The impact of age and comorbidity on effect of treatment, adverse effects, and quality of life in Danish lung cancer patients receiving immunotherapy

As of 2022, lung cancer is Denmark's most frequent and lethal cancer diagnosis. Most patients are ≥ 70 yearsof age at the time of diagnosis, with incurable disease stage at this time. Previously, palliative chemotherapywas the treatment of choice for these patients. However, it did not improve long-term survival significantlysince most patients died within one year after diagnosis. In 2015, treatment with immunotherapy (immunecheckpoint inhibition) (ICI) was approved in Denmark for subgroups of patients with advanced/metastaticnon-small cell lung cancer (NSCLC). Since then, its use in everyday clinical practice keeps increasing. In therandomized clinical trials (RCTs), which led to the approval of ICI, patients' survival improved significantlycompared to standard chemotherapy regimens alone. However, everyday clinical patients (real-world) maydiffer from those eligible for clinical trials. Real-world patients may be older, with more comorbidity or morewidespread cancer disease such as brain metastasis (BrM). Thus, compared to RCT patients, they maydevelop more or different toxicities, including immune-related adverse events (irAEs) or cancer-relateddiseases such as venous thromboembolism (VTE). Therefore, this thesis focuses on uninvestigated subgroupsof “real-world” patients regarding the effect of ICI and its tolerability.

Overall aim and studies included in this thesis: To investigate effect and tolerability of palliative ICI in realworld patients with advanced/metastatic NSCLC and compare them to RCTs.

Results: Study I: Median age for patients was 66 years (IQR 59-71), 39% had Charlson's comorbidity index score (CCIS) ≥ 2, and 18% had known BrM prior to first ICI. With a median time of follow-up of 15.7 months (range 7.0-40.1), a median ll survival (mOS) of 11.5 months for those who received ICI in ≥ 2 line (HR 2.6, [95% CI: 1.3-5.0], p=0.005) was found versus not reached for those who had ICI in first line. For patients with BrM mOS was 8.2 months (HR 1.38 [95% CI: 0.7-2.5], p=0.37). Discontinuation due to irAEs grade 3-5 happened in 24% of patients. Colitis/diarrhoea and pneumonitis were the most frequent adverse events (AEs) regardless of BrM status. Age and BrM at baseline were not associated with more irAE or impaired OS. 

Study II: MR-C screening found that 28% of patients had BrM at ICI initiation. Of those, 43% were symptomatic. Patients with BrM had a significantly lower median age; most were adenocarcinomas. Of those with BrM who received ICI with or without additional local therapy (radiotherapy or surgery) within the first 6 weeks, 38% and 50% had an intracranial response at first MR-C assessment, respectively. Median followup time was 23.3 months [IQR 16.4-30.2], and mOS for patients with BrM was 15.7 months (95% CI 7.8-24.3) and 22.4 months (95% 16.2-26.3) for patients without. For patients with locally untreated BrM, mOS was 20.5 months (95% 4.9-NR) with a duration of intracranial response of 16.7 months (95% 0.7-NR). Lower programmed death ligand-1 (PD-L1) tumour proportional score (TPS) <50% and bone metastasis but not BrM were statistically significantly negative predictors of both PFS and OS. Baseline quality of life (QoL) was comparable among patients with and without BrM. A significant improvement of QoL at week 9 was found for patients without BrM. 

Study III: With a median follow-up for VTE of 16.5 months (IQR 6.7-35.6), the cumulative VTE at 1, 3, 6, and 12 months were 8%, 10%, 15%, 20% for cohort A and 1%, 4%, 5%, 10% for cohort B. Recurrent VTE comprised 52% and 37% in cohort A and B, respectively. Median OS according to VTE development or not was for the pooled cohort A+B, 9.4 months (95% CI: 5.0-16.0) for patients with VTE and 16.2 months (95% CI: 14.3-18.7) for patients without VTE during ICI. In multivariate analysis adjusting for line of treatment, performance status(PS) 0-1 vs ≥ 2 and PD-L1 expression ≥50% vs < 50%: (HR 1.69 (CI 95 % 1.20-2.38), p=0.003). Risk factors for VTE comprised prior VTE and ICI administered in first line. 

Conclusion

Based on the results from this thesis, important differences between real-world and RCT populations exist. Still, survival in real-world patients treated with ICI has improved compared to survival from prior chemotherapy regimens, especially in the first-line setting. However, OS and PFS for ICI in first line in realworld patients are slightly lower compared to RCTs. Real-world patients have more grade 3-4 irAE grade 3-4, leading to ICI discontinuation since 26-40% of the patients discontinue due to irAE alone or in combination with other reasons. However, the tolerability does seem better compared to prior standard chemotherapy regimens. Higher age or BrM did not seem to affect the effect of ICI or increase risk of irAEs grade 3-5. Moderate to severe CCIS did not impair effect, but comorbidities may contribute to an increased risk of irAE. BrM at the time of ICI initiation is very frequent (~25%), and this group of patients is very heterogeneous with different patterns of responsiveness to ICI. For subgroups of patients with BrM, long-term intra cranial response and a good comparable OS to patients without BrM is possible. Thus, patients with BrM should be included in RCTs in sufficient numbers now and in the future. Increased focus on the risk of VTE during ICI is necessary since the incidence is comparable to or even higher than rates reported with chemotherapy, which has thrombosis-promoting properties. Particular focus should be on patients with prior VTE and patients treated with ICI in first line due to their increased risk of VTE during ICI.