The effect of adding immune checkpoint inhibitors on the risk of pneumonitis for solid tumours

Editor’s pick of Sylvie Rottey, MD, PhD, medical oncologist, UZ Gent

Treatment with immune checkpoint inhibitors (ICIs) for solid tumours can result in immune-related adverse events, including pneumonitis. Previous meta-analyses report differing increases in incidence of pneumonitis with the addition of ICIs to conventional therapy. Now, a large meta-analysis of 25 phase III randomised controlled trials concludes that the addition of ICIs to conventional treatment significantly increases the risk of both grade I-V and grade III-V pneumonitis, irrespective of the mechanisms of different ICIs and cancer types.

The incidence of severe adverse events and treatment discontinuation in ICI monotherapy is known to be lower than that in conventional chemotherapy. However, ICIs can produce immune-related adverse events (irAEs), in particular pneumonitis. Confusing matters further, previous meta-analyses have reported different rates and reasons for the occurrence of pneumonitis. These analyses dispute the role that factors such as cancer histology and ICI mechanisms of action have on the incidence of pneumonitis. In light of these previous, contrasting analyses, a meta-analysis now reports the impact of adding ICIs to conventional therapy on the incidence and degree of pneumonitis, amongst patients with solid tumours. A total of 25 randomised controlled trials (RCTs) were identified, representing 16,343 patients with grade I-V pneumonitis, and 23 RCTs of 15,006 patients with grade III-V pneumonitis were evaluated. These RCTs were all phase III studies, and had compared ICIs and conventional therapy to conventional therapy alone.

Significant increase of pneumonitis with the addition of ICIs

A total of 469 events of grade I-V pneumonitis were observed in 16,343 patients from 25 RCTs, and 138 events of grade III-V pneumonitis were observed in 15,006 patients across 23 RCTs. When compared to their respective control arms, the addition of ICIs significantly increased the risk of developing both grade I-V pneumonitis (OR[95%CI]: 2.67[2.12-3.37], P< 0.00001) and grade III-V pneumonitis (OR[95%CI]: 1.83[1.26-2.65], P= 0.001). When comparing the risk of pneumonitis based on ICI type (CTLA-4, PD-1 and PD-L1 inhibitors), each of them significantly increased the risk of grade I-V pneumonitis, with no differences observed among the groups (I²= 0%, P for heterogeneity= 0.89). Conversely, CTLA-4 and PD-1 inhibitors had a significant increase in risk of grade III-V pneumonitis, whilst PD-L1 inhibitors did not. However, no significant difference was observed between each group (I²= 0%, P for heterogeneity= 0.72). Comparing the incidence of pneumonitis between oncology patients with lung cancer and non-lung cancer, no significant difference was observed between the two groups for grade I-V or grade III-V pneumonitis (grade I-V: I²= 0%, P for heterogeneity= 0.74. Grade III-V: I²= 25.9%, P for heterogeneity= 0.25).

The risk of developing pneumonitis with ipilimumab, tremelimumab, nivolumab, pembrolizumab, atezolizumab, avelumab and durvalumab was also evaluated, with no difference observed in the incidence of either grade I-V or grade III-V events (grade I-V: I²= 0%, P for heterogeneity= 0.97. Grade III-V: I²= 0%, P for heterogeneity= 0.97). Subgroup analysis of ICI plus chemotherapy included 14 RCTs, representing 8,933 patients with grade I-V pneumonitis and 12 RCTs with 7,596 patients for grade III-V pneumonitis. Compared to a control arm of chemotherapy alone, the addition of ICIs to chemotherapy significantly increased the incidence of grade I-V pneumonitis (OR[95%CI]: 2.43[1.71-3.46], P< 0.00001). There was also tendency towards an increase in the incidence of grade III-V pneumonitis (OR[95%CI]: 1.71[1.00-2.93], P= 0.05). The test for heterogeneity was not significant in either grade categories (grade I-V: I²= 23%, P for heterogeneity= 0.20. Grade III-V: I²= 0%, P for heterogeneity= 0.88).

CONCLUSION

Contrary to previous meta-analyses, the addition of ICIs to conventional treatment for solid tumours, such as chemotherapy, significantly increased the risk of both grade I-V and grade III-V pneumonitis, irrespective of the mechanisms of ICIs, individual ICIs and cancer types. These results can be used to dictate management decisions in clinical practice, so that the most appropriate treatment is selected on an individual basis. In particular, these results will help guide decision-making for patients with known risk factors for pneumonitis or interstitial lung disease.

Reference

Fujiwara Y, Horita N, Namkoong H, et al. The effect of adding immune checkpoint inhibitors on the risk of pneumonitis for solid tumours: a meta-analysis of phase III randomised controlled trials. Eur J Cancer. 2021; 150(2021):168-178.