Tight Control of Psoriatic Arthritis: Real-World Outcomes From TICOPA
Background
Over nearly 2 decades, tight-control treatment strategies have shown potential to transform management of inflammatory arthritic diseases. By using treatment strategies emphasizing early and tight control of disease activity, along with quantifiable and validated outcome measures, clinicians may limit the extent of disease-related joint damage and preserve patients’ quality of life.
In 2004, Grigor and colleagues published results of the Tight Control for Rheumatoid Arthritis (TICORA) study.1 The results demonstrated that tight-control treatment for 18 months, within such a treat-to-target paradigm, improved outcomes for disease activity and physical function in patients with rheumatoid arthritis (RA), according to prevailing metrics such as disease activity score as well as measures defined by the European Alliance of Associations for Rheumatology (EULAR) and the American College of Rheumatology (ACR). Although anti-tumor necrosis factor (anti-TNF) biologics such as etanercept were available at the time this study was conducted, the investigators used conventional synthetic disease-modifying antirheumatic drugs (csDMARDs) only.
Given these findings in RA, researchers hypothesized that tight control strategies could likewise improve outcomes for patients with psoriatic arthritis (PsA) if a suitable clinical target were specified. Accordingly, in 2010, Coates and Helliwell developed and validated 7 minimal disease activity (MDA) criteria as an objective target for PsA research and treatment goals.2
In the subsequent multicenter Tight Control of Psoriatic Arthritis (TICOPA) trial,3 Coates and colleagues in 2015 compared a cohort of 90 patients with PsA who were randomly assigned to a tight-control intervention against 92 patients with PsA who were assigned to receive standard care. Patients in the tight-control arm were eligible for therapy escalation if they achieved and maintained MDA. While TNF-inhibitor drugs were available to both groups, they were administered more often in the cohort receiving tight-control care. As hypothesized, over a 48-week intervention period, the odds of achieving an ACR 20% response (ACR20) were higher in the tight-control patient group; this group also showed improvement based on several secondary measures compared with the standard-care group.
However, it has been uncertain whether the tight-control paradigm facilitates similar outcomes in a real-world clinical context, outside of a randomized trial, and over longer time periods.
Methods and Results
Coates and colleagues recently addressed this question in a follow-up study 5 years after the completion of TICOPA.4 Drawing from the original TICOPA study populations, they examined clinical treatment and evaluation notes for follow-up eligible patients: 54 from the tight-control group (n=77) and 56 from the standard-care group (n=81). Because the follow-up data were insufficient to enable assessment of MDA, the researchers instead determined whether patients were in a state of “low” disease activity (LDA) based on strict criteria, including absence of joint tenderness and swelling, dactylitis, and enthesitis. Patients with treatment escalation or other significant medication changes were not classified as having LDA.
As Coates et al note in their current study, at the end of the TICOPA trial, relatively few patients in the standard-care cohort were being treated with biologics. However, the authors’ case notes indicate that many of those patients had escalated to biologic treatment by the 5-year mark, at which point approximately 50% of patients in each study group were receiving biologics and the proportion of patients considered to be in LDA was similar in the 2 groups. Methotrexate use had declined markedly overall.
Discussion
Allan Gibofsky, MD, JD
Kevin Yip, MD
Genna Braverman, MD
Allan Gibofsky, MD, JD, professor of medicine at Weill Cornell Medicine and an attending rheumatologist at the Hospital for Special Surgery, and his colleagues, Kevin Yip, MD, and Genna Braverman, MD, who are currently pursuing clinical fellowships in rheumatology, discuss this study and the clinical implications of published research in this area.
Allan Gibofsky, MD, JD: Today we are discussing the follow-up to the TICOPA study. The TICOPA study was an attempt to test the hypothesis that tight control of PsA would yield results as good as those we had seen with tight control of RA, that is, in the TICORA study.
Rheumatology is one of the last specialties to transition from qualitative metrics to quantitative metrics. When I trained, we used terms such as “doing well” to evaluate patients. These days, when you train, we use terms like “The Clinical Disease Activity Index (CDAI) is…”; “The American College of Rheumatology 20% improvement criteria (ACR20) score achieved was…”; “The Disease Activity Score including 28 joints (DAS28) score is…” and so on; that is because a standardized objective metric should mean the same thing to any one of us, as opposed to the qualitative comment “doing well.”
It occurs to me that a clinician would not alter a therapy for the treatment of hypercholesterolemia without knowing the patient’s cholesterol level; nor would they initiate a treatment for diabetes without knowing the patient’s glucose levels and hemoglobin A1c (HbA1c) score or for hypertension without knowing the patient’s blood pressure. Yet for so many years, we rheumatologists used qualitative evaluations rather than quantitative assessments such as joint counts, for example. When we were finally able to use objective metrics in RA, we then looked at the next step in its management: Can we couple a metric with a strategy? What we found in TICORA was that use of a metric plus a strategy — regardless of the type of strategy — was better than use of a strategy alone. We then began moving toward a treatment paradigm that incorporated both a metric and a strategy to get to a target outcome.
The next advancement in RA was the treat-to-target paradigm, which has now been incorporated specifically into our ACR and EULAR guidelines, suggesting that we should use a metric, identify a target, and take steps to move the patient toward that target. We should adjust therapy until we are there, and once there, we maintain treatment and continue to monitor the patient. Fast forward to TICOPA, a study initiated by Coates and colleagues. In TICOPA, the authors demonstrated (as they did more recently in TICORA) that employing a metric and a treat-to-target strategy in patients with PsA achieved a better outcome than using a treatment strategy alone. So that was TICOPA. The article by Coates and colleagues that we are going to discuss now is what happened to TICOPA — the “Where are they now?” question, so to speak. With that background, who wants to take us through the objectives and the methods?
Kevin Yip, MD: The main objective of the study was to see what happened to these patients years after the TICOPA trial. What happened to them, what medications were they on, and how well or poorly had they fared in terms of disease status? The original TICOPA study was a multicenter trial based in the UK. The investigators recruited patients from 8 different community hospitals/secondary care centers. Patients had been diagnosed with PsA within 24 months prior to enrollment and had not been treated with any sort of DMARDs. This study was the opening gambit in the evaluation of tight control vs standard care in the management of PsA.
In their published follow-up paper that we are discussing today, the investigators in 2018 evaluated results from only 2 of the 8 TICOPA study sites: St Luke’s Hospital in Bradford and Chapel Allerton Hospital in Leeds, England. The TIPOCA trial included 101 patients in the tight-control arm and 105 in the standard-care arm, but the 5-year follow-up across 2 sites assessed 77 patients who had been randomly assigned to tight control and 81 who received standard care.4 The researchers reviewed their case notes from the prior study and assessed current patient medications and disease status. They determined the number of patients with MDA at the end of the TICOPA trial and the number with LDA at follow-up. In this study, LDA was based on the absence of swollen or tender joints, enthesitis, and/or dactylitis, and any change or escalation in therapy.
To revisit MDA briefly, it is achieved when 5 of the 7 criteria are met: Tender joint count of 1 or fewer; swollen joint count of 1 or fewer; psoriasis area and severity index (PASI) of 1 or less; patient pain scale less than 15; patient global disease activity scale of 20 or less; Health Assessment Questionnaire (HAQ) less than 0.5; and no more than 1 tender entheseal point. So the LDA criteria were somewhat stricter.
AG: What did Coates and colleagues find, Genna?
Genna Braverman, MD: As Kevin mentioned, at the 2 sites, there were a total of 77 patients randomly assigned to tight control and 81 randomized to standard care. Follow-up data were available for 54 patients in the tight-control group and 56 patients who received standard care, so there was some whittling down from the larger population in the original TICOPA study. As Kevin mentioned, approximately 5 years after the end of the original study, the investigators first assessed the medications that patients were taking. At the end of the TICOPA trial, 93% of patients in the tight-control arm were being treated with methotrexate compared with 80% of patients in the standard-care group. At the time of their follow-up review, methotrexate was being administered to only 44% of patients originally assigned to the tight-control group and 54% of patients assigned to the standard-care group. There was a huge reduction overall in the proportion of patients receiving methotrexate.
Conversely, looking at changes in the use of biologics — and this was very interesting — at the end of the original study, 33% of patients in the tight-control arm were being treated with biologics compared with 9% of patients in the standard-care group. This finding poses interesting questions about the evolution of treatment for PsA, location-specific practices, and regulations related to medication use in this setting. At the time of their 5-year review, about half of the patients in both arms were receiving treatment with biologics: 54% in the tight-control group and 52% in the standard-care group. In the standard-care arm, that jump over the 5-year time period in the percentage of patients on biologics is worth considering.
Then, looking at combination therapy with csDMARDs, 30% of patients at the end of the study in the tight-control arm were receiving combination therapy compared with 15% in the standard-care group; at 5-year follow-up, those percentages had shrunk markedly to 2% and 7%, respectively. I think the meat of what the investigators were interested in was the question of whether the tight-control strategy offers any lasting benefit with respect to disease activity. As Kevin mentioned, the researchers were using 2 different outcome measures to capture disease activity. At the end of the TICOPA study, they were using MDA, and that was achieved by 50% of patients in the tight-control arm vs 32% receiving standard care. In their 5-year follow-up, the outcome measure used was LDA, as captured by routine care metrics obtained from chart review; 69% of patients in the tight-control group and 76% in the standard-care group had LDA by those criteria. Notably, the investigators did not seem to think these results represented a clinically meaningful difference.
To explore predictors of which patients would be using biologics, the investigators performed a logistic regression analysis. They used current biologic use as the dependent variable and evaluated independent variables, including age, gender, characteristics of the arthritis (oligoarthritis vs polyarthritis), patient global visual analog scale, and the Psoriatic Arthritis Disease Activity Score (PASDAS). Age was the only significant predictor of future biologic use, meaning that younger patients were more likely to be receiving a biologic in the future. However, it was only barely significant; the confidence interval for the odds ratio almost crossed 1.0. It was just a small signal, although I feel that it squares with how we think about biologic use with regard to age, especially in older patients who may have more comorbidities: We may look at risk factors for infection5 or sometimes there is concern about certain malignancies.6
AG: It is good that we are not discussing hypertension, because I think we have to take some of these findings with a lot more than 1 grain of salt! A couple of problems occurred to me. The TICOPA study began recruiting in 2008, and the paper was published in 2015. We need to bear in mind that management of PsA changed between 2008 and 2015, as it has between 2015 and 2022. Another thing — and Kevin, maybe you can comment on this — is that the primary outcome for the TICOPA trial was achievement of an ACR20 response. However, this paper says, in effect, “Pay no attention to the ‘ACR20 behind the curtain,'” and changes the discussion to LDA and MDA, metrics that were largely developed and validated since the original TICOPA study was conducted. Kevin, what are your thoughts on that?
KY: It is like apples and oranges, so it is hard to comment on. Generally, patients did well, but this paper does not inform us about improvements specifically as it pertains to the patient’s disease. It is hard to do a direct comparison, which is a shame.
AG: I think one difficulty is that this is really neither a long-term extension study nor an open-label follow-up as much as it is simply a reassessment of a population 5 years later that, as I said before, asks “Where are they now?” We should keep that in mind when evaluating the findings. Genna, you are very charitable in suggesting that maybe one patient group was different from the other. What did the authors conclude? What was their concern?
GB: The authors seemed to recognize some of the study limitations and they did not see much of a residual benefit, so to speak, from tight control in those patients after 5 years. As you mentioned, the TICOPA study was not constructed as an extension study. It simply describes how treatment of PsA evolved over a period of 5 years.
AG: Treatment, yes, and the evolution of new guidelines — promulgated by people including the authors on this paper — all of which have to be taken into account while analyzing pros and cons, given what we see 5 years later.
We have now looked at the shift that occurred in patients’ medications, and we have talked a bit about some of the paper’s strengths and weaknesses. Just getting a study like this done is an achievement, so kudos to Laura and colleagues. But what impact, if any, will the findings have on daily clinical practice?
GB: What I find interesting is that, especially in RA, we have internalized that the standard of care is treat-to-target and working towards tight control; I do think PsA treatment is very much evolving in that regard. This paper demonstrates that if you let go of that goal, patients settle into a sort of “regression to the mean,” and it helps us recognize that all of this is occurring against the backdrop of a dynamic treatment landscape.
AG: Kevin, what are your thoughts?
KY: I have been debating this for a while. I do not know how much I can glean from this study, except that it shows how treatment practice has changed over the last 5 years. As you mentioned, Dr Gibofsky, this is not an extension study, so there are insufficient data from the 5 years in between completion of the TICOPA study and follow-up to inform us as to whether or not one patient group fared better. We know that 1 of the 2 groups was started on a more intensive therapy at one period of time; in those 5 years in between, however, were biologics discontinued but then reinitiated in some patients?
AG: Right! Suppose I told you there are data which suggest that in the US, use of the treat-to-target paradigm in patients with RA and community practice use of objective metrics to assess disease activity in RA only represents — and Ι will be charitable — about 35% of our colleagues.7 Of course both of you use metrics 100% of the time, because if you dared to present the patient to me without a CDAI score, I would send you scurrying back to obtain it. Nevertheless, if I told you that community use of an objective metric was only about 35% in RA, then what do you think it is in PsA?
KY: Oh, even less. The authors did lament their lack of radiographs or other more objective evidence of disease status. It is a shame because in the original study they did have good support for disease status and they noted the proportion of patients with erosive disease, joint space narrowing, and so on. Yet, what if the authors just caught this cohort of patients during a really good summer’s month, but actually their joints were worse in some way? We would not know because the authors had to rely on what was reported to assess LDA.
AG: Yes, and there is another factor. Kevin, as a clinician coming from Britain, what would you say about the impact of changes in insurance on UK-based studies such as this? I know that the UK’s National Health Service is a “one-size-fits-all” system; however, I would expect that over a 12-year period, there have been changes and new restrictions in terms of what treatments patients are allowed to receive.
KY: Yes, a good example is the National Institute for Clinical Excellence (NICE) guidelines. NICE is a federal government body that effectively decides what the indications are for medications in the UK, and under what circumstances clinicians can prescribe them. Unfortunately, in more niche specialties such as rheumatology, a lot of medications are used off-label. You would have to write to NICE and argue your case, explaining why your patient’s therapy should deviate from the standard treatment algorithm. For example, there might be a phase 3 trial that supports your expectations of the outcome, even if the indication is not established. There is a lot of analysis that looks at quality-adjusted life-years or another outcome measure, given some number of British pounds or US dollars spent, to create a certain outcome when using biologics.
Back when I was in medical school, NICE had a very stringent requirement to treat patients with at least 2 DMARDs before initiating therapy with a biologic. Eventually, NICE had to relent and introduce caveats as to when a patient could receive a biologic. As another example, you can see in Table 1 of the current study that almost no patients were on combination therapy at follow-up because we have found it to be ineffective.
AG: The cost-effectiveness of a drug is a dimension of treatment that we do not generally deal with in our approval process in the US. Under most instances, the US Food and Drug Administration is indifferent as to how much a drug is going to cost. In contrast, the environment in which the current study was conducted involved a 2-step process. You had to go through the medicines agencies, as well as through NICE, which was looking at cost-effectiveness — again, something we do not do in the US. I think this is another factor that we have to consider. For our colleagues who will be reading this Journal Club discussion, this is one of a few significant factors that could impact how extrapolatable a trial such as this one is.
GB: I thought that was interesting. Again, we do not know how or why the patients in this study were being treated with biologics by the time the authors performed their analysis. At the time this follow-up study was published, the UK still required that therapy with at least 2 csDMARDs had to have failed before a patient could receive biologic therapy. It would have been helpful to have had more objective measures available to know, for example, if patients had experienced significant radiographic progression and joint damage from their disease.
KY: I also wonder, because the current study looked at results from only 2 of the 8 original sites, if geographic differences or prescribing differences may have had an impact on patient outcomes. Rheumatology departments are very small divisions and just one attending consultant “going rogue” could really skew your results. Also, the patients in the study had been treated at secondary centers — community centers. That makes me wonder if the demographics would be different if you took into account refractory cases or the most severe cases, which might have been referred to the tertiary centers.
AG: That is an excellent point, Kevin. Another thing to consider is, if we were to look at the data from all 8 sites taken together, would we see the same effect longitudinally? Do you think we would see the same effect in terms of the cost of therapy? Is it possible that patients on tight control ultimately rebounded rather than that patients not on tight control did poorly? We do not know because this is a snapshot rather than a long-term extension study, 5 years afterwards. Kevin and Genna, do you have any final comments?
KY: While I thought this concise study had a lot of potential, it felt much more like a qualitative, reflective practice on what practitioners were doing “on the ground.” I do not know if it is going to change how I approach patient care. Taken at face value, this study is saying that if you use tight control, you do all right, and if you do not use tight control, you still do all right; furthermore, at the end of 5 years, you will get the same result either way. I do not know if that is a conclusion I can reach with data on only 100 patients.
GB: I agree with Kevin. I think that the original TICOPA study was compelling and thought-provoking in terms of practice. I find it a bit harder to draw conclusions from the current follow-up study that would change or inform my practice.
AG: I agree, too. I will just read, for clarity, their conclusion, which I think is very telling. The authors write that “5 years after study end, clinical outcomes and therapeutic drug use were similar, but with considerable changes in patients who had been in the TICOPA study, with no obvious long-term clinical advantage to the 48-week tight-control intervention.” Now, here is the comment that stuck out to me: “Without a formal intention to continue a treat-to-target strategy, this result reflects clinical practice in routine rheumatology care.” Sadly, as we have discussed, in community practice there is not a lot of treat-to-target happening and few metrics are being used. Once these patients had completed the study, they were done with tight control and done with metrology. That is telling.
This has been a fascinating discussion. I hope our readers will appreciate the following takeaway: If you want to do a long-term follow-up study, you should do a long-term follow-up and not take an interval follow-up or “Where are they now?” approach.
This article was edited for clarity and length.
Disclosure
Allan Gibofsky, MD, JD, reported affiliations with AbbVie, Inc.; Acquist Therapeutics, Inc.; Amgen Inc.; Biosplice Therapeutics, Inc.; Boehringer Ingelheim Pharmaceuticals, Inc.; Eli Lilly and Company; Pfizer Inc.; Janssen Therapeutics; and Regeneron Pharmaceuticals, Inc.
References
1. Grigor C, Capell H, Stirling A, et al. Effect of a treatment strategy of tight control for rheumatoid arthritis (the TICORA study): a single-blind randomised controlled trial. Lancet. 2004;364(9430):263-269. doi:10.1016/S0140-6736(04)16676-2
2. Coates LC, Helliwell PS. Validation of minimal disease activity criteria for psoriatic arthritis using interventional trial data. Arthritis Care Res. 2010;62(7):965-969. doi:10.1002/acr.20155
3. Coates LC, Moverley AR, McParland L, et al. Effect of tight control of inflammation in early psoriatic arthritis (TICOPA): a UK multicentre, open-label, randomised controlled trial. Lancet. 2015;386(10012):2489-2498. doi:10.1016/S0140-6736(15)00347-5
4. Coates LC, Mahmood F, Freeston J, Emery P, Conaghan PG, Helliwell PS. Long-term follow-up of patients in the TIght COntrol of inflammation in early Psoriatic Arthritis (TICOPA) trial. Rheumatology. 2020;59(4):807-810. doi:10.1093/rheumatology/kez369
5. Singh JA, Cameron C, Noorbaloochi S, et al. Risk of serious infection in biological treatment of patients with rheumatoid arthritis: a systematic review and meta-analysis. Lancet. 2015;386(9990):258-265. doi:10.1016/S0140-6736(14)61704-9
6. Seror R, Mariette X. Malignancy and the risks of biologic therapies: current status. Rheum Dis Clin North Am. 2017;43(1):43-64. doi:10.1016/j.rdc.2016.09.006
7. Curtis JR, Chen L, Danila MI, Saag KG, Parham KL, Cush JJ. Routine use of quantitative disease activity measurements among US rheumatologists: implications for treat-to-target management strategies in rheumatoid arthritis. J Rheumatol. 2018;45(1):40-44. doi:10.3899/jrheum.170548
Posted by Haymarket’s Clinical Content Hub. The editorial staff of Rheumatology Advisor had no role in this content’s preparation.
Reviewed November 2022
