A Dosimetric Comparison of Copper and Cerrobend Electron Inserts

Ben D. Rusk1, Kenneth R. Hogstrom1,2, John P. Gibbons1,2, Robert L. Carver1,2

(1) Louisiana State University, Baton Rouge, LA, (2) Mary Bird Perkins Cancer Center, Baton Rouge, LA

A Dosimetric Comparison of Copper and Cerrobend Electron Inserts

Ben D. Rusk1, Kenneth R. Hogstrom1,2, John P. Gibbons1,2, Robert L. Carver1,2

(1) Louisiana State University, Baton Rouge, LA, (2) Mary Bird Perkins Cancer Center, Baton Rouge, LA

RESULTS AND DISCUSSION

PDDs showed no clinically significant differences

(<1%/1mm) at depths smaller than the practical range (Rp).Copper inserts showed a lower photon contamination

along the central-axis (Dx) than Cerrobend, with maximumdifferences of 0.5%. This difference results from the known

relationship between relative bremsstrahlung photon

production and the ratio of atomic number squared toatomic mass2. Since copper has an effective atomic

number less than half that of Cerrobend, a decrease inphoton production caused a slightly lower Dx value. This

effect increases with energy, and is largest with small fieldinserts in large applicators and at 100 cm SSD (Fig. 2).

Outputs were measured at 100-cm and 110-cm SSD for

both materials and the ratio of Cu/Cerrobend outputs forthe same geometry was used to scale the corresponding

2D dose distributions for absolute comparison.

Compare Absolute 2D Dose Distributions:Dose distributions between matching copper and

Cerrobend inserts were compared using a ±2% of central-

axis maximum dose or ±1 mm distance to agreement (DTA)criteria. The percentage of area passing the criteria was

recorded and any comparisons containing failing pointswere re-analyzed using a 3%/1-mm DTA criteria.

For absolute 2D dose distribution comparisons at

100-cm SSD, 46 of the 48 combinations (energy-field size-applicator) passed the 2%/1-mm criteria for >99% of area.

The two comparisons with <99% points passing were smallfield sizes (≤4x4 cm2) in the 25x25-cm2 applicator at 20 MeV.

These inserts showed no in-field differences >2%, howeverthe decrease in photon dose outside of the field caused

criteria failures near the surface (Fig. 4). All combinations

passed a 3%/1-mm criteria for 100% of area. All 48combinations at 110-cm SSD passed the 2%/1-mm criteria

for 100% of area.

METHODS and MATERIALS

Obtain a Matching Set of Copper and Cerrobend Inserts:

A total of 32 matching pairs of inserts were obtainedfor:

INTRODUCTION

Using third party vendors, such as .decimal, Inc.

(Sandford, FL), to mill custom electron inserts couldeliminate the need for in-house block cutting rooms and

potentially provide improved accuracy through precisionmilling as compared to hand poured Cerrobend inserts.

Specifically, custom electron inserts milled out of copper

may provide a suitable and clinically equivalent alternativeto traditional Cerrobend inserts.

Many linear accelerators are commissioned usingCerrobend inserts, with the data being used for dose

calculations and treatment planning. Implementing copperelectron inserts clinically, without recommisioning,

requires a comparison of standard commissioningmeasurements between copper and Cerrobend inserts.

PURPOSE

To compare percent depth dose curves (PDDs), off-axis

relative dose profiles, output factors, and absolute 2D dosedistributions for a clinically relevant range of insert field

sizes, applicator sizes, energies, and source-to-surfacedistances (SSDs) for matching copper and Cerrobend

inserts 1.

Figure 3: Off-axis relative dose profile at a depth of 0.5 cm for a 12x12-cm2 field

size in a 25x25-cm2 applicator at 20 MeV and 100-cm SSD; red circles highlight

the two areas of dosimetric differences: the out-of-field dose and the beam edge

horns near the edge of the insert.

Figure 2: PDDs measured at various energies using copper and

Cerrobend inserts in the 25x25-cm2 applicator at 100-cm SSD for2x2-cm2 (a) and 10x10-cm2 (b) field sizes.

Off-axis relative dose profiles showed only small in-

field differences (<2%). Increased dose at the field edgefrom the copper inserts was due to the greater amount of

scatter off the insert edge compared to Cerrobend. Adecrease in out-of-field dose under the copper inserts was

due to the reduction in bremsstrahlung photon productionin copper compared to Cerrobend, and was >2% in some

instances. This difference in out-of-field photon dose was

greatest at the highest energies, and for small fields inlarge applicators at 100-cm SSD. These differences are

illustrated in Figure 3.

ACKNOWLEDGEMENTS

This research was supported in part by .decimal, Inc.

(Sanford, FL), who provided gifts in-kind, specificallycopper inserts and aluminum templates for the Cerrobend

inserts.

REFERENCES1Rusk B. D. ,“A Dosimetric Comparison of Copper and Cerrobend Electron

Inserts”, MS Thesis Louisiana State University and Agricultural and

Mechanical College, Baton Rouge, LA (2014).2ICRU. 1972 Radiation dosimetry: electrons with initial energies between 1 and

50 MeV Report No. 21. (Washington, DC).3Khan F M, Doppke K P, Hogstrom K R, et al., “1991 Clinical electron-beam

dosimetry: report of AAPM Radiation Therapy Committee Task Group No.

25” Med Phys 18 73-109 (1991).

CONCLUSIONS

Using custom milled copper inserts for electron beam

therapy planned with standard commissioning datameasured using Cerrobend inserts results in minimal in-

field dosimetric differences (<2%) for standard clinicalapplicators (6x6-25x25 cm2), energies (6-20 MeV), and field

sizes (2x2-20x20 cm2).

The largest dosimetric differences (>2%) result fromthe lower out-of-field dose for copper inserts due to a

reduction in bremsstrahlung production as compared toCerrobend. Clinically, this dosimetric difference is

insignificant, as the use of copper inserts could reduce thedose received by healthy tissue outside of the planned

treatment volume.

Figure 4: (a) Absolute isodose comparison between Cerrobend (solid lines)

and copper (dashed lines) for an 8x8 cm2 insert in a 15x15-cm2 applicator at

12 MeV and 100-cm SSD. (b) Absolute isodose comparison for a 2x2-cm2

insert in a 25x25-cm2 applicator at 20 MeV and 100-cm SSD, with 98.90% of

area passing the 2%/1-mm criteria and red pixels indicating the pixels which

failed. (c) PDD 5 cm off-axis as indicated by the green line.

for:

� Nine different field sizes (2x2 cm2-20x20 cm2)� Five available applicator sizes (6x6 cm2-25x25 cm2)

Measure Dosimetric Data:Measurements were taken on a Varian Clinac 21EX

linear accelerator using an IBA EFD3G Electron DosimetryDiode Detector3. PDDs and outputs measurement

geometries for electron energies of 6, 9 ,12, 16 and 20 MeVfor 100-cm and 110-cm SSD are listed in Table 1. Also listed

are the off-axis relative dose profiles geometries forenergies of 6, 12, and 20 MeV at 100-cm and 110-cm SSD.

Figure 1: (LEFT) Cerrobend poured into a 15x15-cm2 applicator mold to

generate an insert formed by a 4x4-cm2 aluminum negative. (RIGHT)The resulting Cerrobend insert alongside its matching copper insert,

fabricated by .decimal, Inc.

Table 1: Full set of 32 inserts used for measuring PDDs and outputs for all

five energies at 100-cm SSD (X’s); a smaller subset of 16 inserts used for

measuring PDDs and outputs at 110-cm SSD and for off-axis relative dose

profiles at both SSDs for three energies (O’s).