Start Slow and Step Up or Hit Hard, Step Down? Finding the Right Initial Therapy for Chronic Obstructive Pulmonary Disease

Chronic obstructive pulmonary disease (COPD) is the third leading cause of mortality worldwide, accounting for 6% of deaths (1). Global incidence rates of this preventable, common, and treatable disease continue to increase with continued exposure to risk factors (2). Much has been learned about the disease and its treatments since the initial Global Initiative for Chronic Obstructive Lung Disease (GOLD) report was published in 2001 with the goal of treatment to improve symptoms and reduce future risk of exacerbations (3). Data suggest that pharmacotherapy cannot only reduce exacerbations and improve symptoms but can also decrease the rates of lung decline and improve mortality rates (4).

Current GOLD guidelines recommend that patients with a high risk of exacerbations, defined as at least two moderate or at least one severe exacerbation in the previous 12 months (GOLD group E), receive initial dual therapy with a long-acting β-agonist (LABA) and a long-acting muscarinic antagonist (LAMA) (5). The addition of an inhaled corticosteroid (ICS) is reserved for eosinophilic phenotypes, patients with concomitant asthma, or continued exacerbations despite treatment with technique-optimized LABA/LAMA dual therapy. This “start low, step up” strategy of intensifying treatment with the addition of an ICS only if initial dual treatment is ineffective is based in part on evidence associating ICS use with alterations in the airway microbiome and a higher risk of pneumonia (6, 7). Additionally, observational studies suggest that ICS use might also be associated with increased risk of uncontrolled diabetes, cataracts, osteoporosis, and Mycobacterium infection (8).

In this issue of AnnalsATS, Mountain and colleagues (pp. 1139–1146) performed a benefit–harm analysis of starting triple therapy with LAMA/LABA and an ICS as first-line treatment compared with starting guideline-recommended LAMA/LAMA dual therapy and “stepping up” if needed. The target population was Canadian patients aged at least 40 years with COPD treated with a single inhaler agent and at high risk of acute exacerbations (9). The EPIC (Evaluation Platform in COPD) model, a validated whole-disease microsimulation, used inputs for efficacy that were derived from published literature using the SHINE trial and the IMPACT (Informing the Pathway of COPD Treatment) trial (10–12). The input for harm was the risk of pneumonia based on a meta-analysis of triple therapy in the management of COPD (13). The authors found that early initiation of triple therapy was associated with a gain of 4.8 quality-adjusted life-years (QALYs) per 100 patients older than 20 years with COPD compared with standard care. This benefit was robust to various subgroup and sensitivity analyses, with the net QALY gain increasing to 5.9 per 100 patients in the subgroup with a high symptom burden.

Notable strengths of this study include its rigorous use of a validated, open-sourced, policy-focused population-based and dynamic outcome model developed specifically to study Canadian patients with COPD and the use of QALYs to place benefits and harms on the same scale. Limitations of this study include potential oversimplification of the harms associated with ICS use due to limited data on the risks of harms other than pneumonia. It is important to note that benefit–harm analyses like the one by Mountain and colleagues tend to be biased toward benefit because harms are often underreported in trials (9). Additionally, the EPIC model does not include eosinophil burden even though it is an important factor in real-world decision-making for COPD treatment.

How will this study impact practice? Importantly, the results of this study inform the evolving choice between a start low, step up approach and a “hit hard, step down” approach for patients with COPD in GOLD group E. Like COPD, many medical conditions have differing levels of treatment intensity that can be adjusted based on treatment response. The decision to choose to initiate a lower-intensity treatment and step up rather than start a higher-intensity treatment and step down depends on several factors summarized in Table 1. First, the consequences of failure of the initial treatment must be considered. Medical conditions for which treatment failures are minimally consequential favor a step-up approach in treatment, with additional therapies added only if the initial treatments are ineffective. Diseases in which treatment failures have more dire consequences, however, favor starting with a more intense therapy and deescalating if appropriate. Second, the harm associated with the more intense treatment must be considered: when treatments present a high risk of harm, they should generally be reserved for patients in whom it has been proven that an intensive treatment is required to control the disease. Third, the choice of a step-up versus a step-down approach depends on cost. In situations in which a higher-intensity treatment is markedly more expensive, the burden of the cost from patient and societal perspectives may favor the step-up approach. Fourth, patient preference and adherence must be considered. In general, patients prefer less intensive, simpler treatment. Finally, the success of a step-up or step-down approach depends on the existence of reliable, responsive, clinically relevant markers of treatment failure or disease stability to guide decisions.

Table 1.

Factors to consider in choosing “start low, step up” versus “hit hard, step down” and their application to COPD treatment

Factors to Consider	Favors Step-Up Approach	Favors Step-Down Approach	Application to COPD
Consequences of treatment failure	Treatment failures are minimally consequential	Treatment failures have severe or long-lasting health effects	Exacerbations have profound clinical implications for patients
Harms of more intensive treatment	More intensive treatment associated with harm	More intensive treatments have minimal harms	ICS use associated with increased risk of pneumonia
Cost	Less intensive treatment costs less	More intensive treatment has similar cost or shown to be cost-effective	Triple therapy more expensive, but no data on cost effectiveness
Patient preference	Patients prefer less intensive, simpler treatment	Demonstrated acceptability of more intensive treatment	No data, but expect less impact with single inhaler delivery
Maintenance decision-making	Reliable measures to indicate treatment failure and guide step-up	Reliable, clinically relevant measures to indicate disease stability and guide step-down	No robust markers to guide step-up or step-down

Definition of abbreviation: COPD = chronic obstructive pulmonary disease; ICS = inhaled corticosteroid.

Mountain and colleagues provide data to suggest that the benefits from the early addition of an ICS in avoiding future exacerbations outweigh the harm of adverse events from pneumonia, favoring the hit hard, step down approach to the treatment of COPD (9). We currently lack data about other factors such as cost effectiveness and patient preference to help inform the decision. Importantly, there are no robust markers to accurately guide stepping up or stepping down the intensity of treatment for COPD. GOLD recommends that patients with COPD with well-controlled dyspnea and no exacerbations in the previous year may trial withdrawal of the ICS, particularly patients with complications from the ICS or those who are less likely to benefit from the ICS, but there are no robust data to support this approach, and trials of this strategy have produced mixed results (5, 14).

Ultimately, the optimal treatment strategy for COPD remains uncertain because there is still much to learn about the biological and molecular pathways of the disease. It is reasonable to assume that “hitting hard” early in the disease and thus aggressively suppressing pathologic pathways may result in better long-term outcomes. A future clinical trial comparing early aggressive treatment with triple inhaler therapy versus a step-up approach in COPD would be welcome, like the recent PROFILE (PRedicting Outcomes For Crohn’s disease using a moLecular biomarker) trial that found that early and aggressive immunosuppressive therapy with TNF (tumor necrosis factor) inhibitors significantly outperformed conventional therapy for patients with newly diagnosed Crohn’s disease (15). In the mean time, this well-conducted simulation-based study by Mountain and colleagues provides some initial momentum for a hit-hard approach.