Comparison between 2D, 2.5D and 3D PTVs definition in treatment planning for conformal radiotherapy of organ-confined carcinoma of prostate

 

Yasser Assad Rashed, Ph. D. and Tounis El-sayed Ahmed M. D.

The Department of Clinical Oncology and Nuclear Medicine, Faculty of Medicine, Menuofia University.

 

Abstract:

 

Purpose: To evaluate the adequacy of planning target volume (PTV) coverage using a three-dimensional (3D) margin-growing method compared to a two-dimensional (2D) and two and-one-half-dimensional (2.5D) margin-growing methods in the conformal radiotherapy (CRT) planning of prostate cancer.

 

Patients and methods: PTV of 10 patients suffering from organ-confined prostate carcinoma was planned using ARTS 3D treatment planning system. The plan was performed by; 2D, 2.5D, and 3D techniques. The three PTVs were determined by expansion of the gross tumor volume (GTV) by a 1.0-cm margin. The former was obtained by automatically adding the margin in 2.5D and 3D for GTV. In 2D the former was obtained by manual contouring in CT slices. For each patient the three PTVs were compared to assess the deviations of the 2D and 2.5D PTVs from the 3D PTV. For all PTVs conformal plans were designed using the same number of fields for the three PTVs. For each patient dose-volume histograms (DVHs) and isodose distributions were calculated.

 

Results: The 2D and 2.5D margin-growing methods underestimated the PTV by 17.84 % (range 12.1-22) and by 19.62 % (range 13.7-23.2) respectively when compared to the 3D margin-growing method. The underdosage according to 2D, 2.5D, and 3D margins are 2.6 % (range 1-3.23), 2.11 % (range 0-4.30), and 1.29 % (range 1-2.11) respectively.  

 

Conclusion: From the comparison between the DVHs and isodose distributions for different slices, the full 3D GTV-to-PTV expansion is recommended in CRT to avoid underdosage.

 

Keywords: PTV; Conformal radiotherapy; 2D; 2.5D; 3D; prostate carcinoma.

 

 

1. Introduction

Three-dimensional conformal radiotherapy (3D CRT) allows men with prostate cancer to be treated safely with high doses of radiation. Determination of the planning target volume (PTV) is an essential step in 3D CRT [1]. The precision of its delineation directly influences the quality of treatment [2-8]. Delineation of the prostate and seminal vesicles for CRT involves computed tomography (CT). Rectal and bladder contrast are helpful in defining these normal structures [9]. An urethrogram has been used to identify the inferior border of the prostate, which is often difficult on a transverse cut. Image-based beam’s eye view techniques have resulted in changes in 40% of the portal apertures, in comparison to conventional simulator methods of prostate target delineation [10].  The Gross tumor volume (GTV) is taken to be the prostate, seminal vesicles, and puboprostatic ligament. 

The International Commission of Radiation Units and Measurements (ICRU) [11] has defined volumes for treatment planning that take into account the extent of the known gross tumor, the areas of likely microscopic extension, and daily variations in patient setup and tumor position. In prostate cancer, most authors believe that in general 0.5- to 1.0-cm is required to ensure coverage of extracapsular microscopic tumor extension [12-14]. Roach, et al. [12] suggested increasing the margin to 1.0-cm infero- laterally and 1.5-cm postero- laterally at the area of the neurovascular bundles where tumor extension through the capsule is prevalent. To generate the PTV, additional margin is added to account for patient setup variation and prostate movement secondary to rectal and bladder filling. The PTV should be designed so that it covers the clinical target volume (CTV) with high probability for every treatment [15].

There are many methods by which a predetermined margin can be added to the GTV [16-18]. A commonly used method has been to add a uniform margin to each transaxial GTV outline. However, this two-dimensional (2D) method can be inadequate in certain regions of a 3D target volume, as the margin is not added perpendicular to the GTV, therefore, it does not take into account the volumetric nature of the GTV [19, 20]. While most major centers practicing 3D conformal planning utilize 3D volume expansion methods. Many other centers utilize beams-eye-view (BEV) planning in CRT may still be use a laminar 2D margin method. Conformal 3D planning requires 3D spatial evaluation of the tumor borders in order to create the appropriate PTV [21].

This study compares the adequacy of PTV coverage resulting from the application of a uniform margin in 2.5D and 3D with the coverage provided by a conventional 2D margin in the conformal treatment of localized prostate cancer.

 

2. Material and Methods

2.1. Patients, tumors, and radiotherapy details

From October 2002 to March 2004, ten prostate cancer patients treated with CRT using a six-field technique with shaped beams in Menuofia University, Egypt were selected for this study. Six of the patients had stage T1 tumors and four had T2 tumors for which the seminal vesicles were part of the GTV. The median age was 62 years (range 51-74). The patients were simulated in supine position and submitted in the same position to a CT scan of the pelvis with a 0.2-cm step (thickness 0.2-cm) in the region of the prostate and the seminal vesicles and with a 0.5-cm step caudally down to the anus, and cranially up to the sigmoid colon (always including the top of the bladder).

 

2.2. The treatment planning system

The advanced radiation therapy treatment planning system (ARTS-3D TPS, version 2.2, March 2000) was used in this work. It is designed to run on high performance Unix graphics workstations. Advanced Visual System (AVS) enable the system to provide the most versatile 3D visualization and treatment planning features.

          The 3D model for photon beam dose calculation allows for irregular beam shapes and the 3D geometry of the patient. Beam direction can be non-coplanar as defined through gantry, table and collimator rotations. Dose normalization and beam weighting are handled efficiently using a summation process to provide rapid evaluation of 2D and 3D dose distributions.

 

2.3. Analytical strategy

CT images of the clinical and added margin were transferred to the planning workstation. The GTV was outlined in all relevant slices. The GTV covers the prostate gland only (PO) and a GTV covers the prostate with the seminal vesicles (PSV). The position of the apex of the prostate was verified by use of sagittal reconstruction through the prostate. A margin of 1.0-cm was applied for each patient in these two groups PO and PSV to create the PTV. The margin for each GTV was created automatically using both 2.5D and 3D margin-growing algorithms and manually using 2D margin growing method.

 

2.4. PTV definitions

The 2D, 2.5D and 3D PTVs are available in ARTS-3D TPS. The 2.5D technique amounts rolling an ellipse with its major and minor axes defined by the specified margins along the border of the contour template to expand its boundary. This technique will result in ‘exact margins’ in the x–y plane, on a contour-by-contour basis. Margins as viewed on a volume basis will be uneven depending on the sloping of the volume surface.

The 3D technique uses an ellipsoid to roam along the surface of the volume template. This technique guarantees ‘exact margins’ on a volume basis, the margins on the x–y plane, by the technique, will be uneven. The 2D margin-growing method deals with each transaxial slice in succession. The difference between 2D and 2.5D is that the 2D method does not provide a margin superior and inferior to the GTV. To grow the 2D margin on GTV accurately, growing of the 2.5D is done and then the excess slice on the superior and inferior to the GTV is removed. In order to create an appropriate superior margin, the most superior GTV outline is copied onto the end of 2.5D margin. A similar process is used for the inferior region (Figure 1)

2.5. Treatment technique

The importance of the choice of the irradiation technique in CRT of prostate cancer is related to the different doses, which can be delivered to the surrounding critical organs (rectum, bladder and femoral heads). There are many different techniques used in the CRT of the prostate cancer [22-29]. There is no technique better the other when considering all the organs at risk [29]. In this study, all patients were treated by six equally weighted fields (two lateral fields and four oblique fields at gantry angles 30º, 150º, 210º, and 330º) (Figure 2). The six-field arrangement conformed for three volumes; 2D, 2.5D, and 3D PTVs respectively. Figure 3 shows BEV of the conformed field at gantry angle 90º for the three PTVs. In figure 3a and 3b, the shape of the conformed field is almost the same except at the superior and inferior of PTV where the field was conformed to 2D and 2.5D PTVs respectively. But in Figure 3c the shape of the conformed field is wider than in figure 3a and 3b where the field was conformed to the 3D PTV.   

 

2.6. Statistical evaluation

The significance of measured differences between different PTV-volumes of patients was tested with Student’s t-test for a confidence level of 0.05.

 

3. Results

Figure 4 (a, b, and c) shows the axial position and sagittal and coronal reconstructions of the prostate in a plane near the superior part of the prostate for the same patient and the GTV (red line), 2D and 2.5D PTVs (green line) and 3D PTV (yellow line) are depicted. As expected, the 2D and 2.5D PTVs generally follows the GTV at exactly 1.0-cm in lateral and ventro-dorsal directions. Certain parts of the GTV responsible for a GTV-to-PTV extension as seen in the reconstruction may in themselves be invisible in that reconstruction. The 2D extension identify with the 2.5D extension except the regions after the end of the GTV in the cranio-caudal direction. These regions due to the slight variation in the volume of PTV between 2D and 2.5D extensions.

3D PTV

+2.5D=2D PTV

GTV

GTV

2.5D PTV

  

Figure 1: The classification of the treatment volumes; GTV and three PTVs definitions. The GTV is the first degree of dark black, 2.5D is the second degree, and with adding the four corners to 2.5D give us the 2D. Finally the third degree of dark black (light black) is 3D PTV.

 

 

Figure 2: 6-field technique of two lateral fields and four oblique fields at gantry angles 30º, 150º, 210º, 330º.

A                                    B                                    C

 

Figure 3: BEV of the left lateral beam (gantry angle 90º) conformed to the three PTVs; (A) 2D PTV, (B) 2.5D PTV, and (C) 3D PTV. The conformal irradiation field (irregular white line) in A and B almost the same except at the superior and inferior of PTV. In C the conformal field is wider than in A and B.

 

Table 1 summarized the clinical patient prostate volumes generated by 2D, 2.5D, and 3D margin techniques for the clinical GTV. The mean volume of the GTV was 112.7 cc. The mean PTV defined by the 2D margin technique was 240.1 cm3 (range 171.4-309 cm3) and the mean PTV with 2.5D margin was 234.7 cm3 (range 169-300 cm3) while the mean PTV with the 3D margin was 292.9 cm3 (range 202.3-372 cm3).  Overall, the PTVs generated with 3D algorithm were 12.1-22% larger than the PTVs generated by the 2D method and 13.7-23.2 larger than the PTVs generated by the 2.5D algorithm. The mean difference of the PTVs discrepancy between 2D and 3D is 17.8% and between 2.5D and 3D is 19.6%. The difference between the 2D and 2.5D PTVs is very highly significant (p<0.001) because slight variation in its volumes. While the difference between the 3D and both 2D and 2.5D PTVs is significant (p<0.05).

Figure 5 shows DVHs for the PTVs of prostate using the same six-field arrangement. In figure 5a the plan designed for the 2D PTV, in figure 5b the plan designed for the 2.5D PTV, and in figure 5c the plan designed for the 3D PTV. When the plan designed to the 2.5D PTV, the axial dimensions for the six field sizes are 10.4x8.9-, 10.4x9.7-, 10.4x9.9-, 10.4x9.6-, 10.4x8.9-, and 10.3x9.7-cm2 respectively started from the gantry angle 90º and rotated in clockwise direction.

2D & 2.5D

Text Box: GTV

 

Text Box: A

 

 

3D PTV

2D & 2.5D

GTV
Text Box: B

 

 

 

Text Box: 2D & 2.5D

 

3D PTV

GTV
Text Box: C

 

 

 

Figure 4: GTV (red line), 2D and 2.5D PTVs (green line), and 3D (yellow line) show on the three positions; (A) Axial; (B) Sagittal; and (C) Coronal.


 

Table 1: Tumor and planning volumes for the prostate patients.

Case

GTV

(cm3)

PTV (cm3)

3D-2D

3D-2.5D

p-Value

2D

2.5D

3D

diff. (%)

diff. (%)

2D/2.5D

2D/3D

2.5D/3D

1

87

185.5

182

211

12.1

13.7

0.00001

0.048

0.029

2

96.6

202.1

197

249

18.8

20.9

3

138

288.8

280

345

16.3

18.8

4

109

218.5

215

280

22

23.2

5

144

309

300

372

16.9

19.4

6

90

171.4

169

202.3

15.3

16.5

7

96

205.4

200.9

259

20.7

22.4

8

125.1

261

257

308.9

15.5

16.8

9

126

284

278

354

19.8

21.5

10

115

275

268

348

21

23

Mean

112.7

240.1

234.7

292.9

17.8

19.6

 

Applying these beams will therefore result in underdosage of the 3D PTV. This is further illustrated in Figure 6 by the DVHs for the 3D PTV resulting from the treatment plans designed for the 2D, 2.5D, and 3D PTVs, respectively.

          The analysis of these DVHs is summarized in Table 2. Ninety percent of 3D PTV volume covered by 93% dose for the 2D plans designed. For the 3D plans designed it is covered by 96%. From the table the dose delivered depends on the target volume.  Figure 7 shows the change in the isodose distributions with changing the designed plans for the three PTVs displayed on the slice near from the cranial end of the GTV. The dose distribution resulting from the beam shapes determined for the 3D PTV was assumed to be correct delivered to the 3D PTV.       

For the plan designed to the 2D PTV, the axial dimensions for the six field sizes changes to 10.4x8.9-, 10.4x9.7-, 10.4x9.9-, 10.4x9.6-, 10.5x8.9-, and 10.4x9.7-cm2 respectively. But when the plan designed to the 3D PTV, the axial dimensions for the six field sizes are 10.6x9-, 10.7x9.6-, 10.7x9.9-, 10.7x9.6-, 10.7x9-, and 10.7x9.8-cm2 respectively.

Text Box: C

 

Text Box: A

 

 

Figure 5: Cumulative DVHs for the 2D, 2.5D, and 3D PTVs.  The six-beam arrangement designed for (A) 2D PTV; (B) 2.5D; and (C) 3D PTV respectively.

 

Figure 6: Cumulative DVHs for the 3D PTV of the three designed plans. The solid curve represents the DVH of the plans designed for the 2D PTV, the dashed curve represents the DVH of the plans designed for the 2.5D PTV, and the symbol curve represents the DVH of the plans designed for the 3D PTV.  

 

Table 2: The isodose line value for different partial volume of the PTVs for different designed plans. 

 

Volume of PTV (%)

Isodose line (%)

2D Calculation

2.5D Calculation

3D Calculation

2D

2.5D

3D

2D

2.5D

3D

2D

2.5D

3D

60

100

100

99

99

99

99

100

100

99

80

99

98

96

98

98

96

99

99

98

90

96

96

93

95

96

93

97

97

96

95

93

95

90

93

93

89

95

96

93

 

Bladder

Bladder

Text Box: 50%

 

B
Text Box: B

 

Text Box: A

 

         

 

Bladder

Text Box: 50%

 

Text Box: 80%

 

Text Box: 90%

 

Text Box: GTV

 

Text Box: C

 

Text Box: Figure 7: The isodose line distributions for patient calculated by conformal radiotherapy displayed on the slice near from the end of GTV. The beams are conformed to appropriate; (A) 2D margin; (B) 2.5D margin; and (C) 3D margin respectively.  The red line is GTV; the green, light blue, and yellow lines are 90%, 80% and 50% isodose lines respectively. The coved dose of GTV dependent on the volume of PTV.

 

 

 

 

 

4. Discussions

The ICRU Report 50 recognizes that radiotherapy planning requires a planning margin to be specified in all directions [11]. This aspect being more important when considering 3D planning techniques. Defining the appropriate treatment margin can be a complex task as one has to consider, among other factors, the extent of predicated microscopic tumor spread or the CTV, the magnitude and frequency of patient and internal organ movement, as well as the allowance for treatment setup errors. Putting aside these issues, this study compares the validity of an arbitrarily specified margin determined by a 2D margin method with a margin as determined by the 2.5D and 3D margin-growing algorithms.

For prostate radiotherapy planning, when the 2D method is used, the region of the prostatic apex is most susceptible to radial deficiencies in margin width while the superior region demonstrates the largest deficiencies. This can be explained by the methodology of the 2D method which creates a simple chamfered margin at the cranial and caudal ends of the GTV compared to the appropriate volumetric margin produced by the 2.5D and 3D algorithms.

Discrepancies between the three margin methods can be more marked when considering the prostate as the GTV. In this region, the 1.0-cm transaxial margin generated by the 2D and 2.5D methods appears adequate in the transaxial plane as shown in Figure 4, while the margin generated by the 3D margin appears abnormally excessive; however, the sagittal view of this same region clearly demonstrates that the 2D and 2.5D margins are grossly inadequate, as the 1.0-cm margin has not been added perpendicular to the GTV. The main difference between the 3D margin algorithm compared to a conventional 2D method is that for any given CT slice, the 3D algorithm automatically takes into account the tumor outline on the slices above and below it, rather than merely the tumor outline within that given transaxial CT slice only.

When the actual target is larger than the delineated target volume, this results in underdosage in the target and can result in a reduction in Tumor Control Probability (TCP). On the other hand, when the actual target is smaller than the delineated target, this may have consequences for the dose to the surrounding tissues. The uncertainties in delineation should be taken into account during the planning process.

Margins defined only in the transaxial plane underestimate the extent of coverage that is required in other planes. This has the potential of allowing a portion of the target to be missed, thus reducing the probability of local control, and subsequently, cure. This has consequences for CRT and could reduce the efficacy of dose escalation studies. The 3D margin growing algorithm avoids such underestimation of the intended margin be ensuring that the margin is spatially uniform all around the tumor volume. This method of adding a spatially uniform margin may allow critical structures to be included in the PTV [30].   Conventional 2D margin methods are still widely used; therefore, it is important to be aware of the possible deficiencies described. Beam apertures for radiotherapy treatment should be designed by adding a 2D margin to the GTV outline in the BEV rather than by growing a 2D margin in the transaxial plane only, then creating the beam profiles are correct; it does not ultimately provide an accurate PTV against which the calculated dose distribution can be assessed. For this reason, the 3D margin-growing algorithm has now replaced the 2D method used for CRT planning. 

 

5. Conclusion

This study illustrates the problem of assuming that laminar added transaxially on a slice-to-slice basis are adequate to cover a 3D tumor volume. An appreciation of spatial margins in 3D is required if 2D margin drawing method and 2.5D margin-growing algorithm are used. BEV evaluations are also required to ensure adequate coverage of the tumor volume when using 2D margin-drawing method and 2.5D margin-growing algorithm. Utilization of an accurate 3D margin algorithm allows the predetermined margin to be realized in all spatial orientations. We have shown that multiple 2D calculation and 2.5D margin-growing algorithm of PTV margins instead of full 3D calculations can lead to serious underdosage and hence full 3D PTV margin calculations are required when CRT is used.

 

References

1-           Rezart B., Dominique P., Jean-Claude R., and Genevieve G. Automatic three-dimensional expansion of structures applied to determination of the clinical target volume in conformal radiotherapy. Int. J. Radiation Oncology Biol. Phys. Vol. 37; No. 3, 689-696, 1997.

2-           Dische S., Saunders M. I., Williams C., Hopkins A., and Aird E. Precision in reporting the dose given in a course of radiotherapy. Radiother. Oncol. 29; 287-293, 1993.

3-           Gotein M. Calculation of the uncertainty in the dose delivered during radiation therapy. Med. Phys. 12; 608-612, 1985.

4-            Herbert D. E. An extreme value paradigm for the effect of size of target volume on end results in radiation oncology. Med. Phys. 10; 589-604, 1983.

5-           Leunens G., Menten J., Weltens C., Verstraete J., and Van der Schueren E. Quality assessment of medical decision making in radiation oncology: Variability in target volume delineation for brain tumors. Radiother. Oncol. 29; 169-175, 1993.

6-           Mayles W. P. M., Chow M., Dyer J., Fernandez E. M., Heisig S., Knight R. T., Moore I., Nahum A. E., Shentall G. S., Tait D. M. The royal Marsden Hospital pelvic radiotherapy trial: Technical aspects and quality assurance. Radiother. Oncol. 29; 184-191, 1993.

7-           Ten Haken R. K., Thornton A. F., Sandler H. M., LaVigne M. L., Quint D. J., Fraas B. A., Kessler M. L., and McShan D. L. A quantitative assessment of the addition of MRI to CT-based,3-D treatment planning of the brain tumors. Radiother. Oncol. 25; 121-133, 1992.

8-           Urie M., Gotein M., Doppke K., Kutcher J. G., LoSassc T., Mohan R., Munzenrider J. E., Sontag M., and Wong W. The role of uncertainty analysis in treatment planning. Int. J. Radiation Oncology Biol. Phys. Vol. 21; 91-107, 1991.

9-           George T. Y., and Charles A. P. The role of imaging in tumor localization and portal design. In: New York, Alfred R. Smith (ed.) Radiation therapy physics, Springer-Verlag, 1995.

10-      Sandler H., McShan D., and Lichter A. S. Potential improvement in the results of irradiation for prostate carcinoma using improved dose distribution. Int. J. Radiation Oncology Biol. Phys. Vol. 22; 361-367, 1991.

11-      International Commission of Radiation Units and Measurements. Prescribing, recording, and reporting photon beam therapy, Bethesda, MD: ICRU Report 50, 1993.

12-      Roach III M., Pyckett B., Rosenthal S. A., Verhey L. and Pillips T. Defining treatment margins for six-field conformal irradiation of localized prostate cancer. Int. J. Radiat. Oncol. Biol. Phys. 28:267-275, 1994. 

13-      Low N. N., Vijayakumar S., Myrianthopoulos L. C., Sutton H., Krishnasamy R., Spelbring D. R., and Chen G. T. Y. Beam’s eye view based prostate treatment planning: Is it useful? Int J Radiat Oncol Biol Phys 19:759-768, 1990.

14-       Ten R. K., Forman J. D., and Heimburger D. K. Treatment planning issues related to prostate movement in response to differential filling of the rectum and bladder. Int. J. Radiat. Oncol. Biol. Phys. 20; 1317-1324, 1991.

15-      Deborah A. K. and Anas M. E. Cancers of the Genitourinary Tract. In; Treatment planning in Radiation Oncology; Faiz M. Khan (ed.), Williams & Wilkins, London, 1998.

16-      Austin-Seymour M., Kalet I., and McDonald J. Three-dimensional planning target volumes: A model and software tool. Int. J. Radiat. Oncol. Biol. Phys. 33:1073-1080, 1995.

17-      Stroom J. and Storchi P. Automatic calculation of three-dimensional margins around treatment volumes in radiotherapy planning. Phys Med Biol 42:745-755, 1997.

18-      Bedford J. L. and Shentall G. S. A digital method for computing target margins in radiotherapy. Med Phys 25:224-231, 1998.

19-      Ketting C., Austin-Seymour M., and Kalei I. Evaluation of an expert system producing geometric solids as output. Proceedings of the 19lh Annual Symposium on Computer Applications in Medical Care. Philadelphia, PA: Hanley and Belfus Inc.; 1995. p. 683-687.

20-      Ketting C., Austin-Seymour M., and Kalet I. Consistency of three-dimensional planning target volumes across physicians and institutions. Int J Radiat Oncol Biol Phys 1997; 37:445-453.

21-      Ketting C. Austin-Seymour M., and Kalet I. Automated planning target volume generation. An evaluation pitting a computer-based tool against human experts. Int J Radial Oncol Biol Phys 1997; 37:697-704.

22-      Bijhold J., Lebesque J. V., Hart A. A. M., and Vijlbrief R. E. Maximizing setup accuracy using portal images as applied to a conformal boost technique for prostatic cancer. Radiother. Oncol. 24:261-271, 1992.

23-      Lee M., Wynn C., Webb S., Nahum A. E. and Dearnley D. A. Comparison of proton and megavoltage x-ray treatment planning for prostate carcinoma. Radiother. Oncol. 33:239-253, 1994.

24-      Neal A. J., Oldham M. and Dearnaley D. P. Comparison of treatment techniques for conformal radiotherapy using dose-volume histograms and normal tissue complication probabilities. Radiother. Oncol. 37:29-34, 1995.

25-       Oldham M. and Webb S. The optimization and inherent limitations of 3D conformal radiotherapy treatment plans of the prostate. Br. J. Radiol. 68:882-893, 1995.

26-      Soffen E. M., Hands G. E., Hwang C. C. and Chu J. C. H. Conformal static field radiation therapy treatment of early prostate cancer with rigid immobilization. Int. J. Radiat. Oncol. Biol. Phys. 20:141-146, 1990. 

27-      Ten Haken R. K., Perez-Tamayo C., Tesser R. J., McShan D. L., Fraas B. A. and Lichter A. S. Boost treatment of the prostate using shaped fixed fields. Int. J. Radiat. Oncol. Biol. Phys. 16:193-200, 1989.

28-      Leibel S. A. Clinical trials with three-dimensional conformal radiation therapy in carcinomas of the nasopharynx, prostate and lung. In: Acta of ‘International Congress on Advanced Diagnostic Modalities and New Irradiating Techniques in Radiotherapy’, pp. 117-123 Gobbi-Latini, Perugia, 1994.

29-      Claudio F., Michele R., Giovanni M. C., Angelo B., and Riccardo C. Comparing 3-, 4-, and 6-fields techniques for conformal irradiation of prostate and seminal vesicles using dose-volume histograms. Radiother. Oncol. 44, 251-257, 1997.

30-      Vincent S. K., James L., Bedford S. W., and David P. Dearnaley. Comparison of 2D and 3D algorithms for adding a margin to the gross tumor volume in the conformal radiotherapy planning of prostate cancer. Int J Radiat Oncol Biol Phys 42 (3): 673-679, 1998.