with the gentlest clasp. Unless you knew that Mr. Jones had RA, you would probably not have expected this diagnosis. Mr. Jones looked quite healthy, and there was not even a bend or a twist of a finger joint that would be a tell-tale sign of RA.

When I questioned Mr. Jones about his symptoms, he admitted to few complaints, saying that he felt dramatically better compared to the time before he was on methotrexate and prednisone. I then examined him, gently palpating and compressing his joints, starting with the hands, then the wrists, elbows, shoulders, and knees, left and right, all 28 accounted for, keeping a close tally of those that were tender and swollen. I also checked the ankles and feet but left out the counting part.

As it turns out, the house officer was right. There were in fact some joints with synovitis (three metacarpophalangeal joints to be precise), and the tissues surrounding the joints were clearly thickened. When I pressed these joints bit harder, Mr. Jones said, “I can feel it,” which I interpreted as mild tenderness. Not surprisingly, the global assessment was low and, like some stoics and people who want to banish symptoms with positive thought, Mr. Jones rated it only a 10 on a 0 to 100 scale.

Given the physical findings, I expected a DAS in the 3 to 4 range and was prepared to recommend a boost in the methotrexate dose. The difficulty, however, was the erythrocyte sedimentation rate (ESR). The ESR was 1 and, even if the patient had synovitis in up to 10 joints (with a global of 10), an ESR that low produces a DAS in the remission range. I therefore had to ask myself which number was more informative: the joint count or the ESR rate?

Making the Right Decision?

As Ted Pincus has pointed out, many people with putatively active RA have a normal ESR, especially when adjusted for age and gender. However, the observation begs the question of whether, in the range of values we consider normal, there are important physiological or pathophysiological differences that could affect the course of RA and the likelihood for joint damage. Thus, ESR values of 1 and 30 are both “normal,” but a value of 30 may indicate the presence of proteins in the blood that reflect mischief in the joints. Such proteins could fuel inflammation just as they coat red cells to make them propel down a thin tube under the watchful eye of a laboratory technician.

In view of the discrepancy between the clinical and laboratory findings, perhaps Mr. Jones’ disease was inactive, with the ESR of 1 a bona fide measure of the quiescence of his immune system. The swelling that both the house officer and I felt could be fibrosis, indicating a process of healing or repair of past inflammation, a time of nastier events when the synovium teemed with angry macrophages, T cells, and fibroblasts. In this scenario, the tenderness could represent the aftermath of damage. Maybe in RA, joints with erosion are chronically tender, just as joints with osteoarthritis—pock marked with cysts and encrusted with spurs—are tender even if the synovium in osteoarthritis shows only a scant smattering of synovial cells.

After some thought as I weighed in the balance—three tender and swollen joints versus a DAS of 2.80—I decided to increase the methotrexate to 17.5 milligrams a week. That certainly was not a bold move, but I felt better knowing that I had at least done something that could quell the troubles that may have been brewing in three of the Mr. Jones’ joints.

I like the DAS because I have several years experience with its use and, frankly, I like the convenience of the website. I know that the DAS did not make the grade in the deliberations of the ACR/European League Against Rheumatism committee that has defined remission in RA. Nevertheless, as T2T becomes more of a standard, numerous issues will arise about any of the metrics that are used in research or clinical practice in define disease states. Whether it is the DAS28, DAS44, SDAI, CDAI, HAQ, or RAPID3, uncertainty and ambiguities will abound, starting with the extent of synovitis that is still consistent with a state of remission. We will have to contend with the meaning of joint tenderness and swelling and, trust me, answering the question, “How active has your arthritis been in the last seven days?” is not easy, especially when symptoms fluctuate wildly (“I don’t hurt unless I stand”), and the patient does not appreciate or accept the distinction between pain and inflammation.

I am happy that T2T is here to stay, even if an occasional value looks bizarre and will stymie me as I decide whether to increase or decrease an agent or add another therapy. Fortunately, rheumatologists have an increasingly large armamentarium of effective agents to allow target practice (pun intended).

Nevertheless, I have my worries and doubts about relying on existing outcome measures to make important decisions on treatment. Even if the DAS or CDAI provide values with two numbers after the decimal point, how solid—or, to use a popular word, robust—are those values?

Dear reader, I have a question that has been nagging at me since that day in clinic, and I would like your advice. In increasing Mr. Jones’ methotrexate, did I make the right call? In other words, was I on target? the rheumatologist

Dr. Pisetsky is physician editor of The Rheumatologist and professor of medicine and immunology at Duke University Medical Center in Durham, N.C.