Abstract
Immunotherapy based on immune checkpoint inhibitors (ICIs) represents a novel anticancer treatment strategy. Monoclonal antibodies targeting cytotoxic T-lymphocyte antigen-4 (CTLA4), programmed cell death-1 receptor (PD1) and programmed cell death-1 ligand (PD-L1) have shown efficacy and safety in the treatment of various malignancies. Some of them have recently found their place in a routine clinical practice, while others are at different phases of clinical trials. Treatment with ICIs may be accompanied by undesirable impairment of immunotolerance to non-tumoural tissues, leading to a specific side-effect also called immune-related adverse events (irAE). There is an increasing body of evidence that the development of irAEs is associated with a beneficial effect of immunotherapy, thus it has become a hot topic in the field of clinical oncology. This review is focused on data from recently published studies evaluating the association between irAEs and outcome of patients with cancer treated with ICIs.
Immunotherapy based on the blocking of immune checkpoints represents an innovative anticancer treatment strategy that has been currently changing the world of clinical oncology. There are several agents that have shown efficacy and safety in the treatment of various malignancies. Some of them have recently found their place in a routine clinical practice, while others are at different phases of clinical trials. Among the malignancies where ICIs have been well established and have already been widely used in the routine clinical practice include malignant melanoma (MM), lung cancer and renal cell carcinoma (RCC). Furthermore, ICIs show efficacy and have been approved for the treatment of urothelial carcinoma, breast, head and neck squamous cell cancer, gastric cancer, hepatocellular cancer, colorectal cancer with microsatellite instability, Hodgkin lymphoma and Merkel cell carcinoma (1-12).
The most frequently used immune checkpoint inhibitors (ICIs) are monoclonal antibodies against cytotoxic T-lymphocyte antigen-4 (CTLA4), programmed cell death-1 receptor (PD1) and programmed cell death-1 ligand (PD-L1). The most commonly studied CTLA4 inhibitor is ipilimumab; among PD1 inhibitors, nivolumab, pembrolizumab and avelumab; among PD-L1 inhibitors, atezolizumab and durvalumab.
In contrast to chemotherapy or targeted therapies, immunotherapy has several specific characteristics in terms of the clinical response and side-effects. Overcoming tumour immunotolerance induced by ICIs may be accompanied by undesirable impairment of immunotolerance to non-tumoural tissues. The consequence of this is the emergence of specific side-effects also called immune-related adverse events (irAE). The nature of irAEs is similar to an autoimmune disease. These are to a large extent typical for immunotherapy. In general, irAEs can be divided into organ-specific and organ non-specific; according to the timing of onset they can be divided into early and late. IrAEs can result in damage of any organ, including the skin, intestines, liver, kidneys, lungs, eyes, endocrine glands, heart, muscle, and central or peripheral nervous system (13-15). The most commonly seen are dermatological and gastrointestinal irAEs.
There is an increasing body of evidence that the development of irAEs is associated with a beneficial effect of immunotherapy (16). This association has been described so far in malignancies for which there is the greatest experience with immunotherapy, including MM, non-small cell lung cancer (NSCLC) and RCC.
IrAEs and Outcome in MM
The association between irAEs and the efficacy of ICIs was first reported in patients with advanced-stage MM. The published data on treatment with CTLA4 (ipilimumab) and PD1 (nivolumab, pembrolizumab) inhibitors are not entirely consistent.
The first retrospective analysis of patients with metastatic MM (n=56) treated with ipilimumab which was conducted by Attia et al. showed that 36% of patients with grade 3-4 irAEs reached an objective response to the therapy (ORR), whereas among patients without or with a low-grade irAEs, only 5% had a treatment response (p=0.008) (17). These results were subsequently confirmed by a combined analysis of the data obtained from two prospective clinical trials (n=139), which showed that irAEs of any grade in patients with metastatic MM treated with ipilimumab were significantly associated with ORR (p=0.0004) and a significantly longer duration of response was observed in patients with grade 3-4 irAEs as compared to those with low-grade irAEs (34 vs. 11 months) (18). Nevertheless, the study showed that the duration of treatment response was not adversely affected by the use of high doses of corticosteroids to treat irAEs (p=0.23) (18). On the other hand, two retrospective analyses of larger patient cohorts treated with ipilimumab (n=833 and n=298, respectively) did not show any significant association between irAEs and outcome (19, 20). However, the retrospective design of these studies potentially conferring a bias in the form of underestimation and grade underscoring of irAEs should be emphasized (19, 20).
Several studies have shown the impact of specific types of irAEs. A significant association between development of vitiligo and survival was observed in a meta-analysis of published studies in patients with advanced MM treated with various immunotherapy strategies, which was conducted by Teulings et al. (21). Their results showed that vitiligo was associated with longer progression-free survival (PFS) and overall survival (OS) [hazard ratio (HR)=0.51, p=0.005 and HR=0.25, p=0.003, respectively] (21). Similar results have been reported in patients with metastatic MM treated with high-dose interferon or high-dose interleukin-2 (22, 23). Endocrine irAEs, most frequently in the form of hypopituitarism or hypothyroidism, represent another specific type of irAE whose incidence is described in association with a beneficial effect of immunotherapy with ICIs. Fujisawa et al. observed a significant association of endocrine irAEs with OS in patients with unresectable or metastatic MM (n=60) treated with ipilimumab (HR=0.22, p=0.015) (24). A pooled analysis of safety data of nivolumab obtained from four prospective clinical trials in patients with advanced MM (n=576) showed a significantly higher ORR in patients with irAEs of any grade (48.6 vs. 17.8%, p<0.0001) (25). Similarly, in a small retrospective study including patients with advanced MM treated with nivolumab (n=15), Okada et al. reported a higher disease control rate (DCR) (75% vs. 14%, p<0.05) and longer OS (p<0.05) in patients with any irAE compared to those without (26). A retrospective study focusing on the role of irAEs in patients with resected and unresectable MM (n=148) treated with nivolumab was conducted by Freeman-Keller et al. (27). They found a significantly longer OS in patients who experienced an irAE versus those who did not (p<0.001), while the OS benefit was noted in patients who reported three or more irAE events (p<0.001) (27). Moreover, a 12-week landmark survival analysis closely focusing on the role of specific type of irAEs found that only skin rash and vitiligo were significantly associated with OS (p=0.0004 and p=0.028), while no statistically significant differences in OS were seen in patients with any grade of hypothyroidism (p=0.117), hyperthyroidism (p=0.489), diarrhoea (p=0.132), or pneumonitis (p=0.493) (27). Another retrospective study by Maeda et al. also recorded a significant association of endocrine irAEs with DCR (84.6% vs. 37.5%, p=0.004) in patients treated with nivolumab for metastatic MM (n=69); there was also a trend for longer PFS in the 20-week landmark analysis, however, it was not statistically significant (p=0.07) (28). The data published on the linkage between irAEs and outcome in patients treated with pembrolizumab are still limited, however, several studies suggest similar results to those obtained from studies with nivolumab. The retrospective study conducted by Indini et al. including patients with metastatic MM (n=173) treated with pembrolizumab or nivolumab found a higher DCR (p=0.029) and longer PFS and OS in patients with any irAE in the multivariate Cox regression analysis (HR=0.47, p=0.016; HR=0.39, p=0.007) (29). Analysis of prospectively collected data from patients with advanced MM treated with pembrolizumab (n=147) found a higher DCR in patients who developed any irAE (75.0 vs. 17.9%, p=0.02) and the best results were achieved in patients with higher grade irAEs, in whom a significantly longer PFS and OS were found compared to patients without irAEs (HR=0.54, p<0.001; HR=0.51, p<0.001) (30). There was no significant association between the use of corticosteroids and PFS nor OS (p=0.8 and p=0.6, respectively) (30). The results of selected studies on patients with MM are summarised in Table I.
Summary of selected studies in patients with malignant melanoma.
IrAEs and Outcome in NSCLC
In patients with advanced and metastatic NSCLC, the association of irAEs with the effect of immunotherapy has been documented particularly in treatment with anti-PD1 and anti-PD-L1 inhibitors, which are commonly used in advanced NSCLC.
The association of the effect of nivolumab with irAE in patients with metastatic NSCLC (n=40) was first reported by Hasan Ali et al., who observed the association of treatment response with skin toxicity (31). Skin irAEs were associated with ORR (42% in patients with skin irAEs vs. 7% in those without) (31). Sato et al. observed that irAEs were associated with longer PFS (HR=0.10, p<0.001) in patients with advanced NSCLC treated with nivolumab (n=38), however, subsequent 60-day landmark analysis did not show any significant difference (HR=0.28, p=0.13) (32). Longer OS in patients with metastatic NSCLC (n=91) who developed any irAEs during the course of nivolumab treatment as compared to those with no irAEs (HR=2.75, p<0.001) was reported by Owen et al. (33). They did not observe significant differences in a 3-month landmark analysis, which was similar to the study by Sato et al. (32). Another retrospective analysis of patients with advanced NSCLC treated with nivolumab (n=134) showed higher ORR (52.3% vs. 27.9%, p=0.02) in patients with any irAE; moreover, in a Cox multivariable analysis, irAEs remained a significant factor for longer PFS (HR=0.542, p=0.04) as well as for OS (HR=0.285, p=0.003) (34). Regarding the specific type of irAE, skin irAEs were significantly associated with longer PFS (HR=0.476, p=0.03) and OS (HR=0.209, p=0.003), and endocrine irAEs were significantly associated with longer PFS (HR=0.237, p=0.02) (34).
The prognostic role of irAEs developing early after initiation of nivolumab treatment was demonstrated in the analysis of prospective data of patients with advanced NSCLC (n=43) conducted by Teraoka et al. (35). They found a significant correlation of longer PFS with any irAE (p=0.01) and irAE in the form of skin rash (p=0.01) in a 2-week landmark analysis (35). By shifting the boundary of the landmark analysis to 6 weeks, the differences in PFS were no longer significant but there was a significant trend for both parameters (p=0.06 and p=0.08); the shift beyond statistical significance might have been affected by a small group of patients (35). Subsequently, two larger retrospective studies were published. The study conducted by Ricciuti et al., including patients with advanced NSCLC treated with nivolumab (n=195), showed significantly higher ORR (43.5% vs. 10.0%, p<0.0001) and DCR (70.5% vs. 18.1%, p<0.0001), as well as longer PFS and OS (HR=0.41, p<0.0001 and HR=0.33, p<0.0001) in patients with irAEs compared with patients without any (36). The results of the Cox multivariate model showed the following factors to be significant for PFS: any irAE (HR=0.48, p<0.0001), pulmonary (HR=0.56, p=0.038), gastrointestinal (HR=0.055, p=0.021), endocrine (HR=0.59, p=0.011) and skin irAEs (HR=0.57, p=0.031); significant factors for OS were: any irAE (HR=0.38, p<0.0001), pulmonary (HR=0.46, p=0.022), gastrointestinal (HR=0.50, p=0.045), endocrine (HR=0.45, p=0.001) and skin irAEs (HR=0.80, p=0.043) (36). A 6-weeks landmark analysis showed a significant association of irAEs with OS (HR=0.55, p=0.021) and a 12-weeks landmark analysis showed an even deeper and statistically significant association with PFS (HR=0.48, p<0.0001) and OS (HR=0.40, p<0.0001) (36). The results of the study also showed longer PFS and OS in patients with two or more types of irAEs versus those with only one versus those without any irAE (PFS: 8.5 vs. 4.6 vs. 2.0 months, p<0.0001; OS: 26.8 vs. 11.9 vs. 4.0 months, p<0.0001) (36).
The association of the incidence of irAEs with the effect of anti-PD1 (nivolumab, pembrolizumab) and anti-PD-L1 (atezolizumab) inhibitors in patients with advanced NSCLC (n=270) was recently confirmed by a retrospective study by conducted by Grangeon et al. (37). The results of this study showed a significant association between the incidence of irAEs and higher ORR (22.9% vs. 5.7%, p<0.0001), DCR (72.4% vs. 36.7% p<0.001) as well as longer PFS (HR=0.42, p<0.001) and OS (HR=0.29, p<0.001) (37). The study confirmed a significant association of irAE in a form of hypothyroidism with longer PFS (HR=0.56, p=0.005) and OS (HR=0.46, p=0.01) and did not show significant differences according to the grade of irAEs (37). The association of immune-related thyroid dysfunction with effect of ICIs has been observed also in two other studies by Osorio et al. (38) and Toi et al. (39). Moreover, both studies reported that the development of thyroid dysfunction was significantly higher in patients with pre-existing antithyroid antibodies than in those without (80% vs. 8%, p<0.0001; and 40% vs. 2%, p<0.001, respectively) (38, 39).
The largest study focused on the association of irAEs with ICI treatment efficacy to date was recently conducted by Cortellini et al. (40). Their retrospective study included 559 patients with advanced NSCLC treated with nivolumab or pembrolizumab (40). The results of a multivariate analysis showed a higher ORR was related to experiencing any irAE (p<0.0001), and endocrine (p=0.0043) and skin (p=0.0005) irAEs; significantly longer PFS was associated with irAEs of any grade (p<0.0001), endocrine (p=0.0084) and skin (p=0.0001); and significantly longer OS was associated with irAEs of any grade (p<0.0001), endocrine (p=0.0044), gastrointestinal (p=0.0437), and skin (p=0.0006) irAEs. The 6-weeks landmark analysis revealed irAEs of any grade as an independent predictive factor for higher ORR, and longer PFS and OS (40). The results of selected studies on patients with NSCLC are summarised in Table II.
IrAEs and Outcome in RCC
As compared to MM and NSCLC, the data on the relationship between irAEs and outcome of patients with advanced RCC treated with ICIs are limited to several recently published retrospective studies. The study conducted by Ishihara et al. (n=47) showed significantly higher ORR (60.9% vs. 12.5%, p=0.0006), and longer PFS (p<0.0001) and OS (p=0.0072) in patients with irAEs as compared to those without (41). The Cox multivariate model confirmed the occurrence of irAEs as an independent factor for better PFS (HR=0.25, p=0.0009); OS analysis was not performed (41). A landmark analysis after two cycles of treatment (i.e. after 1 month of the treatment) showed significant association only for PFS (p=0.0279); for OS the difference was not statistically significant (p=0.193). Regarding the grade of irAEs, PFS was significantly longer both in patients with grade 1-2 irAEs (p=0.0024) and those with grade 3 or more irAEs (p=0.0023) as compared to patients without irAEs; for OS, there was significant correlation only in patients with grade 1-2 irAEs (p=0.0124). With regard to the specific type of irAEs, a significant association with PFS and OS was demonstrated only for skin irAEs (p=0.011) (41). The results of a recently published retrospective study (n=389) conducted by Verzoni et al. confirmed a significant association of irAEs with OS (p=0.01); furthermore, irAEs remained an independent factor for OS in a multivariable Cox model (HR=0.57, p=0.02) (42). A 6-week OS landmark analysis showed significant association with irAEs (p=0.006). The results of a 1-year OS analysis found no statistically significant difference between those with grade 1-2 and those with grade 3-4 irAEs (60.9% vs. 79.6%, p=0.19), nor between those with early (occurring within 6 weeks from the treatment initiation) and those with late (occurring over 6 weeks from the treatment initiation) irAEs (78.7% vs. 85.2%, p=0.34) (42); neither ORR nor PFS were evaluated in the study. A development of vitiligo was recently reported in a case of metastatic RCC which achieved a durable complete response after treatment with nivolumab, suggesting this type of irAE to be a factor associated with the efficacy of ICIs in RCC, similarly to that in MM and NSCLC (43).
Summary of selected studies in patients with non-small cell lung cancer.
IrAEs and Outcome in Other Malignancies
Gastrointestinal cancer. The relationship between irAEs and the efficacy of anti-PD1 therapy in patients (n=61) with various types of advanced gastrointestinal cancer including hepatocellular carcinoma, colorectal cancer with microsatellite instability and gastric cancer was assessed in a retrospective study (44). The study found significantly longer PFS and OS in patients experiencing irAEs as compared to those who not (p=0.0001 and p=0.0036) (44). A sub-analysis found no statistically significant differences in PFS nor OS according to the grade, timing of onset or use of corticosteroid therapy (44).
A retrospective study focusing on patients with advanced gastric cancer treated with nivolumab (n=65) conducted by Masuda et al. found significantly longer PFS and OS in patients with irAEs compared to those without (p<0.001 and p<0.001, respectively) (45). Furthermore, the results of multivariable Cox model demonstrated that absence of irAEs represented an independent factor associated with a poor prognosis (HR=9.54, p<0.001) (45).
Head and neck cancer. In patients with metastatic head and neck cancer receiving anti-PD1 therapy (n=114), a retrospective study conducted by Foster et al. found higher ORR (p=0.02) and also longer PFS and OS in patients with irAEs compared to those without (p=0.0004 and p=0.007, respectively) (46). The results of multivariate analyses shown that the development of irAEs was independent factor for improved ORR (p=0.03), PFS (p=0.009), and OS (p=0.003) (46).
Urothelial cancer. In patients with metastatic urothelial cancer, Morales-Barrera et al. reported higher DCR (79% vs. 36.3%, p=0.002) and longer OS (21.91 vs. 6.47 months, p=0.004) for those with irAEs as compared to those without in a single-institution retrospective study (n=52) (47). A pooled analysis of seven randomised clinical trials including cisplatin-refractory or cisplatin-ineligible patients with locally advanced or metastatic urothelial cancer treated with pembrolizumab or atezolizumab (n=1,747) was conducted recently by Maher et al. (48). The study focused on the role of irAEs and adverse events of special interest, which were defined separately from irAEs as autoimmune events not requiring corticosteroid therapy (48). They found longer OS in patients with adverse events of special interest (HR=0.45) and those with irAEs (HR=0.53) as compared to those without (48). The duration of response was not affected significantly by the use of corticosteroids in management of irAEs (48).
Conclusion
There is increasing evidence suggesting that irAEs are associated with treatment response and survival of patients with cancer treated with ICIs. This could be used for the estimation of the prognosis of patients treated with ICIs, suggesting development of irAEs as a feasible on-treatment predictive biomarker. Moreover, this can be helpful information for distinguishing between real progression and pseudoprogression. Most of the published studies are recent, thus it is a hot topic in the field of clinical oncology. However, the vast majority of studies published to date are limited by the relatively low number of patients and retrospective design introducing possible bias. There remain many open questions. The detailed pathophysiological mechanisms underlying various irAEs, as well as the association of specific types of irAEs with the efficacy of ICIs in various cancer types, should be elucidated in the future, optimally under the conditions of prospective clinical trials.
Acknowledgements
This study was supported by the National Sustainability Program I (NPU I) Nr. LO1503 provided by the Ministry of Education Youth and Sports of the Czech Republic and by the Charles University Research Fund (Progres Q39) and by the European Regional Development Fund Project “Application of Modern Technologies in Medicine and Industry” (No. CZ.02.1.01/0.0/0.0/17_048/0007280) and by the grant of Ministry of Health of the Czech Republic - 17-30748A.
Footnotes
Authors' Contributions
OF wrote the article with support from OS, JS, RK, OT and JF.
This article is freely accessible online.
Conflicts of Interest
OF received honoraria from Roche, GSK and Pfizer for consultations and lectures unrelated to this project. JF has received honoraria from Astra Zeneca, Roche and Novartis for consultations and lectures unrelated to this project. OS, JS, RK and OT declare that they have no actual or potential conflict of interest including any financial, personal or other relationships with other people or organizations that could inappropriately influence this work.
- Received January 12, 2020.
- Revision received January 21, 2020.
- Accepted January 24, 2020.
- Copyright© 2020, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved
References
In this issue
Jump to section
Related Articles
Cited By...
- Use of Conditioned Extracellular Matrix as a Tissue-engineered Tumor Matrisome for Prostate Cancer and Melanoma Immunotherapy
- Radiotherapy and Immunotherapy: The Power of the Teamwork for the Treatment of NSCLC
- Three Cases of Nivolumab Plus Ipilimumab Therapy in Haemodialysis Patients With Metastatic Renal Cell Carcinoma
- Possible Pseudo-progression of Non-small Cell Lung Carcinoma in a Patient With Clinical Hyper-progression Associated With Trousseau Syndrome Who Was Treated With Pembrolizumab: A Case Report
- Salvage Chemotherapy After Nivolumab for Recurrent or Metastatic Head and Neck Carcinoma
- Association Between Immune-related Adverse Events and Clinical Outcome Following Nivolumab Treatment in Patients With Metastatic Renal Cell Carcinoma
- Proteasomal and lysosomal degradation for specific and durable suppression of immunotherapeutic targets