PowerPoint Presentationamos3.aapm.org/abstracts/pdf/115-31559-387514-118623.pdf · Meningioma reference plan re-optimized Biological treatment planning using physics information Unkelbach,

8/3/2016

1

Clinical Significance of RBE

Variations in Proton Therapy

H. Paganetti PhD Professor, Harvard Medical School

Director of Physics Research, Massachusetts General Hospital, Radiation Oncology

• Prescriptions are based on dose (physics), not

outcome (biology; tumor control probability (TCP) or

normal tissue complication probability (NTCP))

• The dose in proton therapy is prescribed as

Gy(RBE); RBE is a dose modifying factor

• Proton therapy is using a generic RBE of 1.1

Introduction

Why RBE (relative biological effectiveness) ?

• The RBE is defined as the ratio of doses to reach the same

level of effect when comparing two modalities

Introduction

• RBE for TCP could potentially deduced from tumor control

data

• RBE for NTCP is difficult to assess based on clinical data

because photons generally deliver a more uniform dose to

critical structures and the probability of radiation damage for

a specified dose is sensitive to the volume of normal tissues

irradiated

• The majority of laboratory data are on RBE for cell survival

in vitro

8/3/2016

2

Cells with higher repair capacity (low α/β) show higher RBE

S(D) = e-(aD+bD2)

RBE for cell survival – Endpoint dependency

Carbon ions Photons

© M. Scholz, GSI

Uncertainties due to α/β

Carabe, España, Grassberger, Paganetti:

Phys Med Biol 2013 58: 2103-2117 McNamara, Schuemann, Paganetti:

Phys Med Biol 2015 60; 8399-8416


Inter-patient variability on cell survival RBE can be substantial

Liu, Ghosh, Magpayo, Testa, Tang, Biggs, Paganetti, Efstathiou, Lu, Held, Willers:

Int J Radiat Oncol Biol Phys 2015 91: 1081-1089

“Links Fanconi Anemia/BRCA pathway

defects to elevated proton RBE”

Grosse, Fontana, Hug, Lomax, Coray, Paganetti, Sartori, Pruschy:

Int J Radiat Oncol Biol Phys 2014 88: 175-181

“Repair kinetics in HR-deficient cells

were significantly delayed after proton

irradiation, with elevated amounts of

residual gH2AX foci”


8/3/2016

3

RBE relevant for NTCP: Effect of interest (organ level):

• early effects such as erythema

• late effects such as lung fibrosis, lung function, spinal cord

injury, or necrosis

Typically measured other than cell survival (cellular level):

• Double-strand break induction

• Foci formation

• Chromosome aberrations

• Micronuclei formation

• Cell cycle disruption …


0.001

0.01

0.1

1

0 1 2 3 4 5 6 7 8 Dose [Gy]

Cell

Su

rviv

al

g protons

S(D) = e-(aD+bD2

)

RBE L,dH , a b L

a b L 2

4 a b LRBEmax dH 4RBEmin

2 dH2 a b

L

2dH

RBE≈1.3

RBE≈1.6 2 4 6 8 10

0.8

0.9

1

1.1

1.2

1.3

1.4

1.5

Dose (Gy)

RB

E

(a/b)x = 1 Gy

Our Model

Carabe et al.

Wedenberg et al.

2 4 6 8 100.8

0.9

1

1.1

1.2

1.3

1.4

1.5

Dose (Gy)

RB

E

(a/b)x = 2 Gy

2 4 6 8 100.8

0.9

1

1.1

1.2

1.3

1.4

1.5

Dose (Gy)

RB

E

(a/b)x = 10 Gy

2 4 6 8 100.8

0.9

1

1.1

1.2

1.3

1.4

1.5

Dose (Gy)

RB

E

(a/b)x = 15 Gy

LETd = 2.5 keV/mm

McNamara, Schuemann, Paganetti:

Phys Med Biol 2015 60; 8399-8416

RBE for cell survival – Dose dependency

• Most experiments in vitro look at cell

survival

• Precise measurements of cell

survival below 2 Gy are sparse

• Prescription doses are typically

2Gy/fraction

• There are only a few data points

regarding dose dependency of RBE

in vivo below 4 Gy for protons

Aoki-Nakano et al. J Rad Res 2014

RBE for cell survival – Dose dependency

8/3/2016

4

M. K

räm

er e

t al

.:

Tec

hn.

Can

cer

Res

. T

reat

m.

2, 4

27

-43

6, 2

00

3

Photons Low-LET (Protons)

High-LET (12C)

Radiation is more effective when energy depositions are more concentrated in space

0 5 10 15 20

1

2

3

4

LETd (keV/mm)

RB

E

(a/b)x = 1.22 Gy

0 5 10 15 20

1

2

3

4

LETd (keV/mm)

RB

E

(a/b)x = 2 Gy Our model

Carabe et al.

Wedenberg et al.

Guan et al.

0 5 10 15 20

1

2

3

4

LETd (keV/mm)

RB

E

(a/b)x = 3 Gy

0 5 10 15 20

1

2

3

4

LETd (keV/mm)

RB

E

(a/b)x = 3.49 Gy

0 5 10 15 20

1

2

3

4

LETd (keV/mm)

RB

E

(a/b)x = 10 Gy

0 5 10 15 20

1

2

3

4

LETd (keV/mm)

RB

E

(a/b)x = 15 Gy

Dp = 2 Gy

McNamara, Schuemann, Paganetti:

Phys Med Biol 2015 60; 8399-8416

RBE for cell survival – LET dependency P

ag

anetti: P

hys

. M

ed

. B

iol. 1

998;

43,

2147-2

157

Implication of RBE(LET) for RBE(depth)

Dose = Fluence [1/cm2] × LET [keV/cm] / r [g/cm3]

1

1

2

2

3

3

RBE for cell survival – LET dependency

Entrance: ~1.1 Center: ~1.15 Distal edge: ~1.35 Distal fall-off: ~1.7 (values averaged over all cell lines and SOBPs for in vitro cell survival)

Healthy tissue

RBE for cell survival – LET dependency

Paganetti H: Phys Med Biol 2014 59: R419-R472

1.1 is a conservative estimate!

8/3/2016

5

Increased effectiveness as a function of depth

RBE might be higher close to the ‘target’ edge (mainly in OAR)

Average RBE across a typical SOBP is, on average, about 1.1

RBE depends on LET

RBE seems to be higher for tissues with a low a/b ratio (mainly OAR)

RBE values for endpoints other than cell survival are less well known.

The RBE for normal tissue response is unclear

RBE depends on a/b

RBE increases with decreasing dose

Indicates higher RBE for OAR

Measurements (in vitro and in vivo) typically do not provide high resolution below 2 Gy

RBE depends on dose

RBE for cell survival

Dose x 1.1

Gy(R

BE

)

54

5.4

keV

/μm

5

1 LET

Gy

(RB

E)

59.

4

5.4 Dose x RBE

Clinical evidence ?

Gy(R

BE

)

11

-11

Dose(RBE-weighted) – Dose (RBE=1.1)

Gia

nts

oud

i, S

eth

i, Y

eap

, E

ato

n,

Eb

b, C

aru

so, R

ap

alin

o, C

hen,

Ad

am

s, Y

ock,

Tarb

ell,

Pag

anetti, M

acD

onald

: In

t. J

. R

ad

iat. O

ncol. B

iol. P

hys

. 2016 9

6; 287-2

96

Note:

All 119 cases had similar LET distributions

Only 4 with symptomatic treatment change

Only 1 symptomatic change correlated with LET

Correlation of toxicity and LET

RBE increases with LET

LET is not the sole indicator

Clinical evidence ?

Gia

nts

oud

i, S

eth

i, Y

eap

, E

ato

n,

Eb

b, C

aru

so, R

ap

alin

o, C

hen,

Ad

am

s, Y

ock,

Tarb

ell,

Pag

anetti, M

acD

onald

: In

t. J

. R

ad

iat. O

ncol. B

iol. P

hys

. 2016 9

6; 287-2

96

8/3/2016

6

There is currently no clinical evidence for a correlation

between areas of elevated LET (RBE) and toxicities

Should we consider RBE for NTCP in treatment

planning?

Clinical evidence ?

Planning technique maximizing target conformality

Dose x 1.1 LET Dose x RBE

Gy

(RB

E)

50.4

2.52

keV

/μm

6

1

Gy

(RB

E)

55.4

2.52

Brainstem

Target

Brainstem

Target

Brainstem

Target

Giantsoudi, Adams, Kim, MacDonald, Paganetti: Int. J. Radiat. Oncol. Biol. Phys. 2016 under review

RBE considerations in treatment planning

Dose x 1.1 LET Dose x RBE

Gy

(RB

E

)

50.4

2.52

keV

/μm

6

1

Gy

(RB

E

)

55.4

2.52

Brainstem

Target

Brainstem

Target

Brainstem

Target

Planning technique minimizing maximum LET in the brainstem



8/3/2016

7

!!

RBE = 1.1 stop

RBE = 1.1 through

RBE var. stop

RBE var. through



Passive Scattering

Dose x RBE LET Dose x 1.1

Gy

(RB

E)

54

2.7

Pencil beam scanning

LET Dose x RBE Dose x 1.1

Gy

(RB

E)

54

2.7


Gia

nts

oud

i, A

dam

s, M

acD

onald

, P

ag

anetti:

Int. J

. R

ad

iat. O

ncol. B

iol. P

hys

. 2016 in p

rep

ara

tion

CTV Brainstem Chiasm

Passive Scattering Beam Scanning


Gia

nts

oud

i, A

dam

s, M

acD

onald

, P

ag

anetti:

Int. J

. R

ad

iat. O

ncol. B

iol. P

hys

. 2016 in p

rep

ara

tion

8/3/2016

8

Scanning reduces

the OAR dose for

Scanning

increases the

OAR dose for


Gia

nts

oud

i, A

dam

s, M

acD

onald

, P

ag

anetti:

Int. J

. R

ad

iat. O

ncol. B

iol. P

hys

. 2016 in p

rep

ara

tion

Uncertainties !!

PLAN 2

Dose LETd

PLAN 1

IMPT

Gra

ssb

erg

er,

Tro

fim

ov, Lom

ax,

Pag

anetti:

Int. J

. R

ad

iat. O

ncol. B

iol. P

hys

. 2011 8

0: 1559-1

566

Biological treatment planning using physics information

atypical meningioma

CTV overlaps with

• optic nerve

• chiasm

• brainstem


Unkelbach, Botas, Grassberger, Giantsoudi, Paganetti:

Int. J. Radiat. Oncol. Biol. Phys. 2016 under review

8/3/2016

9

physical dose LET x dose

50 Gy 10 Gy

5 Gy

Scaling of LET x dose such that RBE = 1.1 in center of 5cm SOBP




Meningioma

reference plan re-optimized




Base-of-skull Chordoma

reference plan re-optimized




TH-CD-209-6 (Thursday, August 4, 2016)

10:00 AM - 12:00 PM Room: 209

8/3/2016

10

SUMMARY

Proton therapy uses a generic RBE of 1.1 because of substantial

uncertainties in RBE as a function of dose, endpoint and LET

The RBE is potentially higher towards the distal end of an SOBP and for

low α/β.

The relevance of endpoints other than cell survival for defining clinical

RBEs is unclear.

For a given dose and organ, the RBE dependency on LET is monotone

(reasonably linear)

There is no evidence (yet) for a correlation between LET and toxicity or

recurrence

RBE/LET optimization may improve treatment outcome

Inter-patient variability (biomarkers?) is not well understood

MGH Radiation Oncology Monte Carlo and Biophysics Research Team

Aimee McNamara

Maria Marteinsdottir

Tracy Underwood

David Hall Stephen McMahon

Jan Schuemann Harald Paganetti

Drosoula Giantsoudi

Jan Unkelbach

Hakan Oesten

Clemens Grassberger

Changran Geng

Pablo Botas

Jungwook Shin

Documents

PowerPoint Presentationamos3.aapm.org/abstracts/pdf/115-31559-387514-118623.pdf · Meningioma reference plan re-optimized Biological treatment planning using physics information Unkelbach,