Dosimetric Effects of Using

Generalized Equivalent Uniform Dose

(gEUD) in Plan Optimization

Ontida Apinorasethkul, MS, CMD

AAMD – Indianapolis
June 14, 2017
Disclaimer
This presentation is not sponsored by any vendor nor is it
endorsing one treatment planning system over another. The
material presented is based on our planning experience. This
talk is meant for educational purposes.

2
Overview
• What is gEUD?

• α values

• gEUD optimization

• Benefits and drawbacks

• Plan comparisons against dose volume- (DV) based objectives

• Prostate
• Whole pelvis
• Brain
• H&N
• Lung

3
What is gEUD?
• Generalized Equivalent Uniform Dose

• “The dose that, if given uniformly to the entire organ, is

believed to yield the same complication rate as the true dose
distribution” (QUANTEC)
• Accommodates the concept of serial and parallel organs

• Achieve same tumor response, lower normal tissue dose

• More non-uniform dose distribution

• Less dependent on the variability of patient’s shapes and

sizes
• Less trial and error when planning one case to another

4
What is gEUD?
• Eclipse 13.5 and newer
• New photon optimizer (PO) algorithm
– Supports biological optimization objectives
• Implement biological planning in daily treatment planning
– DV = biological response
– gEUD = estimates of the biological outcome
• IMRT and RA optimization

• Shift to a non-uniform target dose distribution?

• Organ motion
• Daily set up uncertainties

• Brachytherapy
• SRS/SBRT
• Simultaneous integrated boost

5
Workout analogy
• Goal: lose 5 pounds in 2 months

• Cardio only
• Weight training only
• Combinations?

• End result: Weight loss is equivalent to 5 pounds

6
gEUD optimization
For tumors, the EUD represents the biologically equivalent dose
which, if given uniformly, leads to the same cell kill in the tumor
volume as the actual non-uniform IMRT dose-distribution.

EUD =

TG-166

7
α values
• gEUD is dependent on α value.
• Tissue-specific parameter that describes the volume effect of gEUD
• Defines where on the DVH the optimization is to be focused on
• Values range from -40 to +40, but can never be zero
• -40 to -1 = tumor range
• 1 = mean objective
• 1 to 40 = organs at risk range

• Low α values: parallel organs

• High α values: serial organs

8
α values
• TG-166 report

9
α values
• Quantec / Burman

1
0
gEUD optimization
• Target gEUD
• α of -40 to -1
• Defines an exact Equivalent Uniform
dose value
– Ex: PTV must receive 95% of
dose.
• Similar to lower objective of DV-
based optimization

• Lower gEUD
• α of -40 to -1
• Defines the minimum Equivalent
Uniform dose value Varian
– Ex: PTV must receive at least
60 Gy.
• Works well on overlapping
structures
1
1
gEUD optimization
• Upper gEUD
• α of 40 to 1
• Defines the maximum Equivalent
Uniform Dose value
– Ex: Inner ear may receive no
more than 25 Gy.

• Lower α value in the upper gEUD

will affect majority of the DVH
curve

• Higher α value in the upper gEUD

will affect max dose of the DVH Varian
curve

1
2
α values
• Different alpha values
• Same priority value

1
3
Benefits and drawbacks
• Typically lower normal tissue dose while maintaining tumor
coverage

• gEUD objectives are generally used for organs at risk

• Instead of several upper objectives on multiple contours
– Smaller number of parameters used, larger space of solutions
become available
– Less ambiguity in less number of objectives
• Limit max dose
• Lower mean dose

• Limited number of parameters controlled by user

• Details of dose distribution can’t be fine tuned

1
4
gEUD optimization
• With our experience…..
• Using only gEUD objectives on target volumes could result in a highly
non-uniform dose distribution
• Avoid target dose heterogeneities that would not be accepted in
clinical plans
– Evaluate plans for non-uniform dose

1
5
gEUD optimization
• With our experience…..
• Use upper and lower objectives of DV-based optimization for a better
control over target dose distribution
• gEUD works better on normal tissue optimizations

1
6
DV-based optimization objectives

1
7
gEUD based optimization objectives

1
8
Plan comparisons

1
9
Plan comparisons
• Research phase

• Same field parameters

• Arcs / collimator angles
• Field gantry

• DV-based optimization
• Multiple upper objectives on organs at risk
• Optimize on crop structures

• gEUD-based optimization
• PTV objectives - same as DV-based optimized plan
• Optimize on whole organs at risk structure
• One upper gEUD objective for each structure*
• α of 1, 20, or 40

2
0
Prostate plan
• DV-based objectives

• gEUD-based objectives

2
1
Prostate plan
DV gEUD

2
2
Prostate plan
DV gEUD

2
3
Prostate plan
DV gEUD

2
4
Prostate plan

Mean dose (cGy) DV gEUD

Bladder 2070 1781

Rectum 2515 2503

2
5
Prostate plan

Mean dose (cGy) DV gEUD

Femoral head_L 1233 870

Femoral head_R 1318 895

2
6
Pelvis plan
• DV-based objectives

2
7
Pelvis plan
• gEUD-based objectives

2
8
Pelvis plan
DV gEUD

2
9
Pelvis plan
DV gEUD

3
0
Pelvis plan
DV gEUD

Dose max 109.5% Dose max 116.2%

3
1
Pelvis plan

Mean dose (cGy) DV gEUD

Small bowel 1710 1460

Large bowel 3087 2637

Rectum 3091 2967

Bladder 3715 3465

3
2
Brain plan
• DV-based objectives

3
3
Brain plan
• gEUD-based objectives

3
4
Brain plan
DV gEUD

3
5
Brain plan
DV gEUD

3
6
Brain plan
DV gEUD

3
7
Brain plan
DV gEUD

3
8
Brain plan

Organs DV gEUD

Brainstem (max) 5753 5264

Temporal lobe_L (mean) 2882 2600

3
9
H&N plan
• DV-based objectives

4
0
H&N plan
• gEUD-based objectives

4
1
H&N plan
DV gEUD

4
2
H&N plan
DV gEUD

4
3
H&N plan
DV gEUD

4
4
H&N plan
DV gEUD

4
5
H&N plan
DV gEUD

4
6
H&N plan

Max dose (cGy) DV gEUD

Brainstem 3526 2832
Cord 4074 2182
Cord+5mm 4623 2757

4
7
H&N plan

Mean dose (cGy) DV gEUD

Oral cavity 1261 1340
Larynx 2223 2511
Constrictor 3271 2952

4
8
H&N plan

Mean dose (cGy) DV gEUD

Submandibular_L 3145 2928
Parotid_L 2871 2642
Parotid_R 1620 1503

4
9
Lung plan
• DV-based objectives

• gEUD-based objectives

5
0
Lung plan
DV gEUD

5
1
Lung plan
DV gEUD

5
2
Lung plan
DV gEUD

5
3
Lung plan
DV gEUD

5
4
Lung plan
DV gEUD

5
5
Lung plan
• IMRT – same fields

Max dose (cGy) DV gEUD

Brachial plexus_R 4419 3695
Cord 3608 1407
Cord+5mm 3977 2083

5
6
Summary
• With one iteration of optimization, gEUD optimized plan
generally shows lower dose to organs at risk

• Fine tuning is still needed

• Higher global max

• In general, gEUD based optimization can help improve normal

tissue sparing while maintaining the same or better target
volume coverage

5
7
Summary
• Use radiobiological based model tools/algorithms to our
benefits – with caution

• Results in non-uniform target dose

• No one method is perfect

• More research needed of new function

• Use of gEUD on target volumes in the future

5
8
Acknowledgements
• Derek Dolney, PhD
• Akhil Tiwari, CMD
• Dimitris Mihailidis, PhD

5
9
6
0