In this study, we examined the survival prognoses of IPF patients in actual clinical practice and compared the survival prognoses depending on the use of antifibrotics. Patients with IPF were categorized by FVC % predicted into mild: FVC % predicted ≥ 80%, moderate: FVC % predicted < 80% and ≥ 50%, and severe: FVC % predicted > 50% groups, which exhibited significantly different survival periods. Significant differences in survival periods were observed between the mild and severe groups and between the moderate and severe groups, but not between the mild and moderate groups. Further, survival was significantly longer in patients on antifibrotic therapy. This effect was most obvious in the moderate group and a tendency was observed in the mild group, with no effect in the severe group.
Although survival was longest in patients in the mild group, the median survival was approximately 4 years, showing that the survival prognosis may not necessarily be good even in patients with FVC % predicted ≥ 80% and mild respiratory problems. Several other studies have compared survival by stratifying FVC. The Australian IPF Registry (AIPFR) compared the prognosis between FVC % predicted ≥ 80% and < 80% groups, showing significantly longer survival in the FVC % predicted ≥ 80% group [14]. In addition, a Japanese study stratified vital capacity (VC) % predicted in > 80%, 60–80%, 40–60%, and < 40% groups, and found significantly better prognosis in the VC % predicted > 80% group [15]. However, in this study, the median survival of the FVC % predicted ≥ 80% (mild) group was approximately 4 years, which is shorter than 4.5 years or more in the AIPFR and 4.8 years in the Japanese study. One reason for this could be the significant number of deaths due to lung cancer in the mild group. Nine of 40 deaths (22.5%) in the mild group occurred due to lung cancer, significantly more than those in the moderate and severe groups. The two studies mentioned above did not report the number of deaths due to lung cancer, though mortality rate due to lung cancer in the SEPAR National Registry was reported as 9.1% [16] and in a Hokkaido study as approximately 11% [17]. Despite the considerable number of deaths due to unknown causes in our study, deaths due to lung cancer constituted a significant proportion of deaths in the mild group (22.5%), which probably could have affected the survival prognosis. If the cause of death in the unknown cases could be determined, the number of deaths due to lung cancer may be even higher. It is unclear why the mild group exhibited pronounced number of deaths due to lung cancer, though it may be associated with the high percentage of men in this group.
Our results showed that administration of antifibrotics significantly extended the survival of IPF patients. Several reports have indicated that antifibrotics can extend the survival of IPF patients. The European IPF registry and AIPFR showed significantly longer survival in patients treated with antifibrotics [14, 18]. For specific a drug, the EMPIRE registry showed that patients treated with pirfenidone exhibited significantly longer survival [19]. In a joint analysis of the phase 2 and 3 trials of nintedanib, the treatment group showed a significantly lower mortality rate than the placebo group [18]. In addition, long-term observations have also revealed longer survival periods for patients on antifibrotic therapy [20]. Compared to these reports, relatively shorter survivals were observed in patients of our study. One reason could be the high mean age of 74 years in our patients. The mean ages of patients in the EMPIRE registry was 67 years [19], in the European IPF registry was 68 years [18], and in the AIPFR was 70.9 years [14]. However, to the best of our knowledge, this is the first study from Japan and Asia.
Antifibrotics showed a statistically significant effect in the moderate group only. An FVC of 50–80% is consistent with the range suggested as the eligibility criteria for several clinical trials [21], and also with the National Institute for Health and Care Excellence guidance on the indications of antifibrotic agents [22, 23]. Antifibrotic therapy has been shown to be useful for extending the prognosis in patients with 50–80% FVC, which is the most commonly used FVC value for studying the efficacy of antifibrotic therapy. In contrast, antifibrotics did not significantly extend the survival in the mild group. A comparison of the survival curves showed that in the group not treated with antifibrotics, the survival was longer in the mild group than that in the moderate group, which could mean that antifibrotics are less likely to show positive effects for patients in the mild group. However, survival of the group on antifibrotic therapy was clearly longer than the group not on antifibrotic therapy, which suggests that antifibrotics do show an effect. It is possible that a statistically significant difference would have been observed if more cases had been examined. It is interesting that antifibrotics did not show an effect in the severe group. Although the sample size was small, the survival curves almost completely overlapped and only few censored cases were evident, suggesting that the findings are true. As the survival duration of patients in this group was short, the duration of antifibrotic therapy was also short, which could account for the absence of any effect. Moreover, advanced cases are more likely to experience side-effects and suspend the antifibrotic therapy, which may have further shortened the duration of treatment [24]. Nonetheless, it can be safely concluded that effective treatment modalities for severe IPF cases with FVC % predicted < 50% should be developed.
This study had several limitations. First, it was a retrospective study, resulting in a considerable proportion of missing data, such as that on DLco. Second, the observations were censored in many cases due to transfer in different hospitals and/or patients shifting to a different location. In many patients, the cause of death could not be identified. Third, the severe group had a small sample size. While the results of survival analysis for the severe group may have been different if more cases were included, only few censored cases were evident in this group, which indicates that the results are accurate. Finally, antifibrotic therapy was defined as treatment with antifibrotic drugs for at least 6 months, and since we did not consider the commencing time of antifibrotic therapy, some lead time bias is possible. Nonetheless, previous studies have also used a duration of 6 months or more to define antifibrotic therapy, which indicates that our definition is valid [25]. A prospective observational study should be conducted to validate the results of this study. One advantage of this study is that it used data from actual clinical practice and included severe IPF cases with FVC % predicted < 50%, which means it reflected real-world conditions. Severe IPF patients with FVC % predicted < 50% are usually excluded from large phase 3 trials, which does not reflect the real-world conditions.
The results of this study showed that the survival prognosis of IPF patients was poor even in the group with maintained pulmonary function at FVC % predicted ≥ 80%. Even early-stage IPF patients show rapid disease progression and unfavorable prognoses [26, 27], which is similar to the results of our study. Our study showed that antifibrotics may show a positive effect on survival even in patients with FVC % predicted ≥ 80%. This indicates that antifibrotic therapy should be considered even for early-stage cases. In the moderate IPF group with FVC % predicted of 50–80%, which showed a poorer prognosis, antifibrotics showed a significantly positive effect on survival, indicating that treatment with antifibrotics is more important for these patients. In contrast, in the severe IPF group with FVC % predicted < 50%, antifibrotics may show a limited effect on survival. Therefore, early antifibrotic therapy before FVC % predicted drops below 50% is the best treatment protocol. However, each individual case should be considered carefully to determine the most appropriate treatment modality.
Our results showed through a real-world analysis that antifibrotics extended the survival of IPF patients. The effect was obvious in the moderate IPF group with FVC % predicted of 50–80%. A tendency was observed in the mild IPF group with FVC % predicted ≥ 80%, while no effect on survival was observed for the severe IPF group with FVC % predicted < 50%.