ORIGINAL ARTICLE, p 375
The introduction of biologics – particularly tumour necrosis factor (TNF) inhibitors – has been a dramatic step forward in psoriasis treatment. TNF inhibitors are highly effective and may be safer than are methotrexate and ciclosporin, but there is still the potential for adverse events such as infections, lupus-like syndrome and malignancy. It is not clear how best to monitor patients treated with TNF inhibitors for these and other adverse outcomes.
In this issue, van Lümig et al. address this important problem, providing valuable information on the value of laboratory monitoring in 162 patients undergoing TNF treatment for 370 patient-years.1 The major finding in this study is that there wasn’t much found. Routine laboratory monitoring found few abnormalities. Better still, the few abnormal laboratory findings that were identified were rarely of any clinical relevance and did not result in permanent discontinuation of treatment in any patient. Is any laboratory monitoring needed at all during TNF inhibitor treatment?
Stepping back, it may help to ask what data we would need to make firm recommendations for or against routine monitoring. The ideal study would randomize large numbers of patients to two groups, one that would undergo routine monitoring and one that would not. Differences in long-term outcomes of treatment – rates of deaths or other severe adverse events – could be assessed between the two groups. Such a study is not likely to be initiated, much less completed, any time soon.
Failing that, it seems entirely reasonable to monitor for things that have been shown to be risky. Of these, tuberculosis is an obvious choice. While baseline tuberculosis screening is accepted as the standard of care (given reports of tuberculosis activation in unscreened patients treated with TNF inhibitors), the need for annual rescreening is not well established.2 Baseline screening for hepatitis B is probably also warranted.2
As of now, the data do not support clear recommendations for any other routine screening tests. van Lümig et al. did not find any value to monitoring, but there could be rare events that would not be detected without a larger and longer trial. Without more data and careful analysis of the downside risks of monitoring, we’re left with uncertainty. Anything from just baseline tuberculosis and hepatitis B testing at one extreme to complete physical examinations, blood counts and chemistries, urinalyses and chest X-rays every 3 months at the other extreme might be justified.
Is there any reason not to err on the side of doing monitoring? Absolutely. First, there is the direct cost and morbidity of needless tests. Worse yet, tests may identify irrelevant abnormalities or give spurious results that would lead down a road of more testing, cost, and morbidity. As was recently demonstrated with regard to prostate-specific antigen screening, we need to remember it’s a lot easier making a well person sick than it is to make them healthier through routine screening.3
Where does that leave us? Where we usually are, needing to use our best judgment in the care of our patients.