The best magazine
Discussion
Persistence for this investigation was defined as continuous use of the same DMT for the duration of follow-up (i.e., 12 months, 24 months, 36 months) regardless of treatment gaps. Among patients with a minimum of 12 months of follow-up, the persistence rate with GA was 79.5% while for the full cohort of DMT-treated patients the persistence rate at 12 months was 68.3%. This can be compared to the persistence investigation conducted by Reynolds et al. using a different US administrative claims database with MS patient-level data from 1996 to 2005. They found the persistence rate at 12 months for GA was 68.5%, and for all DMTs included in their study it was 65.1%. Differences in persistence rates between the studies may be the result of the time windows used for data extraction, and the sizes of the MS patient samples available for study. This study with a data extraction window of 2001 to 2010 included NZ that was not available for the Reynolds study. This could have resulted in more therapy switching and the corresponding washout periods without DMT that would decrease the persistence rates for all DMT. The data extraction time window used by Reynolds et al. (1996–2005) corresponds to the entry to the market of GA, and IFNβ-1a IM in 1996 and IFNβ-1a SC in 2002. Patients beginning new therapies were in the most vulnerable period for interruptions in therapy and that may be reflected in lower persistence rates than those seen in the later time window of this study. Reynolds et al. included 6,134 MS patients in their investigation, while this study had 35,312 MS patients available for the 12-month cohort. The characteristics of the patient samples in terms of age, gender and geographic distribution were very similar.
When patients with a minimum of two years of follow-up were examined for this study, the persistence rate for GA increased slightly to 80.5% and the corresponding rate for the full DMT-treated cohort fell to 53.9%. With 36 months of follow-up, the persistence rate for GA increased modestly to 81.2% and the persistence rate for the full DMT-treated cohort rose to 70.1%. Relative to the 12-month cohort, the persistence rates for GA remained stable for 24 and 36 months with modest increases in therapy switching and decreases in discontinuation. For the full DMT-treated cohort there was greater fluctuation in persistence rates over the three years. The proportion switching therapies increased with duration of follow-up while the proportion discontinuing DMT increased from 12 months to 24 months but declined after 36 months. This may reflect the introduction NZ in 2006 and the fact that patients needed to stop all DMT for several months for a washout period before they initiated NZ therapy. Results from this investigation do show some similarity to those of other investigators. Margolis et al. followed a cohort of newly diagnosed MS patients for an average of 35.7 ± 17.5 months; those who were treated with DMT had a persistence rate of 72.3%. Rio et al. reported that 80.0% of their DMT users continued on the same therapy without interruption after a mean follow-up of 47 months, and Tremlett et al. reported that at the end of study follow-up (mean 2.4 years), the persistence rate for GA was 80.0%.
Gaps in MS therapy are commonplace and this study shows that they increase with the duration of all DMT. However, the proportion of GA users who re-initiated therapy following gaps in therapy increased over time from 64.4% at 12 months to 80.1% at 36 months. This was in contrast to the full DMT-treated cohort that maintained a therapy re-initiation rate of 50–60% over 36 months. The lower re-initiation rate may be attributed to some patients with more highly active disease who were hospitalized for a relapse, treated with corticosteroids, or taken off their DMT to begin another therapy. Regardless of DMT, the relatively high percentage of patients re-initiating the same therapy after a therapy gap may indicate that these patients are experiencing more short-term issues such as forgetfulness, failure to re-order medication, a problem with reimbursement or the desire to take a drug holiday with or without a physician's approval.
The consequences of medication gaps have begun to be reported and gaps have been shown to be associated with an increased risk of MS relapse. Tremlett et al. reported that therapy gaps were associated with a shorter time to first on-study relapse and trend towards future disease progression when compared to patients without missed doses.
Therapy regimen changes such as switching medication and stopping DMT interfere with persistence. In this study the rate of therapy switching increased over time; 4.3% of the GA cohort switched therapies in the first 12 months, 6.3% in 24 months and 8.0% over 36 months. This trend was consistent with the full DMT-treated cohort though the switching proportions for the GA cohort were somewhat lower than those for the overall sample: 10% over 12 months, 16% in 24 months and 20% over 36 months. This may reflect the results of Reynolds et al. who reported the lowest switch rates for GA relative to IFNβ. When compared to other published results of therapy switching in MS, Margolis et al. reported more than twice as much DMT switching (21.3%) at 12 months as the 9.0% reported by Reynolds et al.. and the 10.0% reported in this study. However, for GA users specifically, Reynolds et al. reported somewhat higher switch rates than those seen with this study: 3.9% had switched to another DMT in the first 6 months, 6.4% had switched over 12 months, and 8.3% switched over 18 months. This may be related to the earlier window of data extraction used by Reynolds et al. corresponding to the market introduction of GA.
Unlike therapy switching that increased with duration of therapy, the proportion discontinuing declined with duration of therapy in this study. This is consistent with the findings of others who have noted that the most vulnerable time for therapy discontinuation is the first six months of therapy. There is evidence to suggest that the timing of therapy discontinuation is earlier when the reason is an adverse drug event and later when the reason for stopping is a perceived lack of therapy effect. Several investigators have explored the reasons for therapy discontinuation and they include factors related to the disease such as the type of MS, the level of disability or physician-documented disease progression, adverse effects, patients' perceptions of therapy ineffectiveness, the presence of cognitive dysfunction, and/or depression and therapy cost.
Source: ...