B a s e lin e D a y 1 D a y 2 D a y 3 D a y 40

1 0

2 0

3 0

F o o d In ta k e

Fo

od

in

tak

e (

g/d

ay

) S ham

XRT

*

* **

ObjectiveFatigue, the most common acute and subacute toxicity of

partial or whole brain radiation therapy (WBRT), significantly decreases quality of life (QOL) for patients, abrogating the benefit afforded by improved tumor control. Little is known about the mechanisms underlying brain-radiation induced fatigue, and there are currently no effective treatments. The objective of this study is to evaluate the role of diminished orexin neuron activity in the pathogenesis of fatigue induced by clinically significant WBRT fractionation.

Methods• Adult male Sprague-Dawley rats received whole brain RT (WBRT) 4

Gy x 5 fractions

• Fatigue Behavior• Home Cage Locomotor Activity

• Anthropometrics• Body weight

• Food intake

• Gene expression• Hypothalamic & Hippocampal mRNA (qRT-PCR)

• Inflammation (IL-1β, Tnf, IL-6, Ccl2, IL-1ra, Hmgb1, Gfap, Iba-1, Itgam)

• Oxidative Stress (Nox4, Cyba, Ncf1, Sod1, Sod2)

• CSF collected from 11 pediatric primary brain cancer patients treated with proton radiation at 4 time points: pre-RT, then q 30 days post-RT (post-RT 2, 3, 4),

• CSF cytokine concentrations measured using multiplex magnetic bead immunoassay

• Physician-assessed CTCAE 4.0 fatigue prospectively collected during brain RT CSF Orexin-A concentration• Collected by cisterna magna puncture (rats) or lumbar puncture (patients)

• I125 Radioimmunoassay

• Statistical Analyses• 2-way ANOVA with Bonferroni corrected t-tests

• Student’s t-test

• Logistic Regression for association between fatigue and CSF cytokine levels

• Significance set at p<.05.

Results• LMA was decreased in WBRT-treated rats following the first fraction

and continued to decrease until reaching a nadir following the final fraction. LMA slowly recovered over the next 7 days.

• Food intake and weight gain were significantly reduced in WBRT-treated rats, recovering to match sham rats within 2 weeks of the first fraction.

• Hypothalamic mRNA demonstrated a significant increase in inflammatory gene expression (IL-1b, IL-6, TNF, CCL2, GFAP, Itgam) following 1- and 5-fractions, which largely resolved by fatigue recovery.

• Immunohistochemistry showed increased Iba-1 positive cells with morphology consistent with activated microglia.

• WBRT-treated animals showed decreased CSF [orexin-A] at activity nadir, which recovered with restoration of LMA.

• Orexin levels were decreased and inflammatory cytokine levels increased in CSF samples from pediatric patients for whom fatigue was recorded as a toxicity during treatment.

Conclusions1. WBRT-induced fatigue can be modeled in preclinical research

using home cage LMA.2. The acute fatigue phenotype is observed following the first fraction,

with an activity nadir following the final fraction and recovery approximately one week later.

3. WBRT-induced inflammation follows the behavioral and anthropometric changes, suggesting a possible causative role.

4. Fatigue is associated with decreased CSF orexin levels and increased brain inflammation in rats and patients.

5. Further studies are needed to test the utility of orexin-replacement as a therapeutic strategy for radiation-induced fatigue.

Association of hypothalamic inflammation and cerebrospinal fluid orexin levels with brain radiation-induced fatigue Aaron J. Grossberg MD, PhD1, Die Zhang PhD2, Wei Zhou PhD2, Eric D. Brooks MD3, Connie C. Weng PhD2, Phillip S. Gross BS4, Elisabeth G. VichayaPhD4, Robert Dantzer DVM, PhD4, David R. Grosshans MD, PhD2,3

1Department of Radiation Medicine, Brenden-Colson Center for Pancreatic Care, CEDAR, Oregon Health & Science University, Portland, ORDepartments of 2Experimental Radiation Oncology, 3Radiation Oncology, and 4Symptom Research, MD Anderson Cancer Center, Houston, TX

Figure 3: Hypothalamic inflammatory gene expression. Inflammatory gene expression from hypothalami isolated at fatigue onset, nadir, and recovery. *, p<.05, **, p<.01, ***, p<.001

A

Study Design

B

C D

WBRT 4Gy x 5 Home Cage Locomotor Activity

FATIGUE

Oxidative Damage Inflammatory Cytokine production

Decreased Orexin release

Brain radiation

ROS

IL-1β TNF IL-6

-2 -1 0 1 2 3 4 5 6 7 8 9 1 0 1 10

5 0 0 0

1 0 0 0 0

1 5 0 0 0

T o ta l L o c o m o to r A c t iv ity

D a yT

ota

l A

cti

vit

y(B

ea

m B

rea

ks

/24

h)

S h am

W B R T 4 G y x 5

******

*

X R T

X R T x tim e in te ra c tio n2 -w a y A N O V A p< .0 0 0 1Fatigue Onset

Day 1Fatigue NadirDay 5

Fatigue RecoveryDay 11

Figure 1: Time course of acute WBRT-induced fatigue. Home cage LMA before, during, and after WBRT (4 Gy x 5 fx). N=10/group.

Figure 2: Anthropometrics. Food intake (A, B) and body weight (C, D) decreased by WBRT. *, p<.05, **, p<.01, ***, p<.001

Day 1 (Onset) Day 12 (Recovery)Day 5 (Nadir)

IL 1 b IL 6 T n f C c l2 IL -1 r a H M G B 1 G fa p Itg am Ib a 10

2

47

8

9

1 05 0

6 0

7 0

In f la m m a to ry

Re

lati

ve

E

xp

res

sio

n

S ham

XRT

*** ** ****

IL 1 b IL 6 T N F C C l2 IL -1 r a H M G B 1 G F A P Itg am Ib a 10

2

47

8

9

1 05 0

6 0

7 0

In f la m m a to ry

Re

lati

ve

E

xp

res

sio

n

S ham

XRT

****

****

****

******

**

IL 1 b IL 6 T N F C C l2 IL -1 r a H M G B 1 G F A P Itg am Ib a 10

2

47

8

9

1 05 0

6 0

7 0

In f la m m a to ry

Re

lati

ve

E

xp

res

sio

n

S ham

XRT

**

***

****

****

****

Figure 6: Heatmap of CSF cytokine levels in pediatric patients treated with RT. CSF collected prior to RT, then serially post-RT. Cytokine concentration measured using multiplex magnetic bead immunoassay. All values normalized to pre-RT levels. Fatigue associated with post-RT [sTNFR1] (p=0.04) and [sTNFR2] (p=.04). Statistical analysis using logistic regression.

B a s e line

X R T 1 2 3 4 50

1 0

2 0

3 0

F o o d in ta k e

Fo

od

in

tak

e (

g/d

ay

)

X RT

S h am

****

W e e k s P o s t-X R T

B a s e line

T rea tm

e n tW

k 1W

k 2W

k 3W

k 4W

k 52 0 0

3 0 0

4 0 0

5 0 0

6 0 0

B o d y W e ig h t

Bo

dy

ma

ss

(g

) W B R T - 4 G y x 5 fx

S h am

2 -w a y A N O V A

W B R T x t im e p < .0 0 1

*

***

-3 0 1 2 3 42 2 0

2 4 0

2 6 0

2 8 0

3 0 0

3 2 0

B o d y W e ig h t

D a y

Bo

dy

ma

ss

(g

)

S ham

XRT

*** ****

*

Supported by RSNA R & E Resident Research Grant RR1650

D a y 1 D a y 5 D a y 1 20

5 0 0

1 0 0 0

1 5 0 0

2 0 0 0

C S F [O re x in -A ]

[Ore

xin

-A]

(pg

/mL

)

S ham

XRT

***

X R T x t im e in te ra c tio n p = .0 0 2

Sham

40X

200X

XRT

40X

200X

Figure 4: WBRT activates hypothalamic microglia. Immunohistochemistry for Iba-1 (green) and DAPI (blue).

Iba-1DAPI

Iba-1DAPI

Figure 5: WBRT suppresses orexin release into the CSF. CSF [orexin-A] collected at the end of the dark cycle at each time point. ***, p<.001

Pre-RT Post-RT (2)

Post-RT (3) Post-RT (4)Inflammatory

Non-inflammatory

Pre-RT Post-RT (2)

P re -R T P o s t-R T0

5 0 0

1 0 0 0

1 5 0 0

2 0 0 0

2 5 0 0

F a t ig u e d

[Ore

xin

-A]

(pg

/mL

)

P re -R T P o s t-R T0

5 0 0

1 0 0 0

1 5 0 0

2 0 0 0

2 5 0 0

N o n -F a t ig u e d

[Ore

xin

-A]

(pg

/mL

)

Figure 7: Relationship between changes CSF [Orexin-A] and fatigue. CSF collected prior to RT, then 30 d post-RT. Orexin-A concentration measured using radioimmunoassay. Statistical analysis using 2-way ANOVA.

pinteraction=.026