The document discusses how tumor radiobiology may impact the move toward hypofractionation in radiation therapy. It reviews evidence that the classic linear-quadratic model does not fully capture tumor response at high radiation doses, and that other factors like vascular damage and immune effects become more important. Response heterogeneity between different tumor cell populations may also help explain why survival curves appear more linear at higher doses per fraction.

1. Does the tumour radiobiology
help in the move toward
hypofractionation?
Michael Joiner
Wayne State University
Radiation Oncology
Detroit, Michigan
joinerm@wayne.edu

2. Does the tumour radiobiology
hinder in the move toward
hypofractionation?
Michael Joiner
Wayne State University
Radiation Oncology
Detroit, Michigan
joinerm@wayne.edu

3. What radiobiology?
Historically…
• Research has focused on doses per fraction
of 1–3 Gy: clinically relevant
• Radiobiology is quite mature
• Little incentive/funding for high-dose research
MCJ
Aug 12
3

4. Why reconsider high dose fractions?
• Because we can: Physics
• Patient convenience and demand
• Lower cost of whole treatment
• Evidence that it is very effective
• Evidence of low !/" in some sites
The radiobiology has not yet caught up
MCJ
Aug 12
4

5. Why not LQ at high doses?
• Reponse is really linear at higher doses?
• Vascular damage?
• Increased apotosis?
• Immunological effects?
• Mixed tumor cell populations with different
response characteristics?
Answers likely depend on tissue type
and tumor type / stage
MCJ
Aug 12
5

6. Hypoxic cells in tumors
First demonstration: ~1.5% of the cells were hypoxic
0.015
Mouse
lymphosarcoma
irradiated in vivo
(Powers and Tolmach, 1973)
MCJ
Aug 12
6

7. The classic
concept of
reoxygenation
during
fractionated
radiotherapy
MCJ
Aug 12
7

8. Reoxygenation during fractionation
0
10
No -reoxygenation
-1
10
-2
10
-3
Surviving Fraction
10
-4
10
-5
10
-6
10
-7
10
30 fractions
-8
10
No Hypoxic Cells
-9
10
-10
10
0
10
20
30
40
50
60
Total Dose (2Gy Fractions)
MCJ
Aug 12
8

9. Courtesy:
Bert van der Kogel

10. Courtesy:
Bert van der Kogel
Vessels (CD31, white)
Hoechst 33342, blue
Hypoxic marker 1:
Pimonidazole (-4.5 h)
Hypoxic marker 2:
CCI-103F (-2.5h)
Overlap: yellow

11. Acute hypoxia may not affect proliferation
Proliferating
cells
(Ki67)
blood vessel
(CD31)
Hypoxia
(pimonidazole)
Courtesy:
George Wilson
MCJ
Aug 12
11

12. Total dose (Gy) – various isoeffects
Early
Late
Yes
No
Dose/fraction (Gy)
Thames et al. Int J Radiat Oncol Biol Phys 1982;8:219
MCJ
Aug 12
12

13. Semin Radiat Oncol 2008;18:234–9
1. Mechanistic, biologically based No
2. Few parameters ! practical Yes
3. Other models predict similar dose-fractionation No
4. Well-documented predictive value in Lab Yes
5. Validated up to 10 Gy per fraction, OK to 18 No

14. Semin Radiat Oncol 2008;18:240–3
1. LQ model derived mostly in vitro No
2. LQ underestimates high-dose tumor control ???
3. LQ ignores cell subpopulations Yes
4. LQ mechs don’t reflect vascular, stromal Yes
5. Need understanding mol mechs and stem cells No

15. Linear
Quadratic
model
! ln(S) = " D + # D
!/" = 3 Gy
-1
Effective D0
-2
1 (! + 2" D)
10
10
Surviving fraction
Effective D0
0
10
2
-3
10
LQ
-4
10
-5
10
-6
10
-7
10
-8
10
-9
SF2 = 0.5
10
0
5
10
Dose (Gy)
15
20

16. Effective D0 is too small at high doses
MCJ
Aug 12
16

17. -1
10
-2
10
Surviving fraction
Linear
Quadratic
must
straighten
at high
doses
0
10
-3
10
-4
10
-5
10
-6
10
-7
10
LQ
-8
10
-9
10
0
5
10
Dose (Gy)
15
20

18. -1
10
!ln(S) = "D + #D 2 ! $ D 3
-2
10
Surviving fraction
The
Linear
Quadratic
Cubic
model
0
10
LQC
-3
10
-4
10
-5
10
(
! = " 3DL
-6
10
-7
!/" = 3 Gy
10
LQ
-8
10
SF2 = 0.5
!ln(S) = "D + #D 2
-9
10
0
5
10
Dose (Gy)
15
20
)

19. Curtis' LPL model
Complex DSB
!
Simple DSB
"
Curtis SB. Radiat Res 1986;106:252

20. Response heterogeneity
Alternative damage response pathways
and/or cell types which are dose dependent

21. Vascular effects occur at high doses
0 Gy
5 Gy
2 Gy
10 Gy
Functional
intravascular
volume
Walker 256 tumors (s.c.)
grown in legs of
Sprague-Dawley rats
Single dose radiation
30 Gy
60 Gy
Park HJ et al.
Radiat Res 2012;177:311–27
MCJ
21
Aug 12

22. Vascular effects occur at high doses
Breast cancer patients
Endothelial cells from normal breast or cancer
Normal
Cancer
Normal
Cancer
Park HJ et al. Radiat Res 2012;177:311–27
MCJ
22
Aug 12

23. Vascular effects occur at high doses
hypoxic
oxic
resistant
endothelial
sensitive
endothelial
Park HJ et al. Radiat Res 2012;177:311–27
MCJ
23
Aug 12

24. Endothelial cell apoptosis at high doses
Lung 20 Gy
Wild type
p53 –/–
ASMase –/–
Intestines 15 Gy
Wild type
p53 –/–
ASMase –/–
Kolesnick R & Fuks Z. Oncogene 2003;22:5897–906
MCJ
24
Aug 12

25. Same tumor, different mice
Garcia-Barros M et al. Science 2003;300:1155–59
MCJ
25
Aug 12

26. High doses not low doses per fraction
Fuks Z & Kolesnick R. Cancer Cell 2005;8:89–91
MCJ
26
Aug 12

27. Immunological effects at high doses
A549 Human NSCLC in lungs of nude mice
27 days after 12 Gy single dose
Hematoxylin-Eosin
Masson-Trichrome
IF
H
F
H
x20
Tumor
Small tumor nodules (arrows) with
degenerative changes in nuclei
and cytoplasm. Multiple large
vacuoles, hemorrhages [H] and
scattered inflammatory infiltrates
x40
Normal lung
Heavy infiltration of inflammatory
cells [IF] mostly lymphocytes and
neutrophils. Fibrous tissue [F] in
midst of inflammatory infiltrates
x40
C
Normal lung
Extensive fibrotic tissue [C] and
hemorrhages [H]
Hillman GG et al. Radiother Oncol 2011;101:329–36
MCJ
Aug 12
27

28. Response heterogeneity
Mixed target cell populations with different
sensitivites

29. Radiotherapy and Oncology, 9 (1987) 241- 248
Elsevier
241
RTO 00341
An explanatory hypothesis for early- and late-effect parameter
values in the LQ model
T. E. Schultheiss, G. K. Zagars and L. J. Peters
Division of Radiotherapy, The University of Texas, M. D. Anderson Hospital and Tumor Institute, Houston, TX 77030, U.S.A.
(Received 3 November 1986, revision received 17 February 1987, accepted 17 February 1987)
1. Higher-order terms (e.g. LQC) result from
response heterogeneity
Key words: Cell survival; lsoeffect: LQ model; Late effect
Summary
The isoeffect equation increase in “measured” value linear and
2. Leads dose. derived from the linear-quadratic that higher-order cell survival containsof !/"
to Experimental studies have shown (LQ) model of terms may also be present. These
quadratic terms in
terms have been previously attributed to the fact that the LQ model may be the first two terms in a power
series approximation to a more complex model. This study shows that higher-order terms are introduced as
a result of heterogeneity in the response of the cell population being irradiated. This heterogeneity is modeled
by assuming that the parameters ! and " in the LQ model are distributed according to a bivariate normal
distribution. Using this distribution, the expected value of cell survival contains third- and fourth-order terms
in dose. These terms result in the previously observed downward curvature of Fe plots. Furthermore, these
higher-order terms introduce bias in the estimated values of ! and ", if only the linear and quadratic terms of
the LQ model are used, and higher-order terms are ignored. The bias is such that the estimated value of
! /" is substantially increased. Thus the higher values of ! /" observed for early effects as compared to late
effects may be due to greater heterogeneity of response in early-responding tissues than in later-responding
tissues. This differential effect is maintained even if the two cell populations have the same average values of
! and ".
3. Leads to “linearization” of the
“cell survival curve” at higher doses
4. Explains the higher !/" for early effects
and in some tumors

30. Response heterogeneity
30%
70%
–20
–15
MCJ
Aug 12
30

31. Response heterogeneity
30%
70%
MCJ
Aug 12
31

32. Put more of this heterogeneity modeling in?

33. A model is a lie
that helps you see the truth
- Howard Skipper
Thanks Patrick McDermott

34. D0 = 1.42 Gy

35. In vivo survival curves linear at high dose
Tissue
D0 (Gy)
Ref
1.65
Kember 1965, 1967
1.3-1.4
Withers 1967
Colon
1.42
Withers 1974
Spermatogenic tubules
1.74
Withers 1974
Telogen hair follicles
1.35
Griem 1979
Jejunum
1.3
Hendry 1983
Kidney tubules
1.25
Withers 1986
Cartilage growth plate
Skin
from HD Thames
MCJ
Aug 12
35

36. Why does this make a difference?
High-dose log-linear (HDLL) model predicts higher
isoeffect dose than LQ as the curve goes linear
HDLL
LQ
At what dose
does this
divergence
become
significant?
MCJ
Aug 12
36

37. Approach
• Consider three tumor sites (breast, prostate, lung)
where hypofractionation or SBRT is being used
• Identify relevant !/" values of tumor and NTs, and
doses per fraction used clinically
• Choose HDLL models consistent with parameters
• Estimate size of dose per fraction at which 10% or
greater disparity between predicted isoeffect
doses occurs
• Compare this with doses actually in use clinically
MCJ
Aug 12
37

38. Fractionation in breast cancer
Mean = 4.0 [CL 1.0 – 7.8]

39. Int. J. Radiation Oncology Biol. Phys., Vol. 79, No. 1, pp. 1–9, 2011
Copyright Ó 2011 Elsevier Inc.
Printed in the USA. All rights reserved
0360-3016/$ - see front matter
doi:10.1016/j.ijrobp.2010.08.035
Fraction size 2.66 – 3.3 Gy, 6 Gy in FAST trial
CRITICAL REVIEW
HYPOFRACTIONATED WHOLE-BREAST RADIOTHERAPY FOR WOMEN WITH
EARLY BREAST CANCER: MYTHS AND REALITIES
JOHN YARNOLD, F.R.C.R.,* SØREN M. BENTZEN, D.SC.,y CHARLOTTE COLES, PH.D.,z
{
AND JOANNE HAVILAND, M.SC.
*Section of Radiotherapy, Institute of Cancer Research and Royal Marsden Hospital, Sutton, United Kingdom; yDepartment of Human
Oncology, University of Wisconsin School of Medicine and Public Health, Madison, Wisconsin; zOncology Centre, Cambridge
University Hospitals NHS Foundation Trust, Cambridge, United Kingdom; {Institute of Cancer Research Clinical Trials and Statistics
Unit, Section of Clinical Trials, Sutton, United Kingdom
INTRODUCTION
Hypofractionation for treatment of women with early breast
cancer is being used again after addressing past causes of
failure. Data from randomized trials confirm the safety and
efficacy of schedules using fraction sizes of around 3 Gy,
provided the correct downward adjustments to total dose
are made. Unjustified concerns relating to heart tolerance,
nonuniform dose distribution, and duration of follow-up
need not discourage the routine adoption of a 15- or 16-fraction schedule. Potential benefits of the overall shorter treat-
total dose must be reduced in order to maintain the same
level of antitumor or normal tissue effect. True, the dose reduction is relatively modest for epidermis and some tumors,
as correctly estimated by the Ellis isoeffect formula proposed in the late 1960s (1). When a regimen is changed
from 25 Â 2.0-Gy fractions to a 15-fraction regimen delivered over the same overall treatment time, the Ellis formula
estimated a dose reduction from 50 Gy to 45 Gy in 15 fractions of 3.0 Gy to match acute skin reactions. Ellis felt that
the healing of skin epithelium reflected ‘‘the condition of

40. Fractionation in prostate cancer
Int J Radiat Oncol Biol Phys
2011;79:195–201
Mean = 1.55 [CL 0.46 – 4.52]

41. Fractionation in prostate cancer
1.55 (0.46–4.52) Gy 5093 patients Proust-Lima C
PSA evolution median follow up 4.7 years d/f < 2.8
6 institutional datasets, no risk-group dependence
Int J Radiat Oncol Biol Phys 2011;79:195-201
1.48 Gy
1.4 (0.9–2.2) Gy 5969 patients Miralbell R
Biochem relapse free survival at 5 years d/f < 6.7
7 institutional datasets, no risk-group dependence
Int J Radiat Oncol Biol Phys 2012;82:e17-e24
1.86 (0.7–5.1) Gy 274 patients Leborgne F
Biochem disease free survival at 5 years d/f < 3.15
Single institution, no risk-group dependence
Int J Radiat Oncol Biol Phys 2012;82:1200-7

42. Relevant parameters for prostate trials
!/" values
Tumor: 1.48 Gy
Proust-Lima, Miralbell, Leborgne
GU and GI toxicity: 5 Gy
Joiner & Bentzen 2002, Brenner 2004
Doses per fraction
3.1 Gy Arcangeli et al 2010
3 Gy
Dearnaley et al 2007 Prostate Cancer
Symposium Orlando FL abstract #303
2.7 Gy Pollack et al 2009
MCJ
Aug 12
42

43. Treatment of NSCLC with SBRT
Early-stage or palliative
Reference
(1st author)
Dose per fraction (Gy)
x Number fractions
Minimum total
dose to PTV (Gy)
Timmerman (2006)
20-22 x 3 (80% isodose)
60-66
Timmerman (2003)
18-20 (80% isodose)
54-60
Nyman (2006)
15 x 3 (PTV periph)
45
Xia (2006)
5 x 10 (50% isodose)
50
Zimmermann (2005)
12.5 x 3 (60% isodose)
37.5
Wulf (2005)
12.5 x 3 (65% isodose)
37.5
Hara (2006)
30-34 x 1 (min PTV)
30-34
Fritz (2006)
30 x 1 (isocenter)
24
Hof (2003)
19-26 x 1 (isocenter)
15.2-20.8
Nagata (2005)
12 x 4 (isocenter)
48 to isocenter
MCJ
Aug 12
43

44. Results
Breast: For dose/fraction 2.7-3.3 Gy, LQ predictions of
isoeffect doses are approximately correct for tumor response
and toxicity, but too low for higher doses per fraction (>6 Gy)
Prostate: For dose/fraction of 2.7-3.1 Gy, LQ predictions of
isoeffect doses are approximately correct for tumor response
and toxicity, but too low for higher doses per fraction (>4 Gy).
This undermines to some extent the LQ-predicted therapeutic
gain from hypofractionation
Lung: For dose/fraction < 10 Gy, LQ predictions of isoeffect
doses are approximately correct for tumor response and early
toxicity, but too low for higher doses per fraction (>12 Gy)
Courtesy:
Howard Thames
MCJ
Aug 12
44

45. Hypofractionation can work in 2012 ?
• Physics enables:
- improved dose definition
- image guidance
• Biology is catching up:
- low “!/"” of some tumors…
- vascular and immunological effects
- is LQ good for high dose fractions?
MCJ
Aug 12
45