0022-534 7 /90/1432-0398$02.00/0 Vol. 143, February

THE JOURNAL OF UROLOGY Copyright© 1990 by AMERICAN UROLOGICAL ASSOCIATION, INC.

Printed in U.S.A.

PHOTODYNAMIC THERAPY FOR LOCALISED PROSTATIC CANCER: LIGHT PENETRATION IN THE HUMAN PROSTATE GLAND M. L. PANTELIDES,* C. WHITEHURST, J. V. MOORE, T. A. KING AND N. J. BLACKLOCK From the University Department of Urology, University Hospital of South Manchester, The Wolfson Laser and Fibre Optic Unit, Physics Department, Manchester University, and The Paterson Institute for Cancer Research, Christie Hospital, Manchester, United Kingdom

ABSTRACT

We are investigating the feasibility ofphotodynamic therapy in the treatment of localised prostatic cancer. Of major importance in this form of treatment is the extent to which light penetrates the target organ; hence, our interest in the optical properties of the human prostate gland. We obtained three whole prostates from autopsies of patients who died of non-urological causes. Red light was launched interstitially and detector fibres measured light intensity as a function of distance from the delivery fibre end. The optical constants derived from the three prostates were almost identical and indicated that light was predominantly scattered rather than absorbed (mean absorption and scattering coefficients 0.07 ± 0.02 mm.- 1 and 0.86 ± 0.05 mm.- 1 respectively). In a comparison of the tissue penetration by four different wavelengths, 633 nm red light was found to be transmitted best. Light propagation in the heavily absorbing tissue of the human liver was 4.3 times poorer than in the prostate. Such a combination of low absorption and high scattering characteristics in prostatic tissue would enhance the effectiveness of PDT. The optical constants derived will enable "light treatment planning" in patients with prostatic cancer. (J. Ural., 143: 398-401, 1990) The optimal treatment for locally confined prostatic cancer is still to be decided. Managements advocated at present include hormonal manipulation, radical prostatectomy, external beam or interstitial irradiation and conservative follow up with deferred therapy. The lack of consensus is due to the good prognosis of many patients with early disease, their elderly age and the significant morbidity associated with current radical therapies. Efforts to select patients for treatment by predicting the biological behaviour of prostatic tumors have been largely unsuccessful1· 2 and alternative approaches aimed at minimising the invasiveness or complications from radical therapy deserve consideration. Seeking such a therapeutic option, we are at present investigating the feasibility of photodynamic therapy (PDT), a new treatment modality for solid tumors involving the interaction of visible light with photosensitising drugs such as porphyrins. This gives rise to toxic species of oxygen and chemical radicals that cause tumor necrosis. Interest in PDT has been enhanced by reports of higher porphyrin accumulation in tumor than in adjacent normal tissue. 3 • 4 PDT has been shown to be cytotoxic to Dunning R3327 rat prostatic adenocarcinoma cells both in vitro 5 and in vivo 6 • 7 and has achieved clinical usefulness in patients with predominantly recurrent malignancies refractory to treatment by conventional means. 8 Controlled light energies can now be delivered interstitially by the coupling of optical fibres to suitable lasers. Furthermore, advances in ultrasound scanning technology with the availability of the per-rectal probe, have enabled an accurate delineation of the prostate gland and the percutaneous placement of radioactive grains in relation to a cancer. 9 • 10 We propose to use such means for the percutaneous placement of light delivery optical fibres into a gland sensitised by a porphyrin derivative. The PDT effect on tumor depends on the quantity of light energy available for interaction with the porphyrin molecules. This is proportionate to the applied laser power and the tissue characteristics. The latter are comprised of the absorption Accepted for puolication September 14, 1989. * Requests for reprints: Radiobiology Laboratory, Paterson Institute of Cancer Research, Christie Hospital & Holt Radium Institute, Manchester M20 9BX, United Kingdom.

coefficient, La, which is the probability per unit length (mm.- 1 ) of tissue of a photon being absorbed, the scattering coefficient, Ls, which represents the probability per unit length of tissue (mm. -i) of a photon being scattered, and the degree of anistropy of light within tissue which indicates any preferential directional scatter with inhomogeneity in distribution. The optical characteristics of the human prostate gland have not been reported before and are essential for precise 'light treatment planning' necessary in PDT. MATERIALS AND METHODS

Prostate glands. We obtained three whole prostate glands and also for comparison, one specimen of liver from autopsies of patients who died of non-urological causes during the preceding 24 to 48 hours. The patients' ages were 90, 68 and 73 years and their respective prostate gland weights were 82, 40 and 38 gm. The macroscopic appearance of the largest gland (No. 1) was consistent with benign hyperplasia. Light source. Red light (633 nm wavelength) from a heliumneon (He-Ne) laser was used throughout, except in the investigation into the effects on tissue penetration by different wavelengths of incident light (red, orange, yellow, green), where a different type of He-Ne laser able to deliver light at the above wavelengths was used. Fibre optic delivery and detector systems. Standard optical fibres consisting of a 400 µm. silica core surrounded by a chemically bonded polymeric coating were used (fig. 1). To measure the relevant optical data in the tissue, we used two different types of fibre termination. Firstly, a standard squarely cleaved tip with an acceptance angle of about 20' (fig. 1). (The acceptance angle is the maximum angle of incidence of light on a fibre end which can be transmitted by the fibre"). This arrangement was suitable for measuring light flux in the isotropic or diffusion dominant regime of the tissue to determine the effective attenuation coefficient. The other, was a small aperture detector with an acceptance angle of about 1.5° consisting of the same cleaved fibre, but with an iris diaphragm interposed between the detector fibre and photodiode. This detector could measure the total attenuation coefficient in the absorption dominant regime of the tissue. Light from the He-

398

399

PHOTODYNAMIC THERAPY FOR PROSTATIC CANCER

OPTICAL FIBRE

1 Relative radiance in Prostate vs.distance from delivery fibre in three different directions

1~ \

~~ f

>,

'

-1

10

ii

3

G= acceptance angle Q)

• Forward .. Sideways

-~ ~ -~ -~

leff=0.45mm- 1

!~.

~-~

.2:

-;;;

FIG. 1. Longitudinal section of standard optical fibre demonstrating silica core surrounded by polymeric coating (stippled area).

• Backward

-~~ ·~~.t

~l,

ai 0::

°""!~-

163 L----,--.---,--.--r-,--,---,--,---,---,,-~~-;,,.~~ 0

2

4

6

8

10

12

14

Distance(mm)from end of delivery fibre

FIG. 3. Study of radiance in three different directions in prostate gland No. 1 using 633 nm wavelength red light. TABLE 1.

Optical characteristics of human prostates Prostate Prostate Prostate Mean± SD No. l No. 2 No. 3

Effective attenuation coefficient 2:eff (mm.- 1 ) Penetration depth 1/2:eff (mm.) Scattering coefficient 2:s (mm.- 1 ) Absorption coefficient 2: a (mm.- 1 ) Albedo c (2:s/2:t)

FIG. 2. Experimental set up. Optical fibres within needles fixed in prostate (P) using aluminum frame (A). Needles carrying delivery fibre (D) and backward (B), forward (F) and sideways (S) facing detectors.

Ne laser was coupled into the delivery fibre by means of a lens system, whilst the detector light was focused onto a photodiode and oscilloscope. Experimental techniques. The anterior fibromuscular tissue of each prostate gland was divided along the line of the prostatic urethra. The opened urethra was then placed face down permitting the interstitial placement of light delivery and detector fibres using 19 gauge needles (fig. 2). These were rigidly held in place by an aluminum frame. We have used the method of Doiron, Svaasand and Profio 12 (1983) to obtain light intensity measurements at various points from the delivery fibre end by means of three different detector fibres placed at three different angles (fig. 2). The forward detector end was placed opposite the delivery fibre end, the sideways detector was at 90° and a backward facing detector was almost parallel to the delivery fibre. This system enables the assessment of anisotropy and the calculation of space irradiance which is the absolute quantity of light flux incident from all directions on a sphere of microscopic dimensions divided by its cross-sectional area. The position of the light delivery fibre in relation to the three detectors was altered by advancing it at 1 mm intervals. Thus, measurements of relative radiance, in effect relative light intensity in three different directions, as a function of distance from the delivery fibre end were made. RESULTS

The relative light intensity measurements in prostate No. 1 as a function of distance from the delivery fibre end in the three different directions are demonstrated in figure 3. It can be seen that at distances >4 mm. from the delivery fibre end there was an exponential fall in light intensity and due to

0.45

0.37

0.48

0.43 ± 0.05

2.2 0.88 0.08

2.7 0.91 0.05

2.08 0.80 0.10

2.31 ± 0.3 0.86 ± 0.05 O.o?± 0.02

0.92

0.95

0.89

0.92 ± 0.02

multiple scattering the distribution of light within the prostate gland was isotropic i.e. homogeneous, with both relative radiance and rate of attenuation in the three different directions almost identical. Here, (>4 mm.) diffusion theory laws are applicable and the gradient of the attenuation curves gives the effective attenuation coefficient Leff= 0.45 mm.- 1 In diffusion theory: ~

eff

= (3

~.~.) 112

1

The inverse of Leff, ( eff) represents the penetration depth, which is the thickness of prostatic tissue by which 63% of the incident light is lost. This value in the three glands ranged from 2.08 to 2.7 mm. with a mean of 2.31 ± 0.3 mm. (table 1). When the wavelength of incident light was varied from 633 nm red, to orange, yellow and green (612, 594 and 543 nm wavelengths respectively), the light intensity detected by the forward facing detector in prostate No. 1 decreased, with a corresponding reduction in the penetration depth from 2.2 mm. to 0.55 mm (fig. 4). Thus, the depth of tissue penetration by red light was four times greater than that achieved by green. Similarly, in a comparison of the attenuation of red light between prostatic tissue and human liver (fig. 5) the penetration depth in prostate was 4.3 times greater than in liver. According to the laws of geometrical optics as reported by Doiron et al. 12 (1983), in an absorption dominant situation at distances close to the delivery fibre end, the total attenuation coefficient, (L,), is given by: 1/,,

~, = ~. + ~.

2

This, when derived using a small aperture detector was = 0.96 mm.- 1 (fig. 6). Having evaluated equations 1 and 2 experimentally in prostatic tissue, it was then possible to calculate the two constantsLa and L,-in all three glands (table 1). It can be seen from table 1 that the effective attenuation

400

PANTELIDES AND ASSOCIATES Relative radiance in Prostate vs.distance from delivery fibre

'k·\

!

1/leff

y·\·\

I 1

10

!

at different wavelengths

• 633nm 4 612nm • 594nm x 543nm

t. . ""~·"'· •'-sx\

rn'

DISCUSSION

"'•"

\\

j '

2.2mm 1.6mm 1.1mm 0.55mm

'",







\ \ , "-,.

10 0

4

2

10

8

6

12

14

Distance(mm)from end of delivery fibre

FIG. 4. Relative radiance in prostate v. distance from delivery fibre, at different wavelengths.

•~Companson between radiance 1

in

~ A

..

-~~

:, >,

1/Ieff

Prostate O 45mm- 1 2 2mm • Liver 1 96mm- 1 0.51mm

\

:

prostate and human liver

ir\~

C

\ '-s.."-...,.

-1

10

\. "'. ~

...

\

Cl)

·"1ii

~ ...

Cl)

a:

-3

10

0

2

4

8

6

12

10

14

Distance(mm)from end of delivery fibre

FIG. 5. Comparison of radiance in forward direction between prostate and human liver using 633 nm wavelength red light.

Relative radiance in Prostate vs.distance from delivery fibre 11

Forward detector with small acceptance angle (1.5°)

"'

·1:

Il=Ia+rs mm- 1

:, >,

f

coefficients and penetration depths in the three glands were essentially identical despite the discrepancy in size and pathology. At distances of 2.08 to 2.7 mm. from the emitting fibre end due to an exponential decay, light intensity was reduced by 63%. The high scattering and low absorption coefficients with an albedo o:sh:,) of almost 1 indicate that the prostate gland is predominantly a light scatterer rather than absorber.

= 0.96 mm'

-1

10

~

The delivery of maximum therapeutic dose to tumor, whilst sparing the adjacent normal tissues, is essential to the use of ionising irradiation. With this in mind, Flocks 13 enhanced the dose of radiation delivered to the prostate by intraglandular injection of radioactive colloidal gold and, more recently, radioactive gold (Au 198) and iodine (11 25 ) grains have been placed interstitially within the gland. Dosimetry calculations are based on the half lives of Au 198 and 11 25 and the tissue penetration depth of their radioactivity. Similarly, the depth of penetration of light and photosensitising drug concentration within tumor are essential considerations in PDT where the dose of light available for drug interaction is proportional to both administered laser energy and the tissue optical characteristics. The latter are difficult to measure and are known in a limited number of human tissues (table 2). In most cases, clinical PDT is based on empirically derived values and for optimal use the precise control of both light and drug dosimetries will be necessary. 12 Ideally, light propagation within tumors, should be studied in patients. This is however technically impossible at present and since both ourselves (unpublished data) and others 12• 15 have established that at the wavelength of current interest for interstitial PDT (633 nm) the tissue penetration of light in vivo is very similar to that in vitro, we believe that the values derived, should represent those of prostatic tissue in vivo. During an interstitial light application, photons are either scattered by tissue components or are absorbed by pigments such as hemoglobin. We demonstrated that prostatic tissue is a far better scatterer of light than absorber. A rapid absorption such as in liver tissue, decreases the depth of penetration. Good scattering properties however, are favourable since they enhance the homogeneity in the distribution of light within tissue, and increase the probability of collision between photons and photosensitising drug. In addition, high scattering would promote reflectance from tissue boundaries, such as the prostatic capsule and the fascia of Denonvilliers, with confinement of light and treatment field within the gland. As seen from table 2, the attenuation of light in the prostate gland compares well with that of the two least attenuating human tissues-neonatal brain and human kidney. However, a strict comparison between studies is impossible due to differences in techniques used. For example, it is known that the width of incident light beam would influence the depth of tissue penetration 16 and, whereas Svaasand11 delivered surface illumination and placed detector fibres interstitially, we implanted both delivery and detector fibres intraglandularly since this resembles our proposed method of treatment. Using identical methods in this study, the penetration depth in prostate was

.::: TABLE 2.

Cl) (.)

Attenuation coefficients of human tissues

C

"'

~

10'

Human Tissue

Cl)

·" 1ii Cl)

a:

- 3

10

0

2

4

6

8

10

12

14

Distance(mm)from end of delivery fibre

FIG. 6. Relative radiance in prostate vs. distance from delivery fibre. Forward detector with small acceptance angle (1.5°).

Adult brain Neonatal brain Liver Kidney Lung Lung tumor Bronchial mucosa Liver Prostate gland

Effective Attenuation Coefficient ~eff mm.- 1 of Tissue Using 633-5 nm Wavelength 1.25 0,25

1.1 0.4 1.1 0.63 0.91 1.96 0.37-0.48

Svaasand" (1984) Svaasand11 (1984) Eichler et al. 14 (1977) Eichler et al. 14 (1977) Doiron et al. 12 (1983) Doiron et al, 12 (1983) Doiron et al. 12 (1983) This work This work

PHOTODYNAMIC THERAPY FOR PROSTATIC CANCER

found to be 4.3 times greater than in human and such good light transmittance by prostatic tissue, would enhance the depth of tumor necrosis with PDT. The effective coefficient for human kidney extracted from the literature 14 seems extremely low since kidney cortices are darker and better perfused than even the liver. Values of Ieff from bovine 17 and porcine 12 kidneys were 0.79 mm.- 1 and 0.4 mm.- 1 respectively. Thus, attenuation in animal models is also relatively low and considering the differences in experimental techniques, such results are highly comparable to those in the human. In contrast, the attenuation in liver is far greater than that in kidney. In this study ~)ff for human liver was 1.96 mm.- 1 (fig. 5). The corresponding value according to Eichler et al. 14 was 1.1 mm.- 1 • I;eff in livers from the pig12 and rabbit 15 were 1.3 mm.- 1 and 1.25 mm.- 1 respectively, comparable to the human. Whilst the dark colour of liver tissue is considered responsible for high attenuation, results from the kidney indicate that other contributory factors, for example, the bilirubin and bile salts, might be of importance. From the wavelengths tested, 633 nm red light offered maximum tissue penetration and hence should be optimal for interstitial PDT. Using moderate laser energies of this wavelength (100 J cm.- 2 ), depths of necrosis approximately 0.5 cm. were obtained in experimental tumors. It is clear, therefore, that for percutaneous treatment of localised prostatic cancer with PDT, the placement of multiple fibres will be necessary. The mean penetration depth value of 2.31 mm. for 633 nm red light and the optical coefficients derived, will enable the computer modelling with work on isodose contours for optimisation of the number and disposition of optical fibres within the gland. In view of the often multifocal nature and peripheral location of prostatic tumours, 18 the whole gland and in particular the parts adjacent to the rectum, will require treatment. It is noteworthy that despite the disparity in size and pathology between the three prostate glands, the optical properties were almost identical. Since macroscopically, resectedprostatic adenocarcinoma is generally indistinguishable from benign tissue, 19 the presence of malignancy should not alter the optical properties significantly. However this aspect, together with the transmittance of light to adjacent rectum and bladder, and the possible effects of intraprostatic calcification, requires evaluation. Beisland and Sander 20 reported the use of the NeodymiurnYAG laser to achieve photocoagulation of residual malignant tissue following per urethral resection. In PDT, tissue necrosis is produced by a photochemical process activated by light. We believe that the combination of favourable optical characteristics and a means of accurate ultrasonic guidance for percutaneous light administration, will enable the treatment of localised prostatic tumors with PDT. Acknowledgments. The authors would like to thank Dr N Haboubi for his assistance with specimen collection. Dr J V Moore and Dr C VVhitehurst are funded by the Cancer Research Campaign and the Gardner Bequest respectively. REFERENCES

1. Epstein, J. I., Paull, G., Eggleston, J.C. and Walsh, P. C.: Prognosis of untreated state Al prostatic carcinoma: a study of 94 cases with extended follow up. J. Urol., 136: 837, 1986.

401

2. Ritchie, A. W. S., Dorey, F., Layfield, L. J., Hannah, J., Lovrekovich, H. and de Kernion, J. B.: Relationship of DNA content to conventional prognostic factors in clinically localised carcinoma of the prostate. Br. J. Urol., 62: 254, 1988. 3. Kelly, J. F., Snell, M. E. and Berenbaum, M. C.: Photodynamic destruction of human bladder carcinoma. Br. J. Cancer, 31: 237, 1974. 4. Lipson, R. L., Baldes, E. J. and Olsen, A. N.: Derivative of hematoporphyrin in tumor detection. J. Nat. Cancer Inst., 26: 1, 1961. 5. Camps, J. L., Powers, S. K., Beckman, W. C., Brown, T. J. and Weissman, R. M.: Photodynamic therapy of prostate cancer: an in vitro study. J. Urol., 134: 1222, 1985. 6. McPhee, M. S., Thorndyke, C. W., Thomas, G., Tulip, J., Chapman, D. and Lakey, W. H.: Interstitial applications oflaser irradiation in hematoporphyrin derivative-photosensitized Dunning R3327 prostate cancers. Laser Surg. Med., 4: 93, 1984. 7. Gonzalez, S., Arnfield, M. R., Meeker, B. E., Tulip, J., Lakey, W. H., Chapman, J. D. and McPhee, M. S.: Treatment of Dunning R3327-AT rat prostate tumors with photodynamic therapy in combination with Misonidazole. Cancer Res., 46: 2858, 1986. 8. Dougherty, T. J.: Photodynamic therapy (PDT) for malignant tumors. CRC Crit. Rev. Oncol. Hematol., 2: 83, 1985. 9. Holm, H. H., Juul, N., Pedersen, J. F. and Stroyer, I.: Transperineal 125 Iodine seed implantation in prostatic cancer guided by transrectal ultrasonography. J. Urol., 130: 283, 1983. 10. Crusinberry, R. A., Kramolowsky, E. V. and Loening, S. A.: Percutaneous transperineal placement of gold198 seeds for treatment of carcinoma of the prostate. Prostate, 11: 59, 1987. 11. Svaasand, L. 0.: Optical dosimetry for direct and interstitial photoradiation therapy of malignant tumors. In: Porphyrin Localization and Treatment of Tumors. Edited by Doiron, Gomer. Alan R Liss, Inc., pp. 91-114, 1984. 12. Doiron, D. R., Svaasand, L. 0. and Profio, A. E.: Light dosimetry in tissue: application in photoradiation therapy. In: Porphyrin Photosensitisation (Adv. Exp. Med. Biol. Vol. 160). Edited by Kessel, D. and Dougherty, T. J. Plenum Press, New York, p. 293 (1983). 13. Flocks, R. H.: Interstitial irradiation therapy with a solution of Au 198 as part of combination therapy for prostatic carcinoma. J. Nucl. Med., 5: 691, 1964. 14. Eichler, J., Knof, J. and Lenz, H.: Measurements on the depth of penetration of light (0.35-1.0 µm) in tissue. Rad. Environ. Biophys., 14:. 243, 1977. 15. Wilson, B. C., Jeeves, W. P. and Lowe, D. M.: In vivo and post mortem measurements of the attenuation spectra of light in mammalian tissues. Photochem. Photobiol., 42: 153, 1985. 16. Marijnissen, J.P. A. and Star, W. M.: Phantom measurements for light dosimetry using isotropic and small aperture detectors. In: Porphyrin Localization and Treatment of Tumors. Edited by Doiron, Gomer. Alan R Liss Inc., pp. 133-148, 1984. 17. Preuss, L., Bolin, F. and Cain, B.: Tissue as a medium for laser light transport-implications for photoradiation therapy. In: Proc. Spie-Lasers in Medicine and Surgery, 357. Edited by L. Goldman, pp 77-84, SPIE Bellingham, Wash.1982. 18. Byar, D. P. and Mostofi, F. K. and the Veterans Administration Cooperative Urological Research Group: Carcinoma of the prostate: prognostic evaluation of certain pathologic features in 208 radical prostatectomies. Cancer, 30: 5, 1972. 19. Kahler, J.E.: Carcinoma of the prostate gland: a pathologic study. J. Urol., 41: 557, 1939. 20. Beisland, H. 0. and Sander, S.: First clinical experiences in Neodymium-YAG laser irradiation of localized prostatic cancer. Scand. J. Urol. Nephrol., 20: 113, 1986.

Photodynamic therapy for localised prostatic cancer: light penetration in the human prostate gland.

We are investigating the feasibility of photodynamic therapy in the treatment of localised prostatic cancer. Of major importance in this form of treat...
179KB Sizes 0 Downloads 0 Views