Abstract
Background
A recent phase III trial (NCT01372774) comparing use of stereotactic radiosurgery [SRS] versus whole-brain radiation therapy [WBRT] after surgical resection of a single brain metastasis revealed that declines in cognitive function were more common with WBRT than with SRS. A secondary endpoint in that trial, and the primary objective in this secondary analysis, was to identify baseline biomarkers associated with cognitive impairment after either form of radiotherapy for brain metastasis. Here we report our findings on APOE genotype and serum levels of associated proteins and their association with radiation-induced neurocognitive decline.
Methods
In this retrospective analysis of prospectively collected samples from a completed randomized clinical trial, patients provided blood samples every 3 months that were tested by genotyping and enzyme-linked immunosorbent assay, and results were analyzed in association with cognitive impairment.
Results
The APOE genotype was not associated with neurocognitive impairment at 3 months. However, low serum levels of ApoJ, ApoE, or ApoA protein (all P < .01) and higher amyloid beta (Aβ 1–42) levels (P = .048) at baseline indicated a greater likelihood of neurocognitive decline at 3 months after SRS, whereas lower ApoJ levels were associated with decline after WBRT (P = .014).
Conclusions
Patients with these pretreatment serum markers should be counseled about radiation-related neurocognitive decline.
Keywords: amyloid beta, apolipoproteins, brain metastases, cognitive decline, radiation therapy
Key Points.
Markers can identify those at high risk for neurocognitive impairment following radiotherapy.
Apolipoproteins and their associated isoforms may confer radiotherapy protective effects.
Importance of the Study.
Previous studies have linked apolipoprotein levels with cognitive changes in a variety of diseases; this is, to the best of our knowledge, the first report of their assessment of metastatic cancer. Our findings indicate that lower serum concentrations of selected apolipoproteins (ApoE, ApoA1, and ApoJ), and perhaps higher levels of Aβ 1–42, may be associated with cognitive decline. Future prospective studies to validate these findings are needed. If our findings are validated, they may be useful for counseling patients about the likelihood of their experiencing neurocognitive decline after postoperative radiation therapy for brain metastases. We previously found the likelihood of neurocognitive decline to be higher after whole brain radiation therapy (WBRT); however, for patients whose baseline biomarkers suggest that they may be at high risk for neurocognitive decline regardless of radiotherapy regimen, perhaps WBRT should be considered to reduce the risk of intracranial recurrence.
Brain metastases (BrMs) are a key cause of morbidity and mortality for patients with cancer 1,2. In the primary analysis of our prospective clinical trial (NCT01372774), postoperative whole-brain radiotherapy (WBRT; 30 Gy in 10 daily 3-Gy fractions or 37.5 Gy in 15 daily 2.5-Gy fractions) to the surgical bed of a single resected BrM was found to be associated with significantly greater declines in cognitive function relative to adjuvant stereotactic radiosurgery (SRS; a 12- to 20-Gy single-fraction dose determined by surgical cavity volume), without appreciable differences in overall survival 3. WBRT in this trial was also linked with inferior local control (ie, recurrence of tumor at unresected cranial metastases; 61.8% SRS vs. 89.2% WBRT at 12 months, P < .00016), leading to a general shift away from WBRT for patients with single resected BrM 3. Nevertheless, large numbers of patients still receive WBRT for intracranial control despite the risk of cognitive toxicity 4.
The mechanisms underlying radiation-related declines in neurocognitive function remain ill-defined; however, elevated oxidative damage triggered by ionizing radiation has been linked with the pathogenesis of Alzheimer disease (AD) 5, and ionizing radiation has been linked with increased risks of developing cardiovascular or neurovascular diseases and dementia 6–8. Patients carrying the apolipoprotein E epsilon 4 (APOE4) allele (present in 16% of the general population) may manifest sporadic AD earlier than non-carriers9, and up to 50% of patients with late-onset familial AD also have an APOE4 allele10. A large study from the AD Genetics Consortium confirmed that each additional copy of the APOE4 allele is associated with an increased risk of AD and a younger age at onset11, although carrying the APOE2 allele conferred considerably lower risk of developing AD11. Moreover, increasing evidence suggests a link between APOE genotypes and response to memantine in patients with AD12. Therefore, although a causal relationship between ionizing radiation to the brain and AD has not been established, whether individuals already carrying risk factors for developing AD may also be more prone to radiation-induced cognitive toxicity is unclear. Here, in an attempt to identify patients who may be at greater risk of cognitive decline after brain radiotherapy, we examined APOE genotypes and levels of various apolipoproteins in serum in patients with BrMs who underwent either WBRT or SRS as part of a completed phase III clinical trial.
Materials and Methods
Study Design and Participants
Details of randomization, masking, cognitive-decline–free survival (CDFS), cognitive testing, SRS, and WBRT are discussed in the primary report of this trial3. CDFS at 3 months (CDFS3) was defined as being alive and without cognitive decline at 3 months (+/-3 weeks) after study enrollment. Cognitive decline at 3 months (CDat3) was defined as a decrease of at least one standard deviation in at least one cognitive test result at 3 months (+/-3 weeks) after study enrollment. In the interest of simplifying the analyses, we did not report serum apolipoprotein findings from patients without serum samples at baseline or from those with incomplete cognitive testing.
Genotype Analysis
Genomic DNA was extracted from peripheral blood specimens by using standard protocols. Genotyping for APOE isoform-defining alleles (rs429358 C/T and rs7412 C/T) was done with custom TaqMan Allelic Discrimination Assays on an ABI 7900HT Fast Real-Time PCR system (Applied Biosystems, Foster City, CA, USA) according to the manufacturer’s instructions. Cluster and genotype calling was done with SDS software v2.2.3 (Applied Biosystems, Foster City, California, USA).
Enzyme-linked Immunosorbent Assay Analysis
Initial-screen ELISA was used to detect interleukin-1β (DLB50, R&D Systems), β nerve growth factor (EHNGF, ThermoFisher Scientific), tumor necrosis factor α (DTA00D, R&D Systems), tumor growth factor β (DY240, R&D Systems), angiopoietin 1 (DANG10, R&D Systems), angiopoietin 2 (DANG20, R&D Systems), vascular endothelial growth factor (DVE00, R&D Systems), angiopoietin-1 receptor (Tie-2) (DTE200, R&D Systems), monocyte chemoattractant protein-1 (MCP-1/CCL-2) (DCP00, R&D Systems), insulin-like growth factor 1 (DG100, R&D Systems), interferon γ (DIF50C, R&D Systems), amyloid beta 1-42 (Aβ1–42) (DAB142, R&D Systems), apolipoprotein J (DCLU00, R&D Systems), apolipoprotein A1 (DAPA10, R&D Systems), apolipoprotein A2 (ab229423, Abcam), apolipoprotein B (DAPB00, R&D Systems), apolipoprotein C (EHAPOC3, ThermoFisher Scientific), apolipoprotein E (EHAPOE, ThermoFisher Scientific), and apolipoprotein J (DCLU00, R&D Systems), to see if any of these factors were associated with cognitive decline. Human plasma amyloid beta 1-42 (Aβ 1–42), apolipoprotein J, apolipoprotein A1, and apolipoprotein E concentrations were assayed with the Milliplex MAP corresponding Magnetic Bead Panel (HNABTMAG-68K and APOMAG-62K; MilliporeSigma) according to the manufacturer’s guidelines and measured on a Luminex LX200 analyzer. This assay was done in triplicate and the median value was used. This work was supported by the National Cancer Institute, National Institutes of Health under award number P30 CA016672 and the ORION core facility at MD Anderson Cancer Center.
Statistical Analyses
Categorical variables were compared with Fisher’s exact tests,13 and serum markers were analyzed with unequal-variance t tests14. Recursive partitioning (rpart) 15 Kaplan–Meier 16, and Cox proportional hazards17 analyses were used to identify optimum cutpoints and analyze subsets of patients with different serum protein levels and CDFS outcomes. The purpose of these analyses was to identify groups at high risk cognitive deterioration or death; both within- and across-arm models were performed. In addition, multivariate Cox proportional hazards models17 were used to evaluate time to CDFS adjusting for radiotherapy type and baseline serum marker levels or patient genotype. Corrections for multiple comparisons were not used because of the exploratory nature of the analysis. Data collection and statistical analyses were conducted by the Alliance Statistics and Data Management Center. All statistical analyses were done with R version 4.0.3 (R Core Team Vienna, Austria, 2020). Analyses were based on the study database frozen on February 18, 2017.
Data Availability
The data generated in this study are available upon request from the corresponding author.
Results
Of the initial 194 patients, 175 had samples available for genotype analysis (93 who received SRS to the surgical bed and 82 WBRT). Baseline characteristics were well balanced between the groups, including the primary tumor histologies, ie, lung (SRS 60% vs. WBRT 58%), radioresistant (melanoma, sarcoma, and renal cell carcinoma) (SRS 11% vs. WBRT 11%), and other (SRS 29% vs. WBRT 31%) (Table 1). The frequency and distribution of APOE genotypes (E2E2, E2E3, E2E4, E3E3, E3E4, and E4E4) resembled those in the general population. Baseline serum protein levels were not statistically different by APOE genotype, sex, age, or combined sex and age (Supplementary Figure S1-2). Similar to the overall study results, among patients with available genotype data, median CDFS was longer after SRS than after WBRT3 (Supplementary Figure S3). CDFS3 was significantly lower after WBRT than after SRS, but only for patients with the E3E3 or heterozygous E3 genotypes (E3E3: 9 of 51 [18%] WBRT vs. 27 of 52 [52%] SRS, P < .001; and E3: 16 of 74 [22%] WBRT vs. 41 of 79 [52%] SRS, P < .001) (Supplementary Table S1). On the other hand, when the treatment groups were analyzed separately, the proportions of patients with CDFS3 (vs. without CDFS3) were no different for those with the E3E3 or E3E4 genotypes (Supplementary Table S2). Within-group analyses of specific genotypes indicated that no other genotypes were associated with different CDFS3 rates (all P ≥ .24) (Supplementary Table S3. Similarly, results of within-arm Cox proportional hazard analyses suggested that APOE genotype did not influence risk of cognitive deterioration or death (Supplementary Table S4).
Table 1.
Baseline Patient Characteristics
| Value or no. of patients (%) | ||||
|---|---|---|---|---|
| WBRT group (n = 82) |
SRS group (n = 93) |
All patients (n = 175) | P value | |
| Age, years | .292 | |||
| Mean (SD) | 61. 8 (9.0) | 60.3 (9.4) | 61.0 (9.2) | |
| Range | 41–81 | 26–83 | 26–83 | |
| Age group | .456 | |||
| <60 years | 29 (35) | 38 (41) | 67 (38) | |
| ≥60 years | 53 (65) | 55 (59) | 108 (62) | |
| Sex | .606 | |||
| Female | 42 (51) | 44 (47) | 86 (49) | |
| Male | 40 (49) | 49 (53) | 89 (51) | |
| Duration of extracranial disease control before study entry * | .883 | |||
| ≤3 months | 45 (55) | 50 (54) | 95 (54) | |
| >3 months | 37 (45) | 43 (46) | 80 (46) | |
| No. of brain metastases | .643 | |||
| 1 | 65 (79) | 71 (76) | 136 (78) | |
| 2–4 | 17 (21) | 22 (24) | 39 (22) | |
| Histology of primary disease | .973 | |||
| Lung | 48 (58) | 56 (60) | 104 (59) | |
| Other | 25 (31) | 27 (29) | 52 (30) | |
| Radioresistant (melanoma, sarcoma, and renal cell carcinoma) | 9 (11) | 10 (11) | 19 (11) | |
| Resection cavity diameter | .821 | |||
| ≤3 cm | 48 (58) | 56 (60) | 104 (59) | |
| >3 cm | 34 (42) | 37 (40) | 71 (41) | |
| ELISA findings available at baseline * | .110 | |||
| Yes | 29 (35) | 44 (47) | 73 (42) | |
| No | 53 (65) | 49 (53) | 102 (58) | |
| Baseline ApoA1 | ||||
| n | 29 | 44 | 73 | |
| Median (IQR) | 108.9 (97.3–122.1) | 130.6 (105.6–158.5) | 116.2 (98.8–138.9) | |
| Range | 82.5–138.1 | 84.1–220.3 | 82.5–220.3 | |
| Baseline ApoE | ||||
| n | 29 | 44 | 73 | |
| Median (IQR) | 17.0 (12.4–25.0) | 29.4 (20.7–36.9) | 24.7 (15.0–34.1) | |
| Range | 3.4–52.7 | 10.7–47.4 | 3.4–52.7 | |
| Baseline ApoJ | ||||
| n | 29 | 44 | 73 | |
| Median (IQR) | 115.8 (94.6–145.2) | 134.9 (98.9–158.4) | 130.3 (97.8–150.0) | |
| Range | 84.5-194.6 | 82.8-218.9 | 82.8-218.9 | |
| Baseline amyloid beta | ||||
| n | 29 | 44 | 73 | |
| Median (IQR) | 66.5 (65.3–80.7) | 65.3 (40.7–78.9) | 65.3 (48.3–80.6) | |
| Range | 40.7–107.5 | 39.1–102.9 | 39.1–107.5 | |
| APOE genotype | .178 | |||
| E2E2 | 1 (1) | 0 (0) | 1 (1) | |
| E2E3 | 3 (4) | 12 (13) | 15 (9) | |
| E2E4 | 2 (2) | 3 (3) | 5 (2) | |
| E3E3 | 51 (62) | 52 (56) | 103 (59) | |
| E3E4 | 20 (25) | 15 (16) | 35 (20) | |
| E4E4 | 1 (1) | 2 (2) | 3 (2) | |
| Not measured | 4 (5) | 9 (10) | 13 (7) | |
| ECOG performance status score | .363 | |||
| 0 | 28 (34) | 36 (39) | 64 (36) | |
| 1 | 49 (60) | 47 (50) | 96 (55) | |
| 2 | 5 (6) | 10 (11) | 15 (9) | |
| Extent of surgery | .132 | |||
| Subtotal resection | 12 (15) | 7 (7) | 19 (11) | |
| Total (gross) resection | 70 (85) | 86 (93) | 156 (89) | |
| Surgical approach | .295 | |||
| Not reported | 0 | 1 | 1 | |
| En-Bloc | 51 (62) | 50 (54) | 101 (58) | |
| Piecemeal | 31 (38) | 42 (46) | 73 (42) |
Values are mean pg/mL (SD) for ApoE and amyloid beta, and mg/dL (SD) for ApoA and ApoJ. Abbreviations: ECOG, Eastern Cooperative Oncology Group; ELISA, enzyme-linked immunosorbent assay; SD, standard deviation; SRS, stereotactic radiosurgery; WBRT, whole-brain radiotherapy
*Defined as months of systemic disease control before study entry.
**In the interest of simplifying the analyses, we did not report ELISA (i.e., serum apolipoprotein) findings from patients without serum samples at baseline or from those with incomplete cognitive testing.
Regarding the serum protein analyses, 73 patients (44 SRS, 29 WBRT) had baseline serum samples and cognitive decline data available for analysis (Table 1). SRS patients with CDat3 had lower mean serum ApoE, ApoA1, and ApoJ concentrations (all P < 0.01), and higher amyloid β-protein (Aβ 1–42) concentrations (P = .048), than did SRS patients without CDat3. WBRT patients with CDat3 had significantly lower ApoJ concentrations (P = .014) than did WBRT patients without CDat3 (Table 2). Analysis of within-arm subgroup differences in CDFS using information from recursive partitioning models revealed that SRS patients with high ApoJ and high ApoA1 levels (12/44, 27%) had a median CDFS interval of 12.4 months (95% confidence interval [CI] 11.2–NA), which was longer than for patients with high ApoJ + low ApoA1 (11/44, 25%) (median 6.5 months [95% CI 6.5–NA]; hazard ratio [HR] 8.6 [95% CI 2.3–32.6], P < .01); SRS patients with low ApoJ (21/44, 47%) had the shortest CDFS time (median 3.3 months [95% CI 3.1–3.7]; HR 51.3 [95% CI 11.7–224], P < .01) (Fig. 1a, Supplementary Figure S4.2). Similarly, WBRT patients with high ApoJ values (12/29, 41%) had slightly—but significantly—longer CDFS (median 3.3 months [95% CI 3.3–NA]) than did WBRT patients with low ApoJ levels (17/29, 59%) (median 2.8 months [95% CI 2.8–3]; HR 2.7 [95% CI 1.2–6.2], P = 0.02) (Figure 1b, Supplementary Figure S4.4). Furthermore, in the Cox proportional hazards model with all serum markers and treatment arm (n = 73), the estimated risk of cognitive deterioration or death in patients receiving WBRT was 2.9 times that of SRS patients (HR 2.9 [95% CI 1.6–5.3], P < .001). Also, an increase of one unit of ApoE resulted in an estimated 4.2% decrease in the estimated risk of CDFS (relative risk –4.2% [95% CI –1.6 to –6.7%], P = 0.001). Similarly, an increase of one unit of ApoJ resulted in an estimated 2.7% decrease (relative risk –2.7% [95% CI –1.7 to –3.7%], P < .001) in the estimated risk of CDFS (Supplementary Table S4).
Table 2.
Mean Marker Levels by Treatment Groups Versus Cognitive Decline Status.
| Patient group | |||
|---|---|---|---|
| Serum marker status | No cognitive decline | Cognitive decline | P value* |
| SRS group | No CDat3, n = 25 | CDat3, n = 19 | |
| ApoE | 32.696 (9.986) | 24.305 (9.406) | .007 |
| Amyloid beta | 58.840 (17.927) | 69.916 (17.853) | .048 |
| ApoA1 | 150.644 (32.367) | 113.147 (23.257) | <.001 |
| ApoJ | 151.772 (30.032) | 111.411 (23.872) | <.001 |
| WBRT group | No CDat3; n = 3 | CDat3; n = 26 | |
| ApoE | 22.233 (12.626) | 19.811 (11.544) | .735 |
| Amyloid Beta | 59.867 (17.110) | 70.246 (18.547) | .364 |
| ApoA1 | 102.467 (7.557) | 110.535 (15.774) | .395 |
| ApoJ | 162.467 (30.266) | 118.938 (26.822) | .014 |
Values are mean pg/mL (SD) for ApoE and Amyloid beta, and mg/dL (SD) for ApoA and ApoJ.
For each marker, means and standard deviations were calculated separately for each treatment group for patients without cognitive decline (NoCDat3) versus with cognitive decline at 3 months (CDat3).
*Means were compared with unequal variance t tests within each marker.
Fig. 1.
Kaplan–Meier plot of cognitive decline-free survival by recursive partitioning analysis among patients treated with (a) stereotactic radiosurgery (SRS) or (b) whole-brain radiation therapy (WBRT). Recursive partitioning analysis (rpart) was performed separately within each treatment group using all 4 serum markers (measured at baseline by enzyme-linked immunosorbent assay) to determine if groups (partitions) of patients could be identified who had different median times to cognitive decline–free survival (CDFS). SRS.P1: ApoJ ≥ 131 mg/L + ApoA1 ≥150 mg/dL. SRS.P2: ApoJ ≥131 mg/L + ApoA1 <150 mg/dL. SRS.P3: ApoJ <131 mg/L. WBRT.P1: ApoJ ≥134 mg/L. WBRT.P1: ApoJ <134 mg/L.
Discussion
Previous prospective studies have shown that patients undergoing non-hippocampal–sparing WBRT for BrMs after either SRS or surgery were at higher risk of decline in cognitive function at 3–4 months after treatment than were patients undergoing SRS only3,18. Here we attempted to determine whether a patient’s APOE genotype or serum apolipoprotein levels was associated with their risk of neurocognitive decline after radiation. We found that APOE genotype was not associated with cognitive decline at 3 months after radiation therapy. However, patients in both arms with low baseline serum ApoJ were more likely to experience cognitive decline at 3 months. Recursive partitioning analysis done internally within each arm also suggested that patients in both arms with low ApoJ had lower median CDFS (SRS, ApoJ <131 mg/L; WBRT, ApoJ <134 mg/L). In addition, two potential lower-risk groups were observed among patients treated with SRS: ApoJ ≥ 131 mg/L + ApoA1 ≥ 150, and ApoJ ≥ 131 mg/L + ApoA1 < 150. These groups seemed to have much longer median time to cognitive decline or death (median 12.4 months and 6.5 months, compared with 3.3 months in the highest-risk SRS group). The impact of this high-ApoA1 group could not be assessed in the WBRT group because the maximum observed ApoA1 level the in WBRT group was 138 (Table 1). The mechanism of neurocognitive decline in these patients seems to be independent of that underlying AD but may have a similar overlap with changes observed in apolipoproteins that are not specific to AD in dementia and other neurocognitive diseases.
ApoJ, also known as clusterin, is an acidic glycoprotein and the second major apolipoprotein in the brain19. ApoJ is also referred to as an extracellular chaperone because of its function in preventing the aggregation of non-native proteins20. In one study, in situ hybridization of human brain specimens detected ApoJ mRNA in astrocytes and in a subset of hippocampal neurons21. ApoJ-containing lipoproteins were first isolated from human plasma and are associated with lipid-poor, protein-rich apoA1-containing high-density lipoprotein in addition to other plasma lipoproteins22. Numerous functional properties have been attributed to ApoJ, including roles in innate immune responses such as complement-mediated lysis and complement lysis inhibitor23. It also acts as an extracellular chaperone with increased expression in response to cellular stress24 and displays anti-apoptotic properties25. Apo J is also capable of interacting with Aβ 1–42, which alters its aggregation26 and seems to promote Aβ 1–42 clearance27. In a prospective study of 196 subjects with mild cognitive impairment, the authors found, after adjustment for potential confounders, a two-fold increase in the risk of conversion to dementia in subjects with low ApoJ serum levels28. This finding is in agreement with a previous study showing that plasma ApoJ levels were higher in patients with mild cognitive impairment than in patients with AD29. In the current study, having low baseline levels of ApoJ seems to indicate susceptibility to neurocognitive decline, and having higher serum baseline levels of ApoJ predict the more favorable situation of not having cognitive decline. These findings mirror those of the previous study, in which the follow-up time was nearly 9 years28. In the future, radiation oncologists may want to consider withholding radiotherapy in patients with low baseline serum levels of ApoJ, reserving it for later in the course of the disease.
ApoE is the most abundant apolipoprotein in the brain and has been extensively studied in the context of AD and dementia. ApoE was initially characterized in the context of human hyperlipidemia30. The liver is the largest production site for ApoE, and the brain is the second largest31. Immunohistochemical staining for ApoE in brains shows that this protein is present in astrocytes32, choroid plexus33, and microglia34, particularly reactive microglia33. ApoE expression has also been detected, albeit to a lesser extent, in pericytes and oligodendrocytes35, and neurons have also been reported to produce ApoE in response to injury33. The neuronal uptake of ApoE lipoprotein particles via ApoE receptors has been implicated in brain homeostasis, synaptic integrity, and synaptic function36. ApoE, like ApoJ, also has known functions related to immune response. ApoE has been shown to suppress T-cell proliferation37 and neutrophil activation38, regulate macrophage functions39, facilitate the presentation of the lipid antigen by CD1 molecules to natural killer T cells40,41, and modulate inflammation42. A recent epidemiologic analysis showed an association between low plasma ApoE levels and increased risk of future dementia in the general population; this association was independent of APOE genotype and has been confirmed in different patient populations43.
ApoA1 is one of the most abundant apolipoproteins in the cerebrospinal fluid. Because ApoA1 is thought to be produced mainly in the liver and intestines, the presence of ApoA1 in the CNS is thought to originate from the periphery44. In vivo models have shown that ApoA1 can enter the CNS via the choroid plexus, and it can be taken up by human epithelial and endothelial cells in in vitro models of the blood–brain barrier45. However, prospective studies of plasma ApoA1 concentrations and the risk of dementia or cognitive decline are sparse and have produced inconsistent results. However, in the large-scale Honolulu-Asia Aging Study, higher concentrations of plasma ApoA1 were found to be associated with a lower risk of dementia46.
We acknowledge that the current study had limitations. First, the results were derived from analyses of serum samples from a heterogeneous group of patients, and we could not fully control for cancer type, extent of systemic disease, comorbidities, and systemic cancer treatments that may have affected baseline levels of the measured variables. Second, the relatively small number of patients in specific subgroups also limited our power to make formal comparisons between groups within the Cox proportional hazard models. Third, the numbers of patients with complete data on cognitive decline and serum apoproteins were not equal between arms, which led to unbalanced distribution of potential confounding or prognostic factors. In addition, we observed significant differences in distribution of baseline characteristics of ApoA1 and ApoE between study arms, which limited our ability to make cross-arm conclusions about the influence of recursive partitioning-identified groups on overall CDFS. Finally, the choice to remove patients without complete cognitive testing allows easier interpretation of results but limits full determination of the relationship between biomarker serum profile and cognitive decline–free survival for this study. Further study is warranted before implementing these cutoffs in clinical practice.
In sum, although previous studies have linked levels of ApoE47, ApoA148, ApoJ28, and Aβ 1–4249 with cognitive changes in a variety of diseases, this is, to the best of our knowledge, the first report of their assessment in metastatic cancer50. Our findings indicate that lower serum concentrations of selected apolipoproteins, and perhaps higher levels of Aβ 1–42, may be associated with cognitive decline. Future prospective studies to validate these findings are needed. If our findings are validated, they may be useful for counseling patients about the likelihood of their experiencing neurocognitive decline after postoperative radiation therapy for BrMs. This analysis allowed us to identify observable differences in CDFS among different biomarker-driven subsets. Because of limitations in study design and availability of lab data and complete cognitive test results, our results should not be extrapolated beyond scope of this study. However, this is sufficient evidence to continue to collect serum samples for biomarker analyses in these patients and to analyze their association with CDFS in a larger study.
Supplementary Material
Acknowledgements
The Alliance funder provided peer-reviewed approval of the trial but had no other role in study design, data collection, data analysis, data interpretation, or writing of the report. The Alliance Statistics and Data Center was responsible for the collection, maintenance, and analyses of the data. Data confidentiality was governed by National Institutes of Health policy.
Contributor Information
Kristin Huntoon, Department of Neurosurgery, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA; The Brain Tumor Center, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA.
S Keith Anderson, Alliance Statistics and Data Management Center, Mayo Clinic, Rochester, Minnesota, USA.
Karla V Ballman, Alliance Statistics and Data Management Center, Mayo Clinic, Rochester, Minnesota, USA; Department of Biostatistics and Epidemiology, Weill Medical College of Cornell University, New York, New York, USA.
Erin Twohy, Alliance Statistics and Data Management Center, Mayo Clinic, Rochester, Minnesota, USA.
Katharine Dooley, Alliance Statistics and Data Management Center, Mayo Clinic, Rochester, Minnesota, USA.
Wen Jiang, Department of Radiation Oncology, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA.
Yi An, Department of Therapeutic Radiology, Yale-New Haven Hospital, North Haven, Connecticut, USA.
Jing Li, Department of Radiation Oncology, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA.
Christina von Roemeling, Department of Neurosurgery, University of Florida, Gainesville, Florida, USA.
Yaqing Qie, Department of Neurosurgery, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA; The Brain Tumor Center, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA.
Owen A Ross, Department of Neuroscience, Mayo Clinic, Jacksonville, Florida, USA.
Jane H Cerhan, Department of Psychiatry and Psychology Mayo Clinic, Rochester, Minnesota, USA.
Anthony C Whitton, Department of Radiation Oncology, McMaster University, Hamilton, Ontario, Canada.
Jeffrey N Greenspoon, Department of Radiation Oncology, McMaster University, Hamilton, Ontario, Canada.
Ian F Parney, Department of Neurosurgery, Mayo Clinic, Rochester, Minnesota, USA.
Jonathan B Ashman, Department of Radiation Oncology, Mayo Clinic, Phoenix/Scottsdale, Arizona, USA.
Jean-Paul Bahary, Department of Radiation Oncology, CHUM, Montreal, Quebec, Canada.
Constantinos Hadjipanayis, Department of Neurosurgery, Icahn School of Medicine at Mount Sinai, New York, New York, USA.
James J Urbanic, Department of Radiation Oncology, University of California San Diego, Moores Cancer Center, La Jolla, California, USA.
Elana Farace, Department of Public Health Sciences, Penn State University College of Medicine, Hershey, Pennsylvania, USA.
Deepak Khuntia, Department of Radiation Oncology, Precision Cancer Specialists and Varian Medical Systems, Palo Alto, California, USA.
Nadia N Laack, Department of Radiation Oncology, Mayo Clinic, Rochester, Minnesota, USA.
Paul D Brown, Department of Radiation Oncology, Mayo Clinic, Rochester, Minnesota, USA.
David Roberge, Department of Radiation Oncology, CHUM, Montreal, Quebec, Canada.
Betty Y S Kim, Department of Neurosurgery, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA; The Brain Tumor Center, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA.
Conflicts of Interest
The authors declare no potential conflicts of interest.
Funding
This work was supported by the National Cancer Institute of the National Institutes of Health under award numbers [U10CA180821], [U10CA180882], and [U24CA196171 (to the Alliance for Clinical Trials in Oncology)], [UG1CA233180], [UG1CA233329], [UG1CA232760[, [U10CA180863 (CCTG)], and [P30CA016672]. https://acknowledgments.alliancefound.org. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Author Contributions
Experimental design: KH, SA, KB, WJ, PB, DR, BYSK. Implementation: KH, SA, KB, WJ, YQ, PB, DR, BYSK. Analysis and interpretation of the data: KH, SA, KN, WJ, YA, JLCR, YQ, OR, JC, AW, JG, IP, JA, JB, CH, JU, EF, DK, NL, PB, DR, BYSK. All authors have been involved in the writing of the manuscript at draft and any revision stages and have read and approved the final version.
References
- 1. Aoyama H, Shirato H, Tago M, et al. Stereotactic radiosurgery plus whole-brain radiation therapy vs stereotactic radiosurgery alone for treatment of brain metastases: a randomized controlled trial. JAMA. 2006;295(21):2483–2491. [DOI] [PubMed] [Google Scholar]
- 2. Cagney DN, Martin AM, Catalano PJ, et al. Incidence and prognosis of patients with brain metastases at diagnosis of systemic malignancy: a population-based study. Neuro-oncology. 2017;19(11):1511–1521. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Brown PD, Ballman KV, Cerhan JH, et al. Postoperative stereotactic radiosurgery compared with whole brain radiotherapy for resected metastatic brain disease (NCCTG N107C/CEC.3): a multicentre, randomised, controlled, phase 3 trial. Lancet Oncol. 2017;18(8):1049–1060. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Sperduto PW, Mesko S, Li J, et al. Survival in patients with brain metastases: summary report on the updated diagnosis-specific graded prognostic assessment and definition of the eligibility quotient. J Clin Oncol. 2020;38(32):3773–3784. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Begum N, Wang B, Mori M, Vares G. Does ionizing radiation influence Alzheimer’s disease risk? J Radiat Res. 2012;53(6):815–822. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Asai A, Matsutani M, Kohno T, et al. Subacute brain atrophy after radiation therapy for malignant brain tumor. Cancer. 1989;63(10):1962–1974. [DOI] [PubMed] [Google Scholar]
- 7. Lowe XR, Bhattacharya S, Marchetti F, Wyrobek AJ. Early brain response to low-dose radiation exposure involves molecular networks and pathways associated with cognitive functions, advanced aging and Alzheimer’s disease. Radiat Res. 2009;171(1):53–65. [DOI] [PubMed] [Google Scholar]
- 8. Sugihara S, Ogawa A, Nakazato Y, Yamaguchi H. Cerebral beta amyloid deposition in patients with malignant neoplasms: its prevalence with aging and effects of radiation therapy on vascular amyloid. Acta Neuropathol. 1995;90(2):135–141. [DOI] [PubMed] [Google Scholar]
- 9. Caselli RJ, Dueck AC, Osborne D, et al. Longitudinal modeling of age-related memory decline and the APOE epsilon4 effect. N Engl J Med. 2009;361(3):255–263. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Teunissen CE, de Vente J, Steinbusch HW, De Bruijn C. Biochemical markers related to Alzheimer’s dementia in serum and cerebrospinal fluid. Neurobiol Aging. 2002;23(4):485–508. [DOI] [PubMed] [Google Scholar]
- 11. Reiman EM, Arboleda-Velasquez JF, Quiroz YT, et al. ; Alzheimer’s Disease Genetics Consortium. Exceptionally low likelihood of Alzheimer’s dementia in APOE2 homozygotes from a 5,000-person neuropathological study. Nat Commun. 2020;11(1):667. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Belitskaya-Lévy I, Dysken M, Guarino P, et al. Impact of apolipoprotein E genotypes on vitamin E and memantine treatment outcomes in Alzheimer’s disease. Alzheimer’s Dementia: Transl Res Clin Interventions. 2018;4:344–349. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Fisher RA. On the interpretation of χ 2 from contingency tables, and the calculation of P. J Roy Stat Soc. 1922;85(1):87–94. [Google Scholar]
- 14. Student. The probable error of a mean. Biometrika. 1908;1:25. [Google Scholar]
- 15. Therneau T, Beth A.. rpart: Recursive Partitioning and Regression Trees. R package version 4.1-15; 2019; https://CRAN.R-project.org/package=rpart. [Google Scholar]
- 16. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc. 1958;53(282):457–481. [Google Scholar]
- 17. Cox DR. Regression models and life-tables. J Roy Stat Soc: Ser B (Methodol). 1972;34(2):187–202. [Google Scholar]
- 18. Chang EL, Wefel JS, Hess KR, et al. Neurocognition in patients with brain metastases treated with radiosurgery or radiosurgery plus whole-brain irradiation: a randomised controlled trial. Lancet Oncol. 2009;10(11):1037–1044. [DOI] [PubMed] [Google Scholar]
- 19. Raulin A-C, Martens YA, Bu G. Lipoproteins in the central nervous system: from biology to pathobiology. Annu Rev Biochem. 2022;91. https://www.annualreviews.org/doi/abs/10.1146/annurev-biochem-032620-104801?casa_token=MFXY9YbfglkAAAAA%3ASY8kxGucb-vdutYc5y9XMBHsxKK-x67twFBnRZls-oTJZcDzNcKLIMrUOi49CnAWE5O_zMYyPdNZ. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Jones SE, Clusterin JC. The international journal of biochemistry & cell biology. 2002;34(5):427–431. [DOI] [PubMed] [Google Scholar]
- 21. Pasinetti GM, Johnson SA, Oda T, Rozovsky I, Finch CE. Clusterin (SGP-2): a multifunctional glycoprotein with regional expression in astrocytes and neurons of the adult rat brain. J Comp Neurol. 1994;339(3):387–400. [DOI] [PubMed] [Google Scholar]
- 22. De Silva H, Stuart WD, Duvic C, et al. A 70-kDa apolipoprotein designated ApoJ is a marker for subclasses of human plasma high density lipoproteins. J Biol Chem. 1990;265(22):13240–13247. [PubMed] [Google Scholar]
- 23. Oda T, Wals P, Osterburg HH, et al. Clusterin (apoJ) alters the aggregation of amyloid β-peptide (Aβ1-42) and forms slowly sedimenting Aβ complexes that cause oxidative stress. Exp Neurol. 1995;136(1):22–31. [DOI] [PubMed] [Google Scholar]
- 24. Humphreys DT, Carver JA, Easterbrook-Smith SB, Wilson MR. Clusterin has chaperone-like activity similar to that of small heat shock proteins. J Biol Chem. 1999;274(11):6875–6881. [DOI] [PubMed] [Google Scholar]
- 25. Lamoureux F, Thomas C, Yin M-J, et al. Clusterin inhibition using OGX-011 synergistically enhances Hsp90 inhibitor activity by suppressing the heat shock response in castrate-resistant prostate cancer. Cancer Res. 2011;71(17):5838–5849. [DOI] [PubMed] [Google Scholar]
- 26. Ghiso J, Matsubara E, Koudinov A, et al. The cerebrospinal-fluid soluble form of Alzheimer’s amyloid β is complexed to SP-40, 40 (apolipoprotein J), an inhibitor of the complement membrane-attack complex. Biochem J. 1993;293(1):27–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Matsubara E, Soto C, Governale S, Frangione B, Ghiso J. Apolipoprotein J and Alzheimer’s amyloid β solubility. Biochem J. 1996;316(2):671–679. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Romagnoli T, Ortolani B, Sanz JM, et al. Serum Apo J as a potential marker of conversion from mild cognitive impairment to dementia. J Neurol Sci. 2021;427:117537. [DOI] [PubMed] [Google Scholar]
- 29. Mullan G M, McEneny J, Fuchs M, et al. Plasma clusterin levels and the rs11136000 genotype in individuals with mild cognitive impairment and Alzheimer’s disease. Curr Alzheimer Res. 2013;10(9):973–978. [DOI] [PubMed] [Google Scholar]
- 30. Havel RJ, Kane JP. Primary dysbetalipoproteinemia: predominance of a specific apoprotein species in triglyceride-rich lipoproteins. Proc Natl Acad Sci USA. 1973;70(7):2015–2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Elshourbagy NA, Liao WS, Mahley RW, Taylor JM. Apolipoprotein E mRNA is abundant in the brain and adrenals, as well as in the liver, and is present in other peripheral tissues of rats and marmosets. Proc Natl Acad Sci USA. 1985;82(1):203–207. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Zhao N, Liu C-C, Qiao W, Bu G. Apolipoprotein E, receptors, and modulation of Alzheimer’s disease. Biol Psychiatry. 2018;83(4):347–357. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Xu Q, Bernardo A, Walker D, et al. Profile and regulation of apolipoprotein E (ApoE) expression in the CNS in mice with targeting of green fluorescent protein gene to the ApoE locus. J Neurosci. 2006;26(19):4985–4994. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Huynh T-PV, Wang C, Tran AC, et al. Lack of hepatic apoE does not influence early Aβ deposition: observations from a new APOE knock-in model. Mol Neurodegener. 2019;14(1):1–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Nelissen K, Mulder M, Smets I, et al. Liver X receptors regulate cholesterol homeostasis in oligodendrocytes. J Neurosci Res. 2012;90(1):60–71. [DOI] [PubMed] [Google Scholar]
- 36. Holtzman DM, Pitas RE, Kilbridge J, et al. Low density lipoprotein receptor-related protein mediates apolipoprotein E-dependent neurite outgrowth in a central nervous system-derived neuronal cell line. Proc Natl Acad Sci USA. 1995;92(21):9480–9484. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Avila E, Holdsworth G, Sasaki N, Jackson R, Harmony J. Apoprotein E suppresses phytohemagglutinin-activated phospholipid turnover in peripheral blood mononuclear cells. J Biol Chem. 1982;257(10):5900–5909. [PubMed] [Google Scholar]
- 38. Terkeltaub RA, Dyer CA, Martin J, Curtiss LK. Apolipoprotein (apo) E inhibits the capacity of monosodium urate crystals to stimulate neutrophils. Characterization of intraarticular apo E and demonstration of apo E binding to urate crystals in vivo. J Clin Invest. 1991;87(1):20–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Barger SW, Harmon AD. Microglial activation by Alzheimer amyloid precursor protein and modulation by apolipoprotein E. Nature. 1997;388(6645):878–881. [DOI] [PubMed] [Google Scholar]
- 40. Elzen P, Garg S, León L, et al. Apolipoprotein-mediated pathways of lipid antigen presentation. Nature. 2005;437(7060):906–910. [DOI] [PubMed] [Google Scholar]
- 41. Borg NA, Wun KS, Kjer-Nielsen L, et al. CD1d–lipid-antigen recognition by the semi-invariant NKT T-cell receptor. Nature. 2007;448(7149):44–49. [DOI] [PubMed] [Google Scholar]
- 42. Ophir G, Amariglio N, Jacob-Hirsch J, et al. Apolipoprotein E4 enhances brain inflammation by modulation of the NF-κB signaling cascade. Neurobiol Dis. 2005;20(3):709–718. [DOI] [PubMed] [Google Scholar]
- 43. Rasmussen KL, Tybjærg-Hansen A, Nordestgaard BG, Frikke-Schmidt R. Plasma levels of apolipoprotein E and risk of dementia in the general population. Ann Neurol. 2015;77(2):301–311. [DOI] [PubMed] [Google Scholar]
- 44. Koch S, Donarski N, Goetze K, et al. Characterization of four lipoprotein classes in human cerebrospinal fluid. J Lipid Res. 2001;42(7):1143–1151. [PubMed] [Google Scholar]
- 45. Stukas S, Robert J, Lee M, et al. Intravenously injected human apolipoprotein A-I rapidly enters the central nervous system via the choroid plexus. J Am Heart Assoc. 2014;3(6):e001156. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46. Saczynski JS, White L, Peila RL, Rodriguez BL, Launer LJ. The relation between apolipoprotein AI and dementia: the Honolulu-Asia aging study. Am J Epidemiol. 2007;165(9):985–992. [DOI] [PubMed] [Google Scholar]
- 47. Wang C, Yu JT, Wang HF, et al. Meta-analysis of peripheral blood apolipoprotein E levels in Alzheimer’s disease. PLoS One. 2014;9(2):e89041. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48. Zuin M, Cervellati C, Trentini A, et al. Association between serum concentrations of apolipoprotein A-I (ApoA-I) and Alzheimer’s disease: systematic review and meta-analysis. Diagnostics (Basel). 2021;11(6):984. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49. Nakamura A, Kaneko N, Villemagne VL, et al. High performance plasma amyloid-beta biomarkers for Alzheimer’s disease. Nature. 2018;554(7691):249–254. [DOI] [PubMed] [Google Scholar]
- 50. Luo J, Song J, Feng P, et al. Elevated serum apolipoprotein E is associated with metastasis and poor prognosis of non-small cell lung cancer. Tumour Biol. 2016;37(8):10715–10721. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data generated in this study are available upon request from the corresponding author.