Abstract
Introduction:
Immune checkpoint inhibitors have become standard of care for many patients with non-small cell lung cancer (NSCLC). These agents often cause immune-related adverse events (IRAEs), which have been associated with increased overall survival (OS). Intracranial disease control and OS for patients experiencing IRAEs with metastatic NSCLC and brain metastases have not yet been described.
Methods:
We performed a single-institution, retrospective review of patients with NSCLC and existing diagnosis of brain metastasis, who underwent pembrolizumab treatment and developed any grade IRAE. The primary outcome of the study was intracranial time to treatment failure (TTF), defined from time of pembrolizumab initiation to new intracranial disease progression or death. Kaplan-Meier and Cox proportional hazard analyses were performed.
Results:
A total of 63 patients with NSCLC brain metastasis were identified, and 24 developed IRAEs. Patients with any grade IRAEs had longer OS (21 versus 10 months (p=0.004), systemic TTF (15 versus 4 months, p<0.001) and intracranial TTF (14 versus 5 months, p=0.001), relative to patients without IRAEs. Presence of IRAEs and high PD-L1 (≥50%), but not absent/moderate PD-L1 (0–49%), had a positive association for OS, systemic TTF, and intracranial TTF. Following multivariable analysis, IRAE experienced on pembrolizumab was an independent predictor of OS, systemic TTF, and intracranial TTF.
Conclusions:
In our series of patients with NSCLC and brain metastases treated with pembrolizumab, IRAE presence was associated with a significant increase in OS, systemic TTF, and intracranial TTF. Future studies with increased cohorts will clarify how IRAEs should be interpreted among molecular subtypes.
Keywords: brain metastasis, immune-related adverse events, immunotherapy, non-small cell lung cancer, pembrolizumab, progression
Introduction
Pembrolizumab is an immune checkpoint inhibitor (ICI) and the standard of care for advanced non-small cell cancer (NSCLC) as a monotherapy for high PD-L1 (≥50%) expressing EGFR/ALK wild-type tumors as well as in combination with chemotherapy irrespective of PD-L1 expression.[1–3] ICIs are also known to cause immune- related adverse events (IRAEs) due to lymphocyte mediated cross-reactions with normal, host antigens. [4] These IRAEs affect nearly one-third of patients and range in severity, from relatively benign to more life-threatening conditions. [3, 5] Importantly, IRAEs are thought to be potential harbingers of improved treatment response to ICIs across a variety of cancers. [5–10]
Major NSCLC clinical trials investigated patients with treated or stable brain metastasis (9.1% - 17.5% of cohorts), which made it difficult to determine whether there were also IRAE-associated benefits in this important subset.[3, 11–13] Thus, we sought to examine the outcomes associated with pembrolizumab-induced IRAEs in patients with metastatic NSCLC to the brain. Our hypothesis was that IRAEs experienced in this population would not only enhance systemic disease control, but also intracranial disease control to ICIs.
Methods
Patient Selection and Treatment
An institutional review board-approved retrospective chart review was performed to identify patients with advanced NSCLC and a prior diagnosis of brain metastasis who were treated with a pembrolizumab ICI-containing regimen in any line of therapy between January 2014 and February 2019 (Table 1). IRAEs during pembrolizumab treatment were identified and graded as described by the Criteria for Adverse Events version 4.0 (CTCAE v4.0). Low-grade IRAEs included Grade 1–2 events, and high-grade included Grade 3–4 events. A pre-specified landmark- based assessment was performed for patients alive six weeks after pembrolizumab initiation to ensure adequate treatment exposure. Patients were required survival and follow-up for at least six months from the time of brain tumor diagnosis.[5, 14]
Table 1.
Cohort demographics and treatment characteristics of NSCLC brain metastasis patients treated with pembrolizumab
| Variables | Developed Pembrolizumab-related IRAE? | ||
|---|---|---|---|
| Yes (N = 24) (%) | No (N = 39) (%) | p-values | |
| Median time of follow-up (months) | Median: 34.5 (range: 12–106) | Median: 22 (range: 6–99) | 0.005** |
| Age at Presentation (years) | Median: 65.5 (range: 34–82) | Median: 60 (range:19–83) | 0.447 |
| Female | 11 (45.8) | 20 (51.3) | 0.674 |
| Histology | 0.863 | ||
| Adenocarcinoma | 23 (95.8) | 33 (84.6) | |
| Squamous | 1 (4.2) | 4 (10.3) | |
| Adenosquamous | 0 (0.0) | 1 (2.6) | |
| Mixed | 0 (0.0) | 1 (2.6) | |
| Mutations | |||
| EGFR | 10 (41.7) | 13 (33.3) | 0.505 |
| ALK | 0 (0.0) | 2 (5.1) | 0.521 |
| ROS1 | 0 (0.0) | 1 (2.6) | 1.000 |
| PD-L1 Level | 0.412 | ||
| <1% | 2 (8.3) | 6 (15.4) | |
| 1–49% | 4 (16.7) | 12 (30.8) | |
| ≥50% | 14 (58.3) | 15 (38.5) | |
| Unknown | 4 (16.7) | 6 (15.4) | |
| Received Pembrolizumab 1st line? | 0.137 | ||
| No | 18 (75.0) | 22 (56.4) | |
| Yes | 6 (25.0) | 17 (43.6) | |
| Received targeted therapy 1st line? | 0.667 | ||
| No | 16 (66.7) | 28 (71.8) | |
| Yes | 8 (33.3) | 11 (28.2) | |
| Features at time of BM diagnosis | |||
| Karnofsky Performance Status | 0.874 | ||
| 100 | 2 (8.3) | 5 (12.8) | |
| 90 | 3 (12.5) | 3 (7.7) | |
| 80 | 10 (41.7) | 14 (35.9) | |
| 70 | 2 (8.3) | 7 (18.0) | |
| 50 | 1 (4.2) | 1 (2.6) | |
| Unknown | 6 (25.0) | 9 (23.1) | |
| Neurological deficit | 9 (37.5) | 8 (20.5) | 0.140 |
| Cerebellar involvement | 10 (41.7) | 14 (35.9) | 0.647 |
| Extracranial disease control | 15 (62.5) | 19 (48.7) | 0.287 |
| Procedural history | |||
| Any surgery for BM | 7 (29.2) | 4 (10.3) | 0.086 |
| Radiation at diagnosis | 20 (83.3) | 34 (87.2) | 0.721 |
| Radiation boost after surgery | 2 (8.3) | 2 (5.13) | 0.632 |
| Radiation necrosis | 5 (20.8) | 2 (5.1) | 0.095 |
| Whole brain radiation | 3 (12.5) | 6 (15.4) | 1.000 |
| First SRS session | |||
| Received SRS, n (%) | 19 (79.2) | 33 (84.6) | 0.735 |
| Number of lesions, mean (SD) | 4.4 (4.0) | 4.2 (4.5) | 0.991 |
| Dose (Gy), mean (SD) | 23.3 (2.3) | 23.7 (2.3) | 0.597 |
| Fractions, mean (SD) | 1.3 (0.8) | 1.5 (1.0) | 0.445 |
| Total volume (cm3), mean (SD) | 4.3 (4.3) | 4.9 (11.4) | 0.224 |
| Second SRS session | |||
| Received 2 SRS treatments, N (%) | 11 (45.8) | 20 (51.3) | 0.674 |
| Number of lesions, mean (SD) | 3.2 (4.9) | 3.3 (3.9) | 0.960 |
| Dose (Gy), mean (SD) | 23.5 (1.8) | 23.3 (2.5) | 0.789 |
| Fractions, mean (SD) | 1.3 (0.6) | 1.2 (0.6) | 0.798 |
| Total volume (cm3), mean (SD) | 1.2 (3.1) | 2.4 (5.1) | 0.483 |
BM, brain metastasis; Gy, gray; IRAE, immune related adverse events; SD, standard deviation; SRS, stereotactic radiosurgery
p < 0.05
p < 0.01
p < 0.001
Systemic treatment regimens, both prior to and after the detection of brain metastasis, including targeted therapies (e.g. tyrosine kinase inhibitors, TKIs), were recorded for each patient. The line of each of these treatments was also noted. Surgery and radiation when used for the management of the intracranial disease were also recorded, including the total number of interventions. For patients who underwent surgery, tumor location, pre-operative neurological deficits, and extracranial control status were recorded. For patients who underwent radiation, the radiation technique (post-operative radiosurgery to resection cavity, whole brain radiotherapy, radiosurgery alone), prescription dose, number of fractions, total volume, and number of lesions were recorded. [15] Patients with radiation treatment within 14 days prior to pembrolizumab therapy were assessed for possible synergistic effects.[16]
Molecular characteristics (i.e., EGFR, ALK, ROS1 alterations) as well as PD-L1 protein expression as measured by immunohistochemistry were recorded. Tumor PD-L1 expression levels were categorized as absent (< 1%), intermediate (1–49%), and high (≥ 50%). Given the low proportion of tumors with absent expression, this subgroup was analyzed in combination with intermediate-expressing tumors (0–49%).
The primary outcome analyzed was intracranial time to treatment failure (TTF), which was defined as time from pembrolizumab initiation to new intracranial progression or death. Local failure after radiation was defined as radiographic evidence of tumor progression within the isodose line representing 80% of the prescription dose. Pseudoprogression was distinguished from local failure by stabilization or shrinkage of the lesion on follow-up MRIs over time or by pathologic confirmation of necrosis in the absence of residual tumor in resected lesions. Secondary outcomes included OS and systemic (extracranial) TTF, similarly defined from the date of pembrolizumab initiation. Systemic and intracranial TTF were measured until the date of systemic imaging or brain MRI, respectively, showing progression warranting change in regimen, or measured until date of death. Patients lost to follow-up were censored.
Statistical Analysis
Kaplan-Meier analyses assessed OS and TTF. Multivariable Cox proportional hazard models determined variables associated with OS and TTF. A univariate analysis was first conducted and variables with p < 0.15 were included in the multivariable analysis, and results are presented as hazard ratios (HR) and 95% confidence intervals. Categorical variables were analyzed by the χ2 test or Fisher’s exact test as appropriate. Normally distributed continuous variables were analyzed by Student’s t-test, while non-parametric continuous variables were analyzed by the Wilcoxon rank-sum test. All data were analyzed using STATA 14.2 (StataCorp, USA) and GraphPad Prism 8 (GraphPad Software, Inc.). The significance level was set at a two-sided alpha of 0.05.
Results
Patient Population and Treatment
A total of 63 patients with NSCLC and previously diagnosed brain metastasis were identified, and 24 patients experienced IRAEs. Median time of follow-up for patients with and without IRAEs were 34.5 (range: 12–106) and 22 months (range: 6–99), respectively (p = 0.005). There were no statistically significant differences in baseline demographics, disease presentation, or treatment history between cohorts (Table 1). Median ages were 65.5 and 60 years-old for patients with and without IRAEs, respectively (p = 0.447). Gender was female for 46% and 51%, respectively (p = 0.674). Adenocarcinoma (96% and 85%, respectively), EGFR-positive mutation status (42% and 33%, respectively), and PD-L1 levels ≥ 50% (58% and 39%, respectively) were the most commonly identified histologic and molecular variants.
With regard to treatment among those with and without IRAEs, 33% and 28% of patients, respectively, were administered a targeted agent prior to initiation of pembrolizumab. Pembrolizumab was part of a first-line systemic treatment among 25% and 44% of patients, respectively. Radiation was administered for local control of initial brain metastasis in 83% and 87%, respectively, with similar proportions of ECOG performance status, presence of neurological deficits, and extracranial disease control status between cohorts. Whole brain radiotherapy was administered to 13% and 15%, respectively. Thirteen patients underwent radiation within 14 days before initiating pembrolizumab, two of whom experienced IRAEs. Surgery was performed in 29% and 10%, respectively, for the treatment of brain metastasis (p = 0.086). Overall, twenty-four patients (38%) experienced objectively documented IRAEs. Of these, fifteen were with low-grade IRAEs, and nine were with high-grade IRAEs (Supplemental Table 1).
Survival and Progression Analyses
Kaplan-Meier survival curves were constructed to analyze our primary and secondary outcomes for those with and without IRAEs (Fig. 1, Table 2). Among the entire study population (N = 63), for those with and without IRAEs, the median OS was 21 months versus 10 months (p = 0.004), respectively. The median systemic TTF as measured from time of pembrolizumab initiation for those with and without pembrolizumab-induced IRAEs was 15 months versus 4 months (p < 0.001), respectively. Statistically significant benefits were also observed for our primary outcome variable: duration of intracranial TTF was 14 months versus 5 months (p = 0.001), respectively.
Fig. 1 :
Survival curves for (A) overall survival, (B) systemic time to treatment failure, and (C) intracranial time to treatment failure, stratified by history of immune-related adverse events
Table 2.
Median survival for Kaplan-Meier analyses of NSCLC brain metastasis patients treated with pembrolizumab with and without IRAEs
| Kaplan-Meier Curve | IRAE (months, 95% CI) | No IRAE (months, 95% CI) | p-value |
|---|---|---|---|
| Overall survival | 21 (18-undef.) | 10 (7–14) | 0.004** |
| Systemic TTF | 15 (2–19) | 4 (2–5) | <0.001*** |
| Brain TTF | 14 (5–29) | 5 (4–8) | 0.001** |
| Overall survival: PD-L1 level:<50% | 20 (2-undef.) | 10 (6–25) | 0.744 |
| Overall survival: PD-L1 level:>50% | Undef. (15-undef.) | 13 (5–35) | 0.011* |
| Systemic TTF: PD-L1 level:<50% | 2 (1-undef.) | 4.5 (3–9) | 0.918 |
| Systemic TTF: PD-L1 level:>50% | 16 (5-undef.) | 3 (1–5) | <0.001*** |
| Brain TTF: PD-L1 level:<50% | 6 (1-undef.) | 6 (4–9) | 0.907 |
| Brain TTF: PD-L1 level:>50% | 17 (5-undef.) | 3 (2–12) | 0.005** |
| Overall survival: No IRAE vs. Low-grade IRAE | 21 (15-undef.) | 10 (7–14) | 0.026* |
| Overall survival: No IRAE vs. High-grade IRAE | Undef. (2-undef.) | 10 (7–14) | 0.040* |
| Systemic TTF: No IRAE vs. Low-grade IRAE | 10 (2–16) | 4 (2–5) | 0.015* |
| Systemic TTF: No IRAE vs. High-grade IRAE | 19 (1-undef.) | 4 (2–5) | 0.001** |
| Brain TTF : No IRAE vs. Low-grade IRAE | 15 (5–29) | 5 (4–8) | 0.002** |
| Brain TFF : No IRAE vs. High-grade IRAE | 11 (1-undef.) | 5 (4–8) | 0.069 |
CI, confidence interval; IRAE, immune-related adverse events; TTF, time to treatment failure; Undef., undefined
p < 0.05
p < 0.01
p < 0.001
The effect of IRAEs was also evaluated by PD-L1 expression levels (Fig. 2). Among patients with NSCLC brain metastases and high PD-L1 levels (≥50%), the median OS was significantly greater for those with than without IRAEs (not reached versus 13 months, p = 0.011). Patients with IRAEs were also with longer median systemic TTF (16 months versus 3 months, p < 0.001) and longer intracranial TTF (17 months versus 3 months, p = 0.005), relative to those without IRAEs.
Fig. 2:
Survival curves for (A, D) overall survival, (B, E) systemic time to treatment failure, and (C, F) intracranial time to treatment failure, stratified by history of immune-related adverse events for low (A-C) and high (D-F) PD- L1 expression levels.
We additionally examined if the severity of IRAE impacted outcomes (Fig. 3). Patients with either low- (21 versus 10 months, p = 0.026) or high-grade IRAEs (not reached versus 10 months, p = 0.040) were noted to have significantly increased OS. Patients with either low- (10 months versus 4 months, p = 0.015) or high-grade IRAEs (19 months versus 4 months, p= 0.001) demonstrated longer systemic TTF, relative to patients without IRAEs. Meanwhile, low-grade IRAEs (15 months versus 5 months, p = 0.002) were associated with longer intracranial TTF, relative to those without IRAEs. High-grade IRAEs (11 months versus 5 months, p = 0.069) only trended towards longer intracranial TTF, relative to those without IRAEs.
Fig. 3:
Survival curves for (A, D) overall survival, (B, E) systemic time to treatment failure, and (C, F) intracranial time to treatment failure, stratified by history of immune-related adverse events for low-grade (A-C) and high-grade (D-F) IRAE severity.
Finally, given that some patients received EGFR-targeted therapies as first-line agents and were subsequently given pembrolizumab as salvage therapy, we examined outcomes for patients based on EGFR mutation status. For patients with EGFR wildtype tumors, there was a statistically significant increase in systemic TTF on pembrolizumab (6 months vs 3 months, p = 0.021; Supplementary Fig. 1, Supplementary Table 2), relative to patients with EGFR mutations receiving pembrolizumab. Additionally, there were trends for increased OS (18 months vs 9 months, p = 0.061) and intracranial TTF (9 months and 5 months, p = 0.068) for patients receiving pembrolizumab with tumors that were EGFR wildtype versus EGFR mutant.
Predictors of Outcomes
In a multivariable analysis, the presence of IRAE was an independent predictor for prolonged OS, systemic TTF, and intracranial TTF (HR = 0.08, 95% CI = 0.02–0.27, p < 0.001; HR = 0.33, 95% CI = 0.16–0.69, p = 0.003; HR = 0.15, 95% CI = 0.05–0.39, p < 0.001; Supplementary Table 3–5).
Additionally, OS was negatively predicted by use of targeted therapy for first-line treatment (HR = 10.01, 95% CI = 2.53–39.7, p = 0.001), uncontrolled systemic disease at the time of radiation treatment (HR = 4.65, 95% CI = 1.63–13.23, p = 0.004), and presence of ≥10 lesions at the time of first radiosurgery treatment (HR = 12.70, 95% CI = 1.94–83.17, p = 0.008; Supplementary Table 3). Both systemic TTF (HR = 2.47, 95% CI = 1.05–5.85, p = 0.039; Supplementary Table 4) and intracranial TTF (HR = 3.59, 95% CI = 1.04–12.39, p = 0.043; Supplementary Table 5) were also predicted by use of a targeted agent as a first-line therapeutic.
To confirm that the hypothesized IRAE-associations were not confounded by patients with tumors with targeted mutations receiving first-line targeted treatments, a secondary multivariable analysis was performed examining patients with tumors without targetable mutations. Again, history of IRAE was predicted of improved OS, systemic TTF, and intracranial TTF (HR = 0.19, 95% CI = 0.04–0.87, p = 0.032; HR = 0.31, 95% CI = 0.110.90, p = 0.032; HR = 0.20, 95% CI = 0.07–0.56, p = 0.002; Supplementary Table 6).
Discussion
In this study we show that patients with metastatic NSCLC to the brain can experience extended OS, systemic TTF, and intracranial TTF following pembrolizumab therapy, particularly if they develop pembrolizumab- related IRAEs. These improvements measured to median increases in 11 months, 11 months, and 9 months, respectively. Prolonged outcomes following IRAE appear to be specifically experienced by patients with PD-L1 levels ≥50%. Further stratification based on severity yielded similar benefits to OS, systemic TTF, and intracranial TTF for patients with Grade 1 and 2 IRAEs and supported statistically significant, longer OS and systemic TTF for patients with Grade 3 and 4 IRAEs. Multivariable regression analyses further identified pembrolizumab-associated IRAEs to be an independent predictor of prolonged survival and disease control.
Pembrolizumab-associated IRAEs with improved outcomes for NSCLC with brain metastasis
Given the wide prevalence of NSCLC brain metastasis and its negative prognostication in oncologic natural history, identifying the incremental impact of IRAEs on outcomes can further guide patient counseling. For a historical comparison, only a few prior studies have specifically investigated ICIs for NSCLC with brain metastasis. Among landmark trials, KEYNOTE-189 enrolled the largest number of brain metastasis patients (17.5%) for first-line pembrolizumab-chemotherapy combination therapy and confirmed OS and PFS benefits, lasting 22.0 and 9.0 months, respectively, after 23.1 months median follow-up. [17] Meanwhile in a trial for second-line pembrolizumab, specifically investigating NSCLC with brain metastasis, Goldberg et al. reported OS, systemic PFS, and CNS PFS of 9.9 months, 1.9 months, and 2.3 months, respectively, following salvage pembrolizumab initiation for PD-L1 ≥1%. [2] Our study describes comparatively longer outcomes for patient with brain metastasis and may suggest a means of optimizing outcomes with proper patient selection.
Further work is needed to clarify if higher IRAE severity can signify added intracranial control. Although our own analysis for intracranial TTF among with high grade IRAE did not demonstrate statistical significance, it is possible that in a larger cohort, the existing trend would have been strengthened. Likewise, our sample size precluded a comparison of outcomes between those with high- and low-severity IRAEs. In their NSCLC experience, Baldini et al. described better OS (19.2 versus 11.2 months) and TTF (7.0 months versus 3.2 months) for low- than high-grade IRAEs after nivolumab. [6] Other authors have consistently supported a positive association between low-grade toxicities and treatment outcomes. Skin, for example, exhibits the most common low-grade IRAEs and is antigenically the most similar organ to NSCLC, while brain is among the least similar. [4] When Hosoya et al. focused on NSCLC patients who experienced early rashes (95% low-grade) after second-line nivolumab use, they also exhibited better TTF, relative to those who did not experience them. [8]
Alternatively, a clinically smaller, positive IRAE association may occur with intracranial metastasis following a week immunologic response. Since IRAE severity is not necessarily synonymous with IRAE antigenicity, an extracranial IRAEs may be debilitating without generating a sufficiently strong, primed T-cell response.[18] This could limit the clinical benefit for distant metastases, particularly when across the blood brain barrier.
IRAE-associated outcomes based on molecular profile
Patients with tumors with driver mutations, particularly EGFR and ALK, are known to have poorer clinical responses to ICIs and tend to only receive these treatments as later line, salvage therapy with a lower chance of response.[19–22] However, the presence of a driver mutation has not been shown to correlate with a lower incidence of IRAEs. If patients had primarily been administered pembrolizumab as a salvage therapy, then the corresponding cohort will be confounded with poorer outcomes. In this study, the cohort with IRAEs was actually with greater EGFR representation than that without IRAEs, suggesting a lesser likelihood for our outcomes to be upwardly biased.
Our multivariable regression analyses also supported prolonged IRAE-associated survival and disease control for patients treated with pembrolizumab as initial therapy for metastatic disease. After excluding patients with targetable mutations, the IRAE remains an even more significant predictor. However, additional studies will be necessary to confirm if patients with tumors with targetable mutations also see outcome benefits after IRAEs. This becomes particularly important as ICIs are considered for those with targeted mutations. IRAEs are potentially even more prominent when ICIs are administered in combination therapies.[23, 24] A secondary, multivariable analysis for patients with targetable mutations was not possible in our study due to cohort size. It is important to realize that PD-L1 levels may be high in patients with tumors with actionable mutations, but the actionable mutation may be a stronger predictor of benefit, or lack thereof, than the PD–L1 level or even development of IRAE.
Still, our evidence does strongly support survival and disease control benefits when PD-L1 expression is ≥50%. Previously, Cortellini et al. reported on a large cohort of over 800 NSCLC patients with ≥50% PD-L1 expression and described IRAEs as being associated with improved OS and systemics PFS. [5] Our findings add credence to the existing recommendation for using pembrolizumab as a first-line agent among patients with high PD-L1 levels.[1–3] Specifically, we demonstrate a clinically relevant response to pembrolizumab-related IRAEs by both the extracranial and intracranial diseases, perhaps consistent with prior descriptions of their concordant PD-L1 expression levels. [25, 26]
Limitation
As a retrospective study uniquely characterizing the NSCLC brain metastasis experience, the availability of covariates and documentation precludes comparisons across other studies. Additionally, outcomes measurement may differ due to the lead time bias associated with times of NSCLC diagnosis and initiation of pembrolizumab therapy. Particularly, our methods were restricted to monitoring time to treatment failure rather than progressionfree survival, which is based on RECIST or RANO criteria and used in clinical trials. While systematic application of iRANO criteria is preferred, the unavailability of some baseline imaging precluded such analysis.
There are also intrinsic challenges to studying the brain metastasis population including assurance of adequate follow-up and of adequate cohort size for study. Restricting our study to patients with at least 6 months survival may have presented an upward bias; however, this is offset by a comparable group of patients undergoing active pembrolizumab treatment given its current first-line indications. Thus, more patients over the past few years have been treated accordingly, with the most recent patients being censored for this study.
Since our inclusion criteria permitted pembrolizumab to be used at any stage during treatment, the contributions of chemotherapies before and after cannot be excluded. Nevertheless, by indexing TTF to the start of pembrolizumab initiation, our study captures the immediate, immunologically response. Furthermore, our secondary analyses provided reassurance that targeted therapy did not bias our assessment of a positive association between IRAEs and outcomes. Larger studies will be needed to confirm if those receiving first-line targeted therapies experience augmented outcomes after IRAEs, as our sample size was limited for this subgroup.
Further studies will be important for confirming what impacts exist from interactions with local control. Our cohort presented with a statistically insignificant, but higher presence of surgical resection among patients with IRAEs. Meanwhile, timing of SRS and RT relative to the start of pembrolizumab is of active consideration given possible abscopal effects. We identified patients with radiation treatment within 14 days prior to pembrolizumab initiation; however, further analysis was limited by the small subgroup and its imbalance towards patients without IRAEs (Supplemental Figure 2). Additional permutations in how radiation and surgery are inserted across a treatment course are also not addressed in this study. [27]
Finally, a histological correlate would have helped corroborate our hypothesis that IRAE-associated outcomes apply to intracranial NSCLC metastasis. PD-L1 status for intracranial disease can only be determined with biopsy or resection, which was performed for a minority of our patients. Only two PD-L1 levels from intracranial samples were available. Both were concordant with primary pathology and support our findings that IRAE- associated outcomes generalize between extracranial and intracranial disease.
Conclusions
Patients with NSCLC brain metastasis can benefit from IRAE-mediated effects by prolongation of survival, systemic disease control, and intracranial disease control. In particular, we show these outcomes are all sustained for high PD-L1 expressing NSCLC brain metastases. Likewise, there is a consistent positive association with pembrolizumab-related, low-grade IRAEs. As immune checkpoint inhibitors are increasingly utilized in NSCLC treatment, future work can expect larger cohorts to personalize NSCLC management around histologic criteria.
Supplementary Material
Supplementary Fig. 1A-C: Survival curves for (A) overall survival, (B) systemic time to treatment failure, and (C) intracranial time to treatment failure, stratified by EGFR mutation status.
Supplementary Fig. 2A-C: Survival curves for (A) overall survival, (B) systemic time to treatment failure, and (C) intracranial time to treatment failure, stratified by history of radiation treatment within 14 days before initiating pembrolizumab.
Funding:
No funding was received for the production of this study. MZ receives research funding from the National Institutes of Health (5T32CA009695-27, MPI). SGS receives research funding from Novocure. JWN receives research funding from Adaptimmune Therapeutics plc, Boehringer Ingelheim, Exelixis, Genentech/Roche, GlaxoSmithKline, Merck & Co., Nektar Therapeutics, Novartis, and Takeda Pharmaceuticals. SKP receives research funding from Bayer, Boehringer Ingelheim, and EpicentRx. HW receives research funding from Arrys Therapeutics, Bristol Myers Squibb, Celgene, Clovis Oncology, Eli Lilly and Company, Exelixis, Genentech/Roche, Merck and Co., Novartis, and Pfizer. GL receives research support from Bristol Myers Squibb and Novocure.
Footnotes
Conflicts of interest/Competing interests:
SGS is a consultant for Inovio Pharmaceuticals, Inc. and speaker for Zap Surgical Systems, Inc. SKP is an advisor for AstraZeneca, Blueprint Medicines, and G1 Therapeutics, Inc. KR is an advisor for Dhristi, Inc. and Varian Medical Systems. HW is an advisor for AstraZeneca, Blueprint Medicines, Cellworks Group, Inc., Genentech/Roche, Helsinn Healthcare SA, Mirati Therapeutics, Inc., and the International Thymic Malignancy Interest Group. JWN receives royalties from UpToDate. GL is an advisor for Medtronic, a speaker for DePuy Synthes/Johnson&Johnson, and own stock in Abbott, Biogen, BioMarin Pharmaceutical, Inc., Bristol Myers Squibb, Exact Sciences, GlaxoSmithKline, Gilead Sciences, Inc., Merck & Co., Pfizer, and Stryker.
Ethics approval: Not applicable
Consent to participate: Not applicable
Consent for publication: All authors
Availability of data and material: Not applicable
Code availability: Not applicable
Publisher's Disclaimer: This Author Accepted Manuscript is a PDF file of an unedited peer-reviewed manuscript that has been accepted for publication but has not been copyedited or corrected. The official version of record that is published in the journal is kept up to date and so may therefore differ from this version.
References:
- 1.Gandhi L, Rodríguez-Abreu D, Gadgeel S, Esteban E, Felip E, De Angelis F, Domine M, Clingan P, Hochmair MJ, Powell SF, Cheng SY, Bischoff HG, Peled N, Grossi F, Jennens RR, Reck M, Hui R, Garon EB, Boyer M, Rubio-Viqueira B, Novello S, Kurata T, Gray JE, Vida J, Wei Z, Yang J, Raftopoulos H, Pietanza MC, Garassino MC (2018) Pembrolizumab plus Chemotherapy in Metastatic Non-Small-Cell Lung Cancer. The New England journal of medicine 378: 2078–2092 doi: 10.1056/NEJMoa1801005 [DOI] [PubMed] [Google Scholar]
- 2.Goldberg SB, Schalper KA, Gettinger SN, Mahajan A, Herbst RS, Chiang AC, Lilenbaum R, Wilson FH, Omay SB, Yu JB, Jilaveanu L, Tran T, Pavlik K, Rowen E, Gerrish H, Komlo A, Gupta R, Wyatt H, Ribeiro M, Kluger Y, Zhou G, Wei W, Chiang VL, Kluger HM (2020) Pembrolizumab for management of patients with NSCLC and brain metastases: long-term results and biomarker analysis from a non-randomised, open-label, phase 2 trial. The Lancet Oncology 21: 655–663 doi: 10.1016/s1470-2045(20)30111-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Reck M, Rodríguez-Abreu D, Robinson AG, Hui R, Csőszi T, Fülöp A, Gottfried M, Peled N, Tafreshi A, Cuffe S, O’Brien M, Rao S, Hotta K, Leiby MA, Lubiniecki GM, Shentu Y, Rangwala R, Brahmer JR (2016) Pembrolizumab versus Chemotherapy for PD-L1-Positive Non-Small-Cell Lung Cancer. The New England journal of medicine 375: 1823–1833 doi: 10.1056/NEJMoa1606774 [DOI] [PubMed] [Google Scholar]
- 4.Berner F, Bomze D, Diem S, Ali OH, Fässler M, Ring S, Niederer R, Ackermann CJ, Baumgaertner P, Pikor N, Cruz CG, van de Veen W, Akdis M, Nikolaev S, Läubli H, Zippelius A, Hartmann F, Cheng HW, Hönger G, Recher M, Goldman J, Cozzio A, Früh M, Neefjes J, Driessen C, Ludewig B, Hegazy AN, Jochum W, Speiser DE, Flatz L (2019) Association of Checkpoint Inhibitor-Induced Toxic Effects With Shared Cancer and Tissue Antigens in Non-Small Cell Lung Cancer. JAMA Oncol 5: 1043–1047 doi: 10.1001/jamaoncol.2019.0402 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Cortellini A, Friedlaender A, Banna GL, Porzio G, Bersanelli M, Cappuzzo F, Aerts J, Giusti R, Bria E, Cortinovis D, Grossi F, Migliorino MR, Galetta D, Passiglia F, Berardi R, Mazzoni F, Di Noia V, Signorelli D, Tuzi A, Gelibter A, Marchetti P, Macerelli M, Rastelli F, Chiari R, Rocco D, Inno A, Di Marino P, Mansueto G, Zoratto F, Santoni M, Tudini M, Ghidini M, Filetti M, Catino A, Pizzutilo P, Sala L, Occhipinti MA, Citarella F, Marco R, Torniai M, Cantini L, Follador A, Sforza V, Nigro O, Ferrara MG, D’Argento E, Leonetti A, Pettoruti L, Antonuzzo L, Scodes S, Landi L, Guaitoli G, Baldessari C, Bertolini F, Della Gravara L, Dal Bello MG, Belderbos RA, De Filippis M, Cecchi C, Ricciardi S, Donisi C, De Toma A, Proto C, Addeo A, Cantale O, Ricciuti B, Genova C, Morabito A, Santini D, Ficorella C, Cannita K (2020) Immune-related Adverse Events of Pembrolizumab in a Large Real-world Cohort of Patients With NSCLC With a PD-L1 Expression ≥50% and Their Relationship With Clinical Outcomes. Clinical lung cancer doi: 10.1016/j.cllc.2020.06.010 [DOI] [PubMed] [Google Scholar]
- 6.Baldini E, Lunghi A, Cortesi E, Turci D, Signorelli D, Stati V, Melotti B, Ricciuti B, Frassoldati A, Romano G, Ceresoli GL, Illiano A, Verderame F, Fasola G, Ricevuto E, Marchetti P, Pinto C, Cartenì G, Scotti V, Tibaldi C, Fioretto L, Giannarelli D (2020) Immune-related adverse events correlate with clinical outcomes in NSCLC patients treated with nivolumab: The Italian NSCLC expanded access program. Lung cancer (Amsterdam, Netherlands) 140: 59–64 doi: 10.1016/j.lungcan.2019.12.014 [DOI] [PubMed] [Google Scholar]
- 7.Freeman-Keller M, Kim Y, Cronin H, Richards A, Gibney G, Weber JS (2016) Nivolumab in Resected and Unresectable Metastatic Melanoma: Characteristics of Immune-Related Adverse Events and Association with Outcomes. Clinical cancer research : an official journal of the American Association for Cancer Research 22: 886–894 doi: 10.1158/1078-0432.Ccr-15-1136 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Hosoya K, Fujimoto D, Morimoto T, Kumagai T, Tamiya A, Taniguchi Y, Yokoyama T, Ishida T, Hirano K, Matsumoto H, Kominami R, Tomii K, Suzuki H, Hirashima T, Uchida J, Morita M, Kanazu M, Sawa N, Makio T, Hara S, Tamiya M (2020) Association Between Early Immune- related Adverse Events and Clinical Outcomes in Patients With Non-Small Cell Lung Cancer Treated With Immune Checkpoint Inhibitors. Clinical lung cancer 21: e315–e328 doi: 10.1016/j.cllc.2020.01.003 [DOI] [PubMed] [Google Scholar]
- 9.Hua C, Boussemart L, Mateus C, Routier E, Boutros C, Cazenave H, Viollet R, Thomas M, Roy S, Benannoune N, Tomasic G, Soria JC, Champiat S, Texier M, Lanoy E, Robert C (2016) Association of Vitiligo With Tumor Response in Patients With Metastatic Melanoma Treated With Pembrolizumab. JAMA dermatology 152: 45–51 doi: 10.1001/jamadermatol.2015.2707 [DOI] [PubMed] [Google Scholar]
- 10.Toi Y, Sugawara S, Kawashima Y, Aiba T, Kawana S, Saito R, Tsurumi K, Suzuki K, Shimizu H, Sugisaka J, Ono H, Domeki Y, Terayama K, Nakamura A, Yamanda S, Kimura Y, Honda Y (2018) Association of Immune-Related Adverse Events with Clinical Benefit in Patients with Advanced Non-Small-Cell Lung Cancer Treated with Nivolumab. The oncologist 23: 1358–1365 doi: 10.1634/theoncologist.2017-0384 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Carbone DP, Reck M, Paz-Ares L, Creelan B, Horn L, Steins M, Felip E, van den Heuvel MM, Ciuleanu TE, Badin F, Ready N, Hiltermann TJN, Nair S, Juergens R, Peters S, Minenza E, Wrangle JM, Rodriguez-Abreu D, Borghaei H, Blumenschein GR Jr., Villaruz LC, Havel L, Krejci J, Corral Jaime J, Chang H, Geese WJ, Bhagavatheeswaran P, Chen AC, Socinski MA (2017) First- Line Nivolumab in Stage IV or Recurrent Non-Small-Cell Lung Cancer. The New England journal of medicine 376: 2415–2426 doi: 10.1056/NEJMoa1613493 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Langer CJ, Gadgeel SM, Borghaei H, Papadimitrakopoulou VA, Patnaik A, Powell SF, Gentzler RD, Martins RG, Stevenson JP, Jalal SI, Panwalkar A, Yang JC, Gubens M, Sequist LV, Awad MM, Fiore J, Ge Y, Raftopoulos H, Gandhi L (2016) Carboplatin and pemetrexed with or without pembrolizumab for advanced, non-squamous non-small-cell lung cancer: a randomised, phase 2 cohort of the open-label KEYNOTE-021 study. The Lancet Oncology 17: 1497–1508 doi: 10.1016/s1470-2045(16)30498-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Hellmann MD, Ciuleanu T-E, Pluzanski A, Lee JS, Otterson GA, Audigier-Valette C, Minenza E, Linardou H, Burgers S, Salman P, Borghaei H, Ramalingam SS, Brahmer J, Reck M, O’Byrne KJ, Geese WJ, Green G, Chang H, Szustakowski J, Bhagavatheeswaran P, Healey D, Fu Y, Nathan F, Paz-Ares L (2018) Nivolumab plus Ipilimumab in Lung Cancer with a High Tumor Mutational Burden. New England Journal of Medicine 378: 2093–2104 doi: 10.1056/NEJMoa1801946 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Haratani K, Hayashi H, Chiba Y, Kudo K, Yonesaka K, Kato R, Kaneda H, Hasegawa Y, Tanaka K, Takeda M, Nakagawa K (2018) Association of Immune-Related Adverse Events With Nivolumab Efficacy in Non-Small-Cell Lung Cancer. JAMA Oncol 4: 374–378 doi: 10.1001/jamaoncol.2017.2925 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Verger E, Salamero M, Conill C (1992) Can Karnofsky performance status be transformed to the Eastern Cooperative Oncology Group scoring scale and vice versa? European journal of cancer (Oxford, England : 1990) 28a: 1328–1330 doi: 10.1016/0959-8049(92)90510-9 [DOI] [PubMed] [Google Scholar]
- 16.Antonia SJ, Villegas A, Daniel D, Vicente D, Murakami S, Hui R, Yokoi T, Chiappori A, Lee KH, de Wit M, Cho BC, Bourhaba M, Quantin X, Tokito T, Mekhail T, Planchard D, Kim YC, Karapetis CS, Hiret S, Ostoros G, Kubota K, Gray JE, Paz-Ares L, de Castro Carpeño J, Wadsworth C, Melillo G, Jiang H, Huang Y, Dennis PA, Özgüroğlu M (2017) Durvalumab after Chemoradiotherapy in Stage III Non-Small-Cell Lung Cancer. The New England journal of medicine 377: 1919–1929 doi: 10.1056/NEJMoa1709937 [DOI] [PubMed] [Google Scholar]
- 17.Gadgeel S, Rodríguez-Abreu D, Speranza G, Esteban E, Felip E, Dómine M, Hui R, Hochmair MJ, Clingan P, Powell SF, Cheng SY, Bischoff HG, Peled N, Grossi F, Jennens RR, Reck M, Garon EB, Novello S, Rubio-Viqueira B, Boyer M, Kurata T, Gray JE, Yang J, Bas T, Pietanza MC, Garassino MC (2020) Updated Analysis From KEYNOTE-189: Pembrolizumab or Placebo Plus Pemetrexed and Platinum for Previously Untreated Metastatic Nonsquamous Non-SmallCell Lung Cancer. Journal of clinical oncology : official journal of the American Society of Clinical Oncology 38: 1505–1517 doi: 10.1200/jco.19.03136 [DOI] [PubMed] [Google Scholar]
- 18.Pardoll DM (2012) The blockade of immune checkpoints in cancer immunotherapy. Nature reviews Cancer 12: 252–264 doi: 10.1038/nrc3239 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Lee CK, Man J, Lord S, Cooper W, Links M, Gebski V, Herbst RS, Gralla RJ, Mok T, Yang JC (2018) Clinical and Molecular Characteristics Associated With Survival Among Patients Treated With Checkpoint Inhibitors for Advanced Non-Small Cell Lung Carcinoma: A Systematic Review and Meta-analysis. JAMA Oncol 4: 210–216 doi: 10.1001/jamaoncol.2017.4427 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Oya Y, Kuroda H, Nakada T, Takahashi Y, Sakakura N, Hida T (2020) Efficacy of Immune Checkpoint Inhibitor Monotherapy for Advanced Non-Small-Cell Lung Cancer with ALK Rearrangement. Int J Mol Sci 21 doi: 10.3390/ijms21072623 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Zhao D, Mambetsariev I, Li H, Chen C, Fricke J, Fann P, Kulkarni P, Xing Y, Lee PP, Bild A, Massarelli E, Koczywas M, Reckamp K, Salgia R (2020) Association of molecular characteristics with survival in advanced non-small cell lung cancer patients treated with checkpoint inhibitors. Lung cancer (Amsterdam, Netherlands) 146: 174–181 doi: 10.1016/j.lungcan.2020.05.025 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Mhanna L, Guibert N, Milia J, Mazieres J (2019) When to Consider Immune Checkpoint Inhibitors in Oncogene-Driven Non-Small Cell Lung Cancer? Curr Treat Options Oncol 20: 60 doi: 10.1007/s11864-019-0652-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Oshima Y, Tanimoto T, Yuji K, Tojo A (2018) EGFR-TKI-Associated Interstitial Pneumonitis in Nivolumab-Treated Patients With Non-Small Cell Lung Cancer. JAMA Oncol 4: 1112–1115 doi: 10.1001/jamaoncol.2017.4526 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Oxnard GR, Yang JC, Yu H, Kim SW, Saka H, Horn L, Goto K, Ohe Y, Mann H, Thress KS, Frigault MM, Vishwanathan K, Ghiorghiu D, Ramalingam SS, Ahn MJ (2020) TATTON: a multiarm, phase Ib trial of osimertinib combined with selumetinib, savolitinib, or durvalumab in EGFR-mutant lung cancer. Annals of oncology : official journal of the European Society for Medical Oncology 31: 507–516 doi: 10.1016/j.annonc.2020.01.013 [DOI] [PubMed] [Google Scholar]
- 25.Horn L, Spigel DR, Vokes EE, Holgado E, Ready N, Steins M, Poddubskaya E, Borghaei H, Felip E, Paz-Ares L, Pluzanski A, Reckamp KL, Burgio MA, Kohlhäeufl M, Waterhouse D, Barlesi F, Antonia S, Arrieta O, Fayette J, Crinò L, Rizvi N, Reck M, Hellmann MD, Geese WJ, Li A, Blackwood-Chirchir A, Healey D, Brahmer J, Eberhardt WEE (2017) Nivolumab Versus Docetaxel in Previously Treated Patients With Advanced Non-Small-Cell Lung Cancer: Two-Year Outcomes From Two Randomized, Open-Label, Phase III Trials (CheckMate 017 and CheckMate 057). Journal of clinical oncology : official journal of the American Society of Clinical Oncology 35: 3924–3933 doi: 10.1200/jco.2017.74.3062 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Takamori S, Toyokawa G, Okamoto I, Takada K, Kinoshita F, Kozuma Y, Matsubara T, Haratake N, Akamine T, Mukae N, Hirai F, Tagawa T, Oda Y, Iwaki T, lihara K, Nakanishi Y, Maehara Y (2018) Clinical Significance of PD-L1 Expression in Brain Metastases from Non-small Cell Lung Cancer. Anticancer research 38: 553–557 doi: 10.21873/anticanres.12258 [DOI] [PubMed] [Google Scholar]
- 27.Shaverdian N, Lisberg AE, Bornazyan K, Veruttipong D, Goldman JW, Formenti SC, Garon EB, Lee P (2017) Previous radiotherapy and the clinical activity and toxicity of pembrolizumab in the treatment of non-small-cell lung cancer: a secondary analysis of the KEYNOTE-001 phase 1 trial. The Lancet Oncology 18: 895–903 doi: 10.1016/s1470-2045(17)30380-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary Fig. 1A-C: Survival curves for (A) overall survival, (B) systemic time to treatment failure, and (C) intracranial time to treatment failure, stratified by EGFR mutation status.
Supplementary Fig. 2A-C: Survival curves for (A) overall survival, (B) systemic time to treatment failure, and (C) intracranial time to treatment failure, stratified by history of radiation treatment within 14 days before initiating pembrolizumab.