Advances in Skeletal Dosimetry Through Microimaging

1
Advances in Skeletal Dosimetry Through
Microimaging
Wesley Bolch, PhD, PE, CHP Director, Advanced
Laboratory for Radiation Dosimetry
Studies Department of Nuclear Radiological
Engineering University of Florida, Gainesville,
FL Oak Ridge National Laboratory Tuesday,
August 2 NCI Grant CA96441 and DOE Grant
DE-FG07-02ID14327
2
Strategies for Cancer Therapy

• External Beam Therapy (photons, protons, heavy
  ions)
• Insertion of Radioactive Seeds (brachytherapy)
• Radionuclide Therapy
• Unsealed Sources
• Tagged to bio-molecules (antibodies, peptides,
  etc.)

3
Radionuclide TherapyUnsealed Sources

• Benign Disease
• 131I sodium iodine Graves disease, goiter
• 32P sodium phosphate Polycythemia,
  thrombocythemia
• 90Y silicate colloid Severe arthritis
• 165Dy ferric hydroxide Severe arthritis

4
Radionuclide TherapyUnsealed Sources

• Malignant Disease
• 131I sodium iodine Thyroid cancer, residual
  disease
• 131I MIBG Metastatic neuroblastoma
• 111In octreotide Neuroblastoma
• 32P chronic phosphate Intracavity therapy
• 89Sr strontium chloride Painful skeletal
  metastases
• 153Sm EDTP Painful skeletal metastases
• 186Re HEDP Painful skeletal metastases

5
Radioimmunotherapy (RIT)

• Solid Tumors
• 131I anti-EGFr Recurrent gliomas
• 125I-425 , 131I-BC-2 Glioblastoma multiforme
• 131I-HMFG1, 186Re-NRLU19 Ovarian cancer
• 177Lu-CC49 Breast, colon, lung cancer
• 131I-CC49 Prostate cancer
• 90Y- or 131I-anti-ferritin Hepatoma
• 186Re HEDP Gastrointestinal cancer
• 90Y-ChT84.66 anti-CEA Colon cancer

6
Radioimmunotherapy (RIT) of B-Cell Lymphoma

• Non-myeloablative
• 131I Lym-1, LL2, Anti-B1, MB1
• 90Y B1, 2B8, C2B8
• Myeloablative
• 131I B1, MB1, LL2, 1F5, BC8
• 90Y B1
• 213Bi HuM-195

7
Fundamental Questions in RIT

• What maximal radionuclide administration can I
  deliver to the patient?
• Need to avoid normal organ complications
• Bone marrow, lungs, GI tract wall, kidneys
• How can I predict this maximum-tolerated activity
  in a given patient?
• Dose-response function for marrow toxicity
• Perform patient-specific estimates of marrow dose

8
MIRD Method for Calculating Internal Dose
Integrated Activity Integral no. of decays in
source region rS
S Value Dose to target region rT per decay in
source rS
9
Radionuclide S Values
Absorbed Fraction Fraction of particle energy
emitted in rS that is deposited in rT
10
Source and Target Regions

• Potential Sources ( rS )
• Active bone marrow
• non-specific uptake (blood/fluid spaces) or
  specific antibody binding (cells)
• Osseous tissues of bone
• Uniformly distributed within the bone volumes
• Uniformly distributed on the interior bone
  surfaces
• Potential Targets ( rT )
• Active bone marrow
• Stem cells and their precursors
• Endosteum tissue layer on the bone surfaces
• Single-cell layer containing osteoblasts (bone
  building cells)
• and osteoclasts (bone destroying cells)

11
Bone Structure

• Cancellous Bone
• spongy bone
• trabecular bone

• Cortical Bone
• hard bone
• compact bone

Spongiosa trabecular bone marrow tissues
endosteum
12
Tissues of the Bone Marrow

• Hematopoietic cellular component
• granulocytic, erythroid, and megakaryocytic
  series
• Bone marrow stromal cells and extracellular
  matrix
• adipocytes, reticulum cells, endothelial cells
• Venous sinuses and other blood vessels
• Various support cells
• lymphocytes, plasma cells, mast cells,
  macrophages

13
Formation of Blood Cells
14
Marrow Cellularity
Marrow Cellularity Fraction of total marrow
space occupied by hematopoietic cells
(cellularity factor, CF) 1 - (Fat
Fraction)
bone trabecula
active or red marrow
inactive or yellow marrow (adipocytes)
endosteum
15
Marrow Cellularity with Age and Skeletal Site
16
MIRD Method for Calculating Internal Dose
Integrated Activity Integral no. of decays in
source region rS
S Value Dose to target region rT per decay in
source rS
17
Patient-specific estimates of
(1) Direct NM imaging
(2) Inference from Blood Measurements
RMECFF red marrow extracellular fluid
fraction HCT patients hematocrit
18
Patient-specific estimates of S ?

• Out of necessity, the medical community has
  borrowed the ICRP reference skeletal models
  developed originally for radiological protection
• The ICRP model has two components
• Reference skeletal masses from the work of
  Mechanic (1926)
• Reference absorbed fractions from the work of
  Spiers (early 1970s)
• Important Point AF data come from an entirely
  different anatomic source than those used to
  define reference tissue masses

19
Study by Mechanik (1926) as summarized by
Woodward (1960)

• 6 male cadavers and 7 female cadavers (18 to 86
  y)
• Senile marasmus (4 cases), tuberculosis (3
  cases), heart disease (2 cases), and malaria (1
  case)
• The bodies appear to have been somewhat but not
  excessively, emaciated, the weights of the women
  ranging from 43.5 to 55.2 kg, and those of the
  men from 59.6 to 65.0 kg.
• It is most unfortunate that the bodies of
  previously healthy victims of accidents or other
  causes of sudden death were not chosen for
  study

20
FW Spiers at the University of Leeds

• Original anatomic source for current absorbed
  fractions
• Single 44-year-old male (skeletal reference man)
• Contact radiographs taken
• Parietal bone, cervical vertebra, lumbar
  vertebra, rib, iliac crest, femur head, and femur
  neck
• Optical scanning system developed
• Chord length distributions were obtained
• Marrow cavities
• Bone trabeculae

21
Spiers Optical Scanning System
22
Current Models of Skeletal Dosimetry

• 1D models of particle transport
• Only 7 skeletal sites
• Single 44y male
• Masses of target tissues taken for other studies

23
Leeds Marrow Chord Distributions
24
Leeds Bone Trabeculae Distributions
25
Applications for a and b particles

• CBIST modeling approach
• Chord-Based Infinite Spongiosa Transport
• Both a and b particles are following through an
  infinite expanse of spongiosa (interior tissues
  of trabecular bone) until their full emission
  energy is expended
• No accounting for electron escape to cortical bone

26
Marrow Spatial Model
70
X 4
50 30
27
Alpha-Particles Active MarrowComparisons to
ICRP 30 and OLINDA
28
Applications for photons

• Absorbed dose to active bone marrow
• Fluence-to-dose response function (DRF)
• Based upon CBIST electron results
• Mass energy absorption coefficient (MEAC) ratio
• CT number method
• Provides for a unique composition per skeletal
  voxel
• Absorbed dose to endosteum
• Fluence-to-dose response function (DRF)
• Based upon CBIST electron results
• Homogeneous bone dose approximation

29
Moving Beyond CBIST in Skeletal Dosimetry
30
3D Image-Based Skeletal Dosimetry UF Adult Male
Model

• Cadaver selection (66 yr, 68 kg, 173 cm, 22.7 kg
  m-2)
• Whole-body CT imaging ( 1 mm3 voxels)
• Bone site harvesting (13 major sites of adult
  active bone marrow)
• Ex-vivo CT imaging of each excised skeletal site
• Image processing ? volumes of spongiosa (1 mm x
  0.3 mm2)
• Spongiosa ? combined tissues of trabeculae,
  endosteum, active and inactive marrow
• Section skeletal sites cubes of spongiosa
• Microimaging of spongiosa
• NMR microscopy or mCT (30 80 mm3)
• Radiation transport simulation of electron/betas

31
PIRT Model of the Lumbar Vertebra
32
PIRT Model of the Lumbar Vertebra (70
cellularity)
33
PIRT Model of the Proximal Femur
34
PIRT Model of the Pelvis
35
PIRT Model of the Cranium
36
PIRT Model of the Ribs
37
PIRT Model of the Ribs (70 cellularity)
38
ICRP and UF Active Marrow Masses by Skeletal Site
39
Skeletal-Averaged Absorbed FractionsUF Model and
the 2000 Eckerman Model (MIRDOSE3)
40
Skeletal-Averaged Absorbed FractionsUF and 2003
Eckerman Model (OLINDA)
41
How can one adjust this image-based skeletal
model to the individual patient?

• Factors to consider
• Physical stature (size of the skeleton)
• Decreases in physical stature will result in
  higher electron escape to cortical bone
• Bone mineral status
• Decreases in BMD are primarly associated with
  thinning and eventual loss of bone trabeculae.
  Loss of bone mass is usually accompanied by
  increases marrow fat (MVF x CF perhaps remains
  constant)
• Marrow cellularity
• Changes in marrow cellularity can be accounted
  for explicitly in the PIRT or other image-based
  models

42
Clincial Input Data for patient-specific model
adjustment

• Pelvic SPECT-CT of RIT Patient
• SPECT image
• quantify marrow / skeletal uptake in sacrum or
  lumbar vertebrae
• CT image
• Make skeletal size measurement (e.g., pelvic
  height)
• CT image
• Using a calibration curve from a previously
  imaged BMD phantom, assess the patients
  volumetric BMD (femoral neck, lumbar vertebrae)
• MR imaging or BM biopsy
• Assess marrow cellularity of the patient,
• Assume reference values or some proportional
  change thereof

43
PIRT ModelAdjustments for skeletal stature
From previous cadaver skeletal studies
where AP anthropometric parameter such as total
body height or head circumference
IBP image-based parameter such as pelvic
height
PIRT model run at size specific to the patient
44
PIRT ModelAdjustments for skeletal stature
VBIST Results
45
PIRT ModelAdjustments for bone mineral status
From previous cadaver skeletal studies
where BMD volumetric CT-based bone mineral
density measured at the femoral
head/neck and lumbar vertebrae
PIRT model run using microCT images from a
reference library
Normal BMDv mCT Images
Osteopenic BMDv mCT Images
Osteoporotic BMDv mCT Images
46
PIRT ModelAdjustments for bone mineral status
47
PIRT ModelAdjustments for marrow cellularity
48
Scalability of Image-Based ModelsImproved
Patient Specificity
49
PIRT ModelAdjustments for location of cellular
targets
CD34 staining of BM biopsies