0360.3016/91
$3 00 + .lc
l Phase I/II Clinical Trials
PHOTODYNAMIC
THERAPY FOR CHEST WALL RECURRENCE IN BREAST CANCER
PAUL W. SPERDUTO, M.D., THOMAS F. DELANEY, M.D., GUNTER THOMAS, PAUL SMITH, PH.D., LAURA J. DACHOWSKI, R.N., ANGELO Russo, M.D., PH.D., ROBERT BONNER, PH.D. AND ELI GLATSTEIN, M.D. RadiationOncology Branch, Clinical Oncology Program, Division of Cancer Treatment, National
Cancer Institute,
National
Institutes
of Health,
Bethesda,
MD 20892
Photodynamic therapy is the use of a sensitizer (dihematoporphyrin ethers) which is preferentially retained in tumor cells and activated by subsequent light delivery resulting in a selective tumoricidal effect. Between1986 and 1989, we treated 20 patients with photodynamic therapy for chest wall recurrence of breast cancer. Responses were seen (20% complete response, 45% partial response, 35% no response), but the duration of response was short (average 2.5 months). Complications, in decreasing frequency, included pain, ecchymoses, blistering, ulceration and necrosis in the area of tumor involvement on the chest wall. One patient required skin flap reconstruction for full thickness necrosis. A limitation to this mode of therapy is that the sensitizer currently used is activated by light at a wavelength of 630 nm. This light can penetrate to a tissue depth of only 0.5 to 1.0 cm; thus, deeper disease cannot be treated. Future research must focus on the development of a clinically useful photosensitizer that can be activated by light at longer wavelengths and thereby achieve deeper tissue penetration. This would greatly expand the patient population for which this therapy is useful.
Hematoporphyrinderivative,Breast cancer,Laser, Photosensitizer. INTRODUCTION Much research has focused on the development of cancer treatment modalities that selectively destroy tumor, without disruption of normal cell and tissue function (7). Light-activated photosensitizers demonstrate relatively selective tumoricidal effects. Many believe this is due to the photosensitizer being retained longer in higher concentrations in neoplastic tissue than in most normal tissues (6). Subsequent light activation preferentially destroys sensitized tumor cells and not the relatively unsensitized surrounding normal tissues. The use of light-activated chemicals in biologic systems dates back to 1900 when Raab reported the lethal effect of light on paramecium treated with an acridine dye ( 14). When given alone, neither the light nor the dye were cytotoxic. Since then, a variety of sensitizers have been investigated but the focus of most research has been on the porphyrins. The mechanism of cell death is the oxidation of vital cellular structures, postulated to be critical mitochondrial or plasma membrane targets. This cytotoxicity is thought to be mediated by singlet oxygen, a reactive oxygen species generated by the interaction of ground state
Reprint requests to: Dr. Sperduto, Rm B3-B69, Bethesda, MD 20892.
NIH, NCI-ROB,
molecular oxygen with photosensitizer which has been activated by the appropriate wavelength visible light (1, 4, 10). Hematoporphyrin derivative (HpD) or the more active preparation of dihematoporphyrin ethers (DHE) have been used for tumor detection and in the treatment of tumors of skin (2), bladder ( 13) brain (8) and bronchus (5, 1 I), as well as gynecologic cancers (15), peritoneal tumors ( 18), and locally recurrent breast carcinoma (3). 630 nm wavelength red light from a laser source has generally been used in the clinic to activate the photosensitizer because of its relatively favorable tissue penetration compared to other wavelengths which may also activate the sensitizer ( 1, 4, 10). Schuh et al. reported 14 patients who received 30 treatments of photodynamic therapy (PDT) for locally recurrent breast carcinoma: 2 of 30 (7%) had a complete response (CR), 22 of 30 (73%) had a partial response (PR), and 4 of 30 ( 13%) had no response (NR) ( 16). The duration of response varied from 6 weeks to 8 months. The purpose of this communication is to report our experience at the National Cancer Institute with PDT for locally recurrent breast carcinoma.
Accepted
Bldg. 10,
441
for publication
25 January
199 1.
1. J. Radiation Oncology 0 Biology 0 Physics
442 METHODS
AND
MATERIALS
Patients Between September 1986 and January 1989,20 women with histologically proven recurrent breast cancer of the chest wall were treated with photodynamic therapy at our institution as part of a Phase I study. All patients had failed conventional therapy (surgery in 17, radiation in 19, and chemotherapy in 17). Table 1 presents the patient characteristics. All had multiple cutaneous and subcutaneous nodules, ranging in size from 2 mm to confluent tumor-infiltrated rashes spanning more than 5 cm (Fig. 1). Patients were selected for treatment on the protocol if grossly visible or imageable disease could be adequately treated by the known 0.5- 1.O cm light penetration of the 630 nm external light beam used. Patients had to have adequate hepatic function (normal or common toxicity criteria grade 1 elevation of liver function tests) because of the hepatic clearance of the photosensitizer. At the time of the first PDT treatment, 8 patients had local chest wall disease only and 12 had distant metastases in addition to chest wall disease.
Pt
Age
Extent of disease
Prior treatment
parameters
HpD dose w/kg
and response
Number of treatments
Fig. 1. The chest wall of a patient (#9) prior to photodynamic therapy with multiple cutaneous nodules and a confluent biopsyproven tumor-infiltrated rash.
planation of the treatment and possible side-effects. They were specifically counseled about photosensitivity and the need to exercise caution, or avoid whenever possible, direct sunlight and bright lights for 6 weeks after therapy. Photosensitizer The photosensitizer used in the trial* was a preparation of dihematoporphyrin ethers (DHE) containing a higher concentration of the photosensitizing component of the hematoporphyrin derivative (HpD) (9) formulated as a
Informed consent Informed consent in conformity with guidelines established by the Institutional Review Board and the NIH Clinical Center was obtained from all patients after ex-
Table 1. Treatment
July 1991, Volume 21, Number 2
to photodynamic
Number of cycles
Total light dose J/cm’
therapy
Power density mW/cm2
for chest wall recurrence
Postinjection interval days
Response
Duration of response months
1
57
DM
mc
1.5
2
1
20
80
2. 3
CR
1.5
2 3 4 5 6 7 8
58 45 46 59 73 56 13
L DM L DM DM DM DM
mrch
1.5 1.5
2 2
1 1 1 1
1
40 24 30 30 35 35 27
33 17 39 18 76 24 50
3. 4 3. 9 2 2. 3 2 3 2. 3. 4
PR NR NR PR NR CR PR
1.0 0 0 2.0 0 14 4.0
9
40
DM
10 11 12
55 72 44
13 14 15 16 17 18 19 20
ITWC
of breast cancer
Complications e. e. e. e. e. e. e. e.
P P. p. P. P, P, P. P.
b b. u n u, n b, n n n
K
1.5
1
mrch
1.5 1.5 1.5 1.5
2 1 1 3
IllIT
1.5 .75
6
3
42
I
2, 3. 4
PR
4.0
e. P, n
DM L L
mrh mrh lmrch
1.5 1.5 1.5
4 10 10
2 3 3
34 145 178
2 4 5
2. 3. 4 2. 3. 4 2. 3.4
NR CR PR
0 6.5 1.0
e. P e. p, b. n e. P. u. n
44 55 69 53 41
L L L DM DM
mrc mrch mrch mrc mrch
1.5 1.5 1.5 1.5 1.5
2 5 14 5 3
1 2 4 2 1
26 73 359 81 48
7 4 7 4 5
2. 3 2, 3.4 2. 3. 4 2, 3, 4 2, 3, 4
NR PR CR PR PR
0 3.0 6.3 2.0 1.5
e. e. e, e. e
39 16 58
DM L DM
IIlK
1.5 1.5 1.5
2 7 3
1 3 1
34 139 36
5 6 4
2. 4 2, 3, 4 2, 3, 4
PR NR NR
2.5 0 0
e. P e. P. n e, P
K
mrh mrch
lrch IIlK
I I
L = Local disease only; DM = distant metastases; m = mastectomy; c = chemotherapy; r = radiation therapy; h = hormonal therapy; = photofrin II drug dose in mg/kg; CR = complete response; PR = partial response; NR = no response; e = erythemia: p = pain; b = blistering;
* Photofiin@ II, Quadra British Columbia, Canada.
Logic
Technologies,
Vancouver,
Comments
Skin flap required
P P. u. n P, b, u P
I = lumpectomy; u = ulceration;
No pain
DHE dose n = necrosis.
Chest wall PDT 0 P. W. SPERDUTO rful.
sterile solution of 2.5 mg/ml dissolved in 0.9% NaCl solution. The drug was administered as an investigational agent under the IND held by the Division of Cancer Treatment, National Cancer Institute 48-72 hr prior to planned light delivery. The photosensitizer was given to the patients as outpatients at a dose of 1.5 mg/kg in 50 ml of normal saline by intravenous infusion over 15 min. Light administration Light was delivered with 600 micron optical fibers coupled to a Coherent Innova20W argon laser/Model 599 Dye Laser System, emitting up to 5.0 W at a wavelength of 630 f 2 nm. Surface illumination was provided via 600 micron flat cut quartz fibers with loops placed between the laser coupling and the tip to produce uniform power density across the treatment field. Table 1 presents the photodynamic therapy details in our patients. Each patient received DHE 1.5 mg/kg IV at time zero. Initially patients were treated with a single fraction of light 72 hr after drug administration. Rather than setting a doseescalation policy prior to initiation of this highly experimental protocol, we allowed our early experience to define later dose escalation. We reasoned that the best approach, extrapolating from experience with external beam irradiation, would be to treat the entire chest wall and sequentially escalate the dose as tolerated. It was found that a single dose of 35 joules/cm2 would take nearly an entire day to deliver and cause full thickness chest wall necrosis. Accordingly, in the interest of shortening treatment time and maximizing the opportunity for repair of sub-lethal damage in normal tissues, we then adopted a multiple fraction treatment program. We injected the DHE on day 0 and delivered light on days 2, 3, and 4. This approach was further justified because there is no reliable method of measuring the level of photosensitizer in the tissue or tumor. Highly variable levels are suggested by the variable responses to similar doses of light. Thereafter, the light dose was sequentially escalated by 3-5 joules/cm2/treatment in successive patients. The only deviations from this schema resulted from: a) patient intolerance (when debilitating chest wall discomfort occurred after one or two fractions, the third fraction was deleted or reduced), orb) inadequate treatment effect (when little effect was seen after two fractions, relative to other patients at similar intervals after treatment, the dose of the third fraction was increased and/or additional fractions were given, as detailed in Table 1). Whenever possible, it was attempted to deliver at least two cycles of treatment per patient with the second or subsequent cycles of PDT at monthly intervals (Table I). A cycle was defined as the period between an IV injection of the photosensitizer to the next IV injection of the photosensitizer, regardless of the number of light treatments following the photosensitizer. Light fields were designed to include all visible disease with at least 3 cm margins. These fields encompassed the ipsilateral chest wall, extending laterally to the mid-axillary line, superiorly to
443
above the clavicle, inferiorly several centimeters below the contralateral inframammary fold and medially to the midline (Fig. 2). The total light dose varied from 20 to 359 joules/cm2. The power density (light dose rate) varied inversely with field size and ranged from 1.2 to 80.0 mW/ cm2. At power densities less than approximately 100 mW/ cm*, there is minimal elevation in temperature and the tissue reaction is felt to represent a photodynamic reaction, not a thermal one ( 17). Photographs were obtained before and after each cycle and during follow-up (see Fig. 1). Punch biopsies (3 mm) were obtained before and at the 4-6 week follow-up visit in at least one site of disease. Because most of these patients had diffuse dermal involvement without measurable nodules, we have defined complete response as clinical and pathologic (biopsyproven) regression of all tumor in the treatment field. A partial response was defined as greater than 50% reduction of measurable nodules or a complete clinical regression with residual microscopic disease. A no response was defined as less than a 50% reduction, no change, or progression of disease. RESULTS Follow-up in all patients extended beyond the time of local recurrence or until death. This period ranged from 1- 18 months with a median follow-up of 6 months. All patients were seen at 4-6 week intervals after photodynamic therapy was completed. Twenty patients received a total of 85 treatments in 34 cycles. A summary of individual treatment results is presented in Table 1. In terms of dose response, our data showed the following: of the 14 patients who received a total light dose under 50J/cm’, 6 had no response, 6 had a partial response, and 2 had a complete response; of the 2 patients who received a total light dose between 50 and lOOJ/cm*, both had a partial response; of the 4 patients who received a total light dose of more than lOOJ/cm’, 1 had no response, 1 had a partial response, and 2 had a complete
Fig. 2. The chest wall of the same patient 24 hr after photodynamic therapy ( 18 joules/cm’ in two fractions). Light fields are demarcated by lines. Tumor appears ecchymotic; little reaction is seen in adjacent normal tissue which received the same light dose.
444
I. J. Radiation Oncology 0 Biology 0 Physics
The reasons why non-responders or partial reresponse. sponders were not simply given higher doses included the following, in order of relative frequency: a) progression of disseminated disease, and b) prohibitive skin toxicity. The large variation in total light dose reflects the large variation in patients responses to similar treatment. The large variation in the minimum power densities reflects variation in field size. Overall, 4 of 20 (20%) achieved a complete response. 9 of 20 (45%) achieved a partial response (Fig. 3) and 7 of 20 (35%) achieved no response to therapy. The duration of complete response ranged from 1.5 to 14 months (average 7 months). The duration of partial response ranged from 1 to 4 months (average 2.3 months). Overall, the median and average duration of response were 1.5 and 2.5 months, respectively. One patient with complete response and one patient with a partial response to PDT died of disseminated disease with no evidence of local progression at 1.5 and 4 months, respectively. For the CR, PR, and NR groups, respectively, the average total dose was 140, 6 1, and 46 joules/cm2; the average number of treatments was 7, 4, and 3; the average number of cycles was 2, 2, and 1; the average dose per cycle was 12, 26, and 20 joules/cm2 (see Table 1). The average dose/fraction for the CR, PR, and NR groups were 2 1, 14, and 19 joules/cm2, respectively. The minimum power density for the CR group ranged from 3.9 to 80.0 mW/cm’ (average 29), for the PR group, 1.2 to 50 mW/cm2 (average 14) and for the NR group, 1.9 to 76 mW/cm’ (average 2 1). Complications included pain and erythema as well as ecchymoses, blistering, ulceration and necrosis in areas of skin grossly involved by tumor. The incidence and severity of these toxicities were related directly to total light dose but not related to prior treatment, dose/fraction, or power density. All patients developed a brisk erythema, usually within 10 minutes, involving the area covered by
Fig. 3. The chest wall of the same patient 31 weeks after completion of photodynamic therapy. The peripheral nodules have resolved and the central necrotic lesion, managed with silvadene dressings only, is healing well. This was classified as a partial response. Progression of this local disease occurred one month later in conjunction with clinical deterioration due to distant metastases.
July 1991, Volume 21, Number 2
the light field, that is, the entire chest wall on the involved side. Of interest, it was observed that when the power density exceeded 7 mW/cm2, the patients complained of a prickly heat sensation. Other complications occurred within the following time course: edema and ecchymoses within l-2 days (Fig. 2), blistering within 2-3 days, pain requiring narcotics within l-4 days, and ulceration and necrosis within 5 days. Nineteen of 20 patients developed moderate to severe discomfort in the treatment field. This was controlled in 17 of 19 with oral narcotics. The other two patients required admission to hospital for 48 hr for i.v. or i.m. medication. One patient denied pain despite a marked erythematous reaction. Superficial necrosis of tumor that could be managed with silver sulfadiazine and sterile dressings was encountered in 8/20 patients. One patient required surgical debridement for a focal area of skin necrosis and one patient developed a 5 X 5 cm area of full thickness necrosis of heavily tumor-infiltrated skin that required surgical debridement and skin flap reconstruction. This was a patient treated with a single fraction of light 35 J/cm2 to a chest wall twice previously irradiated to a total cumulative dose of 56 Gy. The resected area was histologically free of tumor. There were no significant differences in the complications observed when patients were grouped by response (CR, PR, NR). DISCUSSION Chest wall recurrence of breast cancer is a distressing problem for both patient and physician. This study illustrates both the potential and the difficulties of photodynamic therapy when conventional modes of therapy (surgery, irradiation, chemotherapy, and hormonal therapy) fail to control local disease. These patients were treated as part of a Phase I study of photodynamic therapy for surface malignancies, designed to define dose, response, and toxicity parameters in selected anatomic sites, including skin. Although we achieved 20% complete response rate, response duration was short. Thirty-five percent showed no response. One limitation of this treatment is skin toxicity; 45% of our patients showed some necrosis. This was seen, however, only in tumor-infiltrated tissue, not the surrounding normal skin. Furthermore, even the necrotic areas healed well with conservative management as detailed above, except for the one patient who required surgical debridement and skin flap reconstruction. Another limitation of successful PDT to the chest wall appears to be the inadequate penetration of light beyond about 5 mm. It appears that superficial tumor can be eradicated but tumor cells infiltrating beyond 5 mm survive and cause disease recurrence after photodynamic therapy (see Fig. 4). Application of light at a longer wavelength would facilitate deeper tissue penetration, but the photosensitizer used in this study, dihematoporphyrin ethers (DHE), is not excitable at such longer wavelengths. In fact, hematoporphyrin excitability peaks around a wavelength of 400 nm and shows only moderate excitability at 630 nm. Vis-
Chest wall PDT 0 P. W.
Fig. 4. An example of tissue fluorescence 48 hr after injection of the hematoporphyrin derivative in patient #9. The pattern of fluorescence is consonant with the clinical disease as demonstrated by comparison to a simultaneous photograph, Figure 1.
ible red light at a wavelength of 630 nm was used in this study as a compromise; 630 nm light penetrates deeper than 400 nm light but does not take full advantage of the peak excitability of the photosensitizer. Current research is seeking photosensitizers, such as metallothallocyanines, chlorines, and other porphyrins, with excitable ranges at longer wavelengths. The anatomic characteristics of the recurrent disease on the chest wall also appear to influence the likelihood that photodynamic therapy will yield worthwhile tumor regression. In our patients, recurrent disease often involved the dermal lymphatics in association with an erythematous, inflammatory involvement of the skin of the chest wall. After PDT, they often recurred both within and adjacent to our already generous light fields (all visible disease). This suggests the presence of occult microscopic disease both deep to and outside the range of the light used. The size of the area involved will influence the total dose and therefore tolerance of that area. Small tumor nodules (less than 1.5 cm in diameter) with grossly intact skin surrounding them can be treated to very high light doses with less concern about the risk of full thickness necrosis of the tumor nodule. Such small focal areas of necrosis will re-epithelialize from the adjacent, grossly uninvolved skin. Many of the patients in this series, however, had large areas of the chest wall grossly involved by tumor. Distinct from ionizing radiation where such areas will promptly re-epithelialize, our experience with PDT is that areas of skin visibly involved by tumor will display the characteristics of second degree burns. We observed vesiculation and desquamation followed by re-epithelialization over several weeks, despite meticulous care. Such reactions occurred after single fractions of light as small as 9 joules/cm*. Doses up to 30 joules/cm2 appear to yield a transient complete response. Disease, however, promptly recurs, presumably from repopulation from the unaffected deep margin. Attempts to treat to higher light doses with 630 nm red light appear to cause full thickness necrosis and a consequent loss of potential for spontaneous re-
SPERDUTOet al.
445
epithelialization. Furthermore, such efforts produce only a marginal increase in dose at the deep margin because of the exponential fall off in light dose. In the future, this problem may be obviated by the use of other sensitizers, excitable at longer wavelengths and therefore capable of deeper penetration. In this series, we were unwilling to subject our patients to the attendent morbidity (necrosis, weeks of burn care, and possible plastic surgery) of higher doses, especially when the deep margin was suspect or when there was disseminated disease. Another limitation to PDT is the variable retention of the photosensitizer in comparable tissues. This is most manifest by the patient (Table 1, No. 8) who developed a full thickness necrosis in only part of the treatment field. The mechanism of DHE distribution and retention in tumor are not well characterized. There is no currently available method to determine accurately the DHE distribution or concentration in tissue. Although the tissue fluorescence pattern after hematoporphyrin injection is consonant with the pattern of gross disease (compare Figs. 1 and 5) it remains uncertain whether tumor fluorescence accurately reflects DHE concentration. As Schuh et al. note, fluorescence may simply reflect DHE in the blood and interstitial spaces and not necessarily the intracellular concentration. This may explain the variable response in comparable tissues. Similarly, there is no method at present to assess singlet oxygen, a more direct measure of tumoricidal potential. Clearly, oxygen delivery is essential to yield singlet oxygen and a photodynamic reaction. The relationship between tumor response and tumor oxygenation and blood flow remains to be characterized. All of the above need further illumination before optimal treatment parameters can be confirmed. The average total dose increased linearly with the degree
Fig. 5. An MRI of the chest demonstrates one of the limitations of photodynamic therapy; the tumor involvement of the left chest wall delineated by the bright signal extends approximately 4 cm in depth. The light currently used (630 nm wavelength) can only penetrate 0.5 to 1.O cm. Because of the depth of chest wall involvement and the parenchymal nodules seen in the right lung, this patient was not treated with photodynamic therapy.
446
I. J. Radiation Oncology 0 Biology 0 Physics
of response; however, the multiple variables discussed above preclude the simplistic deduction that more dose would yield better response. The dose/fraction and the power density do not show a clear relationship to response. The power density is inversely related to the size of the field. The power density could theoretically limit the therapeutic effect if it was inadequate to activate a photoreaction. Although power density has only been shown to be a limiting factor at power densities below 0.3 mW/ cm2 (12), these low power densities may prevail at the deep margins of the treatment field. Accordingly, we have listed the minimum, not average, power density in Table 1. On the other hand, there was no evidence to suggest that high power density led to more severe complications.
July 1991, Volume 21. Number 2 It would be of great value to be able to identify patients a priori who may achieve complete responses. Toward this end, we reviewed the pre-treatment characteristics common among our subset of complete response patients and found only that all were post-menopausal and 2 of 5 had not received prior chemotherapy. Extent of disease did not distinguish this group. Photodynamic therapy has the potential to be effective local treatment for recurrent breast cancer, but not until a sensitizer that absorbs light at longer, deeper penetrating wavelengths is used will durable complete responses be seen. Future work must address the above mentioned problems and seek to expand the spectrum of clinical applications.
REFERENCES 1. Dougherty, T. J. Hematoporphyrin derivative for detection and treatment of cancer. J. Surg. Oncol. 15:209-2 10; 1980. 2. Dougherty, T. J. Photosensitization of malignant tumors. Sem. Surg. Oncol. 2:24-37; 1986. 3. Dougherty, T. J.; Lawrence, G.; Kaufman, J.; Boyle, D.: Weishaupt, K.; Goldfarb, A. Photoradiation in the Treatment of Recurrent Breast Carcinoma. JNCI 62:231-237; 1979. 4. Dougherty, T. J.; Weishaupt, K. R.; Boyle, D. G. Photodynamic sensitizers. In: DeVita, V. T., Jr., Hellman, S., Rosenberg, S., eds. Cancer: principles and practice of oncology, Vol. 2, 2nd edition. Philadelphia: Lippincott; 1985:22722279. 5. Edell, E. S. Bronchoscopic phototherapy with hematoporphyrin derivative for treatment of localized bronchogenic carcinoma: a S-year experience. Mayo Clin. Proc. 62:8- 14; 1987. 6. Gomer, C. J.; Dougherty, T. J. Determination of (3H) and (14C) hematoporphyrin derivative distribution in malignant and normal tissue. Cancer Res. 39: 146- 15 1; 1979. 7. Hall, E. J. Chemical and pharmacologic modifiers. In: Hall, E. J., ed. Radiobiology for the radiologist. Hagerstown: Harper&Row; 1978:171-194. 8. Kaye, A. H.; Morstyn, G.; Brownbill, D. Adjuvant highdose photoradiation therapy in the treatment of cerebral glioma; a phase l-2 study. J. Neurosurg. 67:500-505; 1987. 9. Kessel, D.; Cheng, M. On the preparation of dihematoporphyrin ether, the tumor-localizing component of HpD. Photochem. Photobio. 41:277-282; 1985. 10. Moan, J. Porphyrin photosensitization and phototherapy. Photochem. Photobio. 43:681-690; 1986.
11. Pass, H. I.; DeLaney, T. F.; Smith, P. D.; Bonner, R.; Russo, A. Bronchoscopic phototherapy at comparable dose rates: Early results. Ann. Thoracic Surg. 47:693-99; 1989. 12. Matthews, W.; Cook, J.; Mitchell, J.; Perry, R.; Evans, S.; Pass, H. Photodynamic therapy with dihematoporphyrin ether (DHE) exhibits dose-rate effects. Cancer Research 49: 718-721: 1989. 13. Prout, G. R.; Lin, C. W.; Benson, R.; Nseyo, J. J.; Griffin, P. P.; Kinsey, J.; Tian, M.; Lao, Chen, X.; Ren, F.; Qiao, S. Photodynamic hematoporphyrin derivative in the treatment transitional cell carcinoma of the bladder. N. 531733-736: 1987.
U. 0.; Daly, Y.; Mian, Y.; therapy with of superficial Engl. J. Med.
14. Raab, 0. Uber die Wirkung fluorescirender fusorien: Z. Biol. 39:524-546; 1900.
Stoffe auf In-
15. Rettenmaier, M. A.; Berman, M. L.; DiSaia, P. J.; Burns, R. G.; Berns, M. W. Photoradiation therapy of gynecologic malignancies. Gynecol. Oncol. 17:200-206; 1984. 16. Schuh, M.; Nseyo, U. 0.; Potter, W. R.; Dao, T. L.; Dougherty, T. J. Photodynamic therapy for palliation of locally recurrent breast carcinoma. J. Clin. Oncol. 5:1766-1770; 1987. 17. Svaasand, 1. 0.; Dorran, D. R.; Dougherty, T. J. Temperature rise during photoradiation therapy of malignant tumors. Med. Phys. 1O:lO; 1983. 18. Tochner, Z.; Mitchell, J. B.; Smith, P.; Harrington, F.; Glatstein, E.; Russo, D.; Russo, A. Photodynamic therapy of ascites tumours within the peritoneal cavity. Br. J. Cancer 53:733-736; 1986.
$3 00 + .lc
l Phase I/II Clinical Trials
PHOTODYNAMIC
THERAPY FOR CHEST WALL RECURRENCE IN BREAST CANCER
PAUL W. SPERDUTO, M.D., THOMAS F. DELANEY, M.D., GUNTER THOMAS, PAUL SMITH, PH.D., LAURA J. DACHOWSKI, R.N., ANGELO Russo, M.D., PH.D., ROBERT BONNER, PH.D. AND ELI GLATSTEIN, M.D. RadiationOncology Branch, Clinical Oncology Program, Division of Cancer Treatment, National
Cancer Institute,
National
Institutes
of Health,
Bethesda,
MD 20892
Photodynamic therapy is the use of a sensitizer (dihematoporphyrin ethers) which is preferentially retained in tumor cells and activated by subsequent light delivery resulting in a selective tumoricidal effect. Between1986 and 1989, we treated 20 patients with photodynamic therapy for chest wall recurrence of breast cancer. Responses were seen (20% complete response, 45% partial response, 35% no response), but the duration of response was short (average 2.5 months). Complications, in decreasing frequency, included pain, ecchymoses, blistering, ulceration and necrosis in the area of tumor involvement on the chest wall. One patient required skin flap reconstruction for full thickness necrosis. A limitation to this mode of therapy is that the sensitizer currently used is activated by light at a wavelength of 630 nm. This light can penetrate to a tissue depth of only 0.5 to 1.0 cm; thus, deeper disease cannot be treated. Future research must focus on the development of a clinically useful photosensitizer that can be activated by light at longer wavelengths and thereby achieve deeper tissue penetration. This would greatly expand the patient population for which this therapy is useful.
Hematoporphyrinderivative,Breast cancer,Laser, Photosensitizer. INTRODUCTION Much research has focused on the development of cancer treatment modalities that selectively destroy tumor, without disruption of normal cell and tissue function (7). Light-activated photosensitizers demonstrate relatively selective tumoricidal effects. Many believe this is due to the photosensitizer being retained longer in higher concentrations in neoplastic tissue than in most normal tissues (6). Subsequent light activation preferentially destroys sensitized tumor cells and not the relatively unsensitized surrounding normal tissues. The use of light-activated chemicals in biologic systems dates back to 1900 when Raab reported the lethal effect of light on paramecium treated with an acridine dye ( 14). When given alone, neither the light nor the dye were cytotoxic. Since then, a variety of sensitizers have been investigated but the focus of most research has been on the porphyrins. The mechanism of cell death is the oxidation of vital cellular structures, postulated to be critical mitochondrial or plasma membrane targets. This cytotoxicity is thought to be mediated by singlet oxygen, a reactive oxygen species generated by the interaction of ground state
Reprint requests to: Dr. Sperduto, Rm B3-B69, Bethesda, MD 20892.
NIH, NCI-ROB,
molecular oxygen with photosensitizer which has been activated by the appropriate wavelength visible light (1, 4, 10). Hematoporphyrin derivative (HpD) or the more active preparation of dihematoporphyrin ethers (DHE) have been used for tumor detection and in the treatment of tumors of skin (2), bladder ( 13) brain (8) and bronchus (5, 1 I), as well as gynecologic cancers (15), peritoneal tumors ( 18), and locally recurrent breast carcinoma (3). 630 nm wavelength red light from a laser source has generally been used in the clinic to activate the photosensitizer because of its relatively favorable tissue penetration compared to other wavelengths which may also activate the sensitizer ( 1, 4, 10). Schuh et al. reported 14 patients who received 30 treatments of photodynamic therapy (PDT) for locally recurrent breast carcinoma: 2 of 30 (7%) had a complete response (CR), 22 of 30 (73%) had a partial response (PR), and 4 of 30 ( 13%) had no response (NR) ( 16). The duration of response varied from 6 weeks to 8 months. The purpose of this communication is to report our experience at the National Cancer Institute with PDT for locally recurrent breast carcinoma.
Accepted
Bldg. 10,
441
for publication
25 January
199 1.
1. J. Radiation Oncology 0 Biology 0 Physics
442 METHODS
AND
MATERIALS
Patients Between September 1986 and January 1989,20 women with histologically proven recurrent breast cancer of the chest wall were treated with photodynamic therapy at our institution as part of a Phase I study. All patients had failed conventional therapy (surgery in 17, radiation in 19, and chemotherapy in 17). Table 1 presents the patient characteristics. All had multiple cutaneous and subcutaneous nodules, ranging in size from 2 mm to confluent tumor-infiltrated rashes spanning more than 5 cm (Fig. 1). Patients were selected for treatment on the protocol if grossly visible or imageable disease could be adequately treated by the known 0.5- 1.O cm light penetration of the 630 nm external light beam used. Patients had to have adequate hepatic function (normal or common toxicity criteria grade 1 elevation of liver function tests) because of the hepatic clearance of the photosensitizer. At the time of the first PDT treatment, 8 patients had local chest wall disease only and 12 had distant metastases in addition to chest wall disease.
Pt
Age
Extent of disease
Prior treatment
parameters
HpD dose w/kg
and response
Number of treatments
Fig. 1. The chest wall of a patient (#9) prior to photodynamic therapy with multiple cutaneous nodules and a confluent biopsyproven tumor-infiltrated rash.
planation of the treatment and possible side-effects. They were specifically counseled about photosensitivity and the need to exercise caution, or avoid whenever possible, direct sunlight and bright lights for 6 weeks after therapy. Photosensitizer The photosensitizer used in the trial* was a preparation of dihematoporphyrin ethers (DHE) containing a higher concentration of the photosensitizing component of the hematoporphyrin derivative (HpD) (9) formulated as a
Informed consent Informed consent in conformity with guidelines established by the Institutional Review Board and the NIH Clinical Center was obtained from all patients after ex-
Table 1. Treatment
July 1991, Volume 21, Number 2
to photodynamic
Number of cycles
Total light dose J/cm’
therapy
Power density mW/cm2
for chest wall recurrence
Postinjection interval days
Response
Duration of response months
1
57
DM
mc
1.5
2
1
20
80
2. 3
CR
1.5
2 3 4 5 6 7 8
58 45 46 59 73 56 13
L DM L DM DM DM DM
mrch
1.5 1.5
2 2
1 1 1 1
1
40 24 30 30 35 35 27
33 17 39 18 76 24 50
3. 4 3. 9 2 2. 3 2 3 2. 3. 4
PR NR NR PR NR CR PR
1.0 0 0 2.0 0 14 4.0
9
40
DM
10 11 12
55 72 44
13 14 15 16 17 18 19 20
ITWC
of breast cancer
Complications e. e. e. e. e. e. e. e.
P P. p. P. P, P, P. P.
b b. u n u, n b, n n n
K
1.5
1
mrch
1.5 1.5 1.5 1.5
2 1 1 3
IllIT
1.5 .75
6
3
42
I
2, 3. 4
PR
4.0
e. P, n
DM L L
mrh mrh lmrch
1.5 1.5 1.5
4 10 10
2 3 3
34 145 178
2 4 5
2. 3. 4 2. 3. 4 2. 3.4
NR CR PR
0 6.5 1.0
e. P e. p, b. n e. P. u. n
44 55 69 53 41
L L L DM DM
mrc mrch mrch mrc mrch
1.5 1.5 1.5 1.5 1.5
2 5 14 5 3
1 2 4 2 1
26 73 359 81 48
7 4 7 4 5
2. 3 2, 3.4 2. 3. 4 2, 3, 4 2, 3, 4
NR PR CR PR PR
0 3.0 6.3 2.0 1.5
e. e. e, e. e
39 16 58
DM L DM
IIlK
1.5 1.5 1.5
2 7 3
1 3 1
34 139 36
5 6 4
2. 4 2, 3, 4 2, 3, 4
PR NR NR
2.5 0 0
e. P e. P. n e, P
K
mrh mrch
lrch IIlK
I I
L = Local disease only; DM = distant metastases; m = mastectomy; c = chemotherapy; r = radiation therapy; h = hormonal therapy; = photofrin II drug dose in mg/kg; CR = complete response; PR = partial response; NR = no response; e = erythemia: p = pain; b = blistering;
* Photofiin@ II, Quadra British Columbia, Canada.
Logic
Technologies,
Vancouver,
Comments
Skin flap required
P P. u. n P, b, u P
I = lumpectomy; u = ulceration;
No pain
DHE dose n = necrosis.
Chest wall PDT 0 P. W. SPERDUTO rful.
sterile solution of 2.5 mg/ml dissolved in 0.9% NaCl solution. The drug was administered as an investigational agent under the IND held by the Division of Cancer Treatment, National Cancer Institute 48-72 hr prior to planned light delivery. The photosensitizer was given to the patients as outpatients at a dose of 1.5 mg/kg in 50 ml of normal saline by intravenous infusion over 15 min. Light administration Light was delivered with 600 micron optical fibers coupled to a Coherent Innova20W argon laser/Model 599 Dye Laser System, emitting up to 5.0 W at a wavelength of 630 f 2 nm. Surface illumination was provided via 600 micron flat cut quartz fibers with loops placed between the laser coupling and the tip to produce uniform power density across the treatment field. Table 1 presents the photodynamic therapy details in our patients. Each patient received DHE 1.5 mg/kg IV at time zero. Initially patients were treated with a single fraction of light 72 hr after drug administration. Rather than setting a doseescalation policy prior to initiation of this highly experimental protocol, we allowed our early experience to define later dose escalation. We reasoned that the best approach, extrapolating from experience with external beam irradiation, would be to treat the entire chest wall and sequentially escalate the dose as tolerated. It was found that a single dose of 35 joules/cm2 would take nearly an entire day to deliver and cause full thickness chest wall necrosis. Accordingly, in the interest of shortening treatment time and maximizing the opportunity for repair of sub-lethal damage in normal tissues, we then adopted a multiple fraction treatment program. We injected the DHE on day 0 and delivered light on days 2, 3, and 4. This approach was further justified because there is no reliable method of measuring the level of photosensitizer in the tissue or tumor. Highly variable levels are suggested by the variable responses to similar doses of light. Thereafter, the light dose was sequentially escalated by 3-5 joules/cm2/treatment in successive patients. The only deviations from this schema resulted from: a) patient intolerance (when debilitating chest wall discomfort occurred after one or two fractions, the third fraction was deleted or reduced), orb) inadequate treatment effect (when little effect was seen after two fractions, relative to other patients at similar intervals after treatment, the dose of the third fraction was increased and/or additional fractions were given, as detailed in Table 1). Whenever possible, it was attempted to deliver at least two cycles of treatment per patient with the second or subsequent cycles of PDT at monthly intervals (Table I). A cycle was defined as the period between an IV injection of the photosensitizer to the next IV injection of the photosensitizer, regardless of the number of light treatments following the photosensitizer. Light fields were designed to include all visible disease with at least 3 cm margins. These fields encompassed the ipsilateral chest wall, extending laterally to the mid-axillary line, superiorly to
443
above the clavicle, inferiorly several centimeters below the contralateral inframammary fold and medially to the midline (Fig. 2). The total light dose varied from 20 to 359 joules/cm2. The power density (light dose rate) varied inversely with field size and ranged from 1.2 to 80.0 mW/ cm2. At power densities less than approximately 100 mW/ cm*, there is minimal elevation in temperature and the tissue reaction is felt to represent a photodynamic reaction, not a thermal one ( 17). Photographs were obtained before and after each cycle and during follow-up (see Fig. 1). Punch biopsies (3 mm) were obtained before and at the 4-6 week follow-up visit in at least one site of disease. Because most of these patients had diffuse dermal involvement without measurable nodules, we have defined complete response as clinical and pathologic (biopsyproven) regression of all tumor in the treatment field. A partial response was defined as greater than 50% reduction of measurable nodules or a complete clinical regression with residual microscopic disease. A no response was defined as less than a 50% reduction, no change, or progression of disease. RESULTS Follow-up in all patients extended beyond the time of local recurrence or until death. This period ranged from 1- 18 months with a median follow-up of 6 months. All patients were seen at 4-6 week intervals after photodynamic therapy was completed. Twenty patients received a total of 85 treatments in 34 cycles. A summary of individual treatment results is presented in Table 1. In terms of dose response, our data showed the following: of the 14 patients who received a total light dose under 50J/cm’, 6 had no response, 6 had a partial response, and 2 had a complete response; of the 2 patients who received a total light dose between 50 and lOOJ/cm*, both had a partial response; of the 4 patients who received a total light dose of more than lOOJ/cm’, 1 had no response, 1 had a partial response, and 2 had a complete
Fig. 2. The chest wall of the same patient 24 hr after photodynamic therapy ( 18 joules/cm’ in two fractions). Light fields are demarcated by lines. Tumor appears ecchymotic; little reaction is seen in adjacent normal tissue which received the same light dose.
444
I. J. Radiation Oncology 0 Biology 0 Physics
The reasons why non-responders or partial reresponse. sponders were not simply given higher doses included the following, in order of relative frequency: a) progression of disseminated disease, and b) prohibitive skin toxicity. The large variation in total light dose reflects the large variation in patients responses to similar treatment. The large variation in the minimum power densities reflects variation in field size. Overall, 4 of 20 (20%) achieved a complete response. 9 of 20 (45%) achieved a partial response (Fig. 3) and 7 of 20 (35%) achieved no response to therapy. The duration of complete response ranged from 1.5 to 14 months (average 7 months). The duration of partial response ranged from 1 to 4 months (average 2.3 months). Overall, the median and average duration of response were 1.5 and 2.5 months, respectively. One patient with complete response and one patient with a partial response to PDT died of disseminated disease with no evidence of local progression at 1.5 and 4 months, respectively. For the CR, PR, and NR groups, respectively, the average total dose was 140, 6 1, and 46 joules/cm2; the average number of treatments was 7, 4, and 3; the average number of cycles was 2, 2, and 1; the average dose per cycle was 12, 26, and 20 joules/cm2 (see Table 1). The average dose/fraction for the CR, PR, and NR groups were 2 1, 14, and 19 joules/cm2, respectively. The minimum power density for the CR group ranged from 3.9 to 80.0 mW/cm’ (average 29), for the PR group, 1.2 to 50 mW/cm2 (average 14) and for the NR group, 1.9 to 76 mW/cm’ (average 2 1). Complications included pain and erythema as well as ecchymoses, blistering, ulceration and necrosis in areas of skin grossly involved by tumor. The incidence and severity of these toxicities were related directly to total light dose but not related to prior treatment, dose/fraction, or power density. All patients developed a brisk erythema, usually within 10 minutes, involving the area covered by
Fig. 3. The chest wall of the same patient 31 weeks after completion of photodynamic therapy. The peripheral nodules have resolved and the central necrotic lesion, managed with silvadene dressings only, is healing well. This was classified as a partial response. Progression of this local disease occurred one month later in conjunction with clinical deterioration due to distant metastases.
July 1991, Volume 21, Number 2
the light field, that is, the entire chest wall on the involved side. Of interest, it was observed that when the power density exceeded 7 mW/cm2, the patients complained of a prickly heat sensation. Other complications occurred within the following time course: edema and ecchymoses within l-2 days (Fig. 2), blistering within 2-3 days, pain requiring narcotics within l-4 days, and ulceration and necrosis within 5 days. Nineteen of 20 patients developed moderate to severe discomfort in the treatment field. This was controlled in 17 of 19 with oral narcotics. The other two patients required admission to hospital for 48 hr for i.v. or i.m. medication. One patient denied pain despite a marked erythematous reaction. Superficial necrosis of tumor that could be managed with silver sulfadiazine and sterile dressings was encountered in 8/20 patients. One patient required surgical debridement for a focal area of skin necrosis and one patient developed a 5 X 5 cm area of full thickness necrosis of heavily tumor-infiltrated skin that required surgical debridement and skin flap reconstruction. This was a patient treated with a single fraction of light 35 J/cm2 to a chest wall twice previously irradiated to a total cumulative dose of 56 Gy. The resected area was histologically free of tumor. There were no significant differences in the complications observed when patients were grouped by response (CR, PR, NR). DISCUSSION Chest wall recurrence of breast cancer is a distressing problem for both patient and physician. This study illustrates both the potential and the difficulties of photodynamic therapy when conventional modes of therapy (surgery, irradiation, chemotherapy, and hormonal therapy) fail to control local disease. These patients were treated as part of a Phase I study of photodynamic therapy for surface malignancies, designed to define dose, response, and toxicity parameters in selected anatomic sites, including skin. Although we achieved 20% complete response rate, response duration was short. Thirty-five percent showed no response. One limitation of this treatment is skin toxicity; 45% of our patients showed some necrosis. This was seen, however, only in tumor-infiltrated tissue, not the surrounding normal skin. Furthermore, even the necrotic areas healed well with conservative management as detailed above, except for the one patient who required surgical debridement and skin flap reconstruction. Another limitation of successful PDT to the chest wall appears to be the inadequate penetration of light beyond about 5 mm. It appears that superficial tumor can be eradicated but tumor cells infiltrating beyond 5 mm survive and cause disease recurrence after photodynamic therapy (see Fig. 4). Application of light at a longer wavelength would facilitate deeper tissue penetration, but the photosensitizer used in this study, dihematoporphyrin ethers (DHE), is not excitable at such longer wavelengths. In fact, hematoporphyrin excitability peaks around a wavelength of 400 nm and shows only moderate excitability at 630 nm. Vis-
Chest wall PDT 0 P. W.
Fig. 4. An example of tissue fluorescence 48 hr after injection of the hematoporphyrin derivative in patient #9. The pattern of fluorescence is consonant with the clinical disease as demonstrated by comparison to a simultaneous photograph, Figure 1.
ible red light at a wavelength of 630 nm was used in this study as a compromise; 630 nm light penetrates deeper than 400 nm light but does not take full advantage of the peak excitability of the photosensitizer. Current research is seeking photosensitizers, such as metallothallocyanines, chlorines, and other porphyrins, with excitable ranges at longer wavelengths. The anatomic characteristics of the recurrent disease on the chest wall also appear to influence the likelihood that photodynamic therapy will yield worthwhile tumor regression. In our patients, recurrent disease often involved the dermal lymphatics in association with an erythematous, inflammatory involvement of the skin of the chest wall. After PDT, they often recurred both within and adjacent to our already generous light fields (all visible disease). This suggests the presence of occult microscopic disease both deep to and outside the range of the light used. The size of the area involved will influence the total dose and therefore tolerance of that area. Small tumor nodules (less than 1.5 cm in diameter) with grossly intact skin surrounding them can be treated to very high light doses with less concern about the risk of full thickness necrosis of the tumor nodule. Such small focal areas of necrosis will re-epithelialize from the adjacent, grossly uninvolved skin. Many of the patients in this series, however, had large areas of the chest wall grossly involved by tumor. Distinct from ionizing radiation where such areas will promptly re-epithelialize, our experience with PDT is that areas of skin visibly involved by tumor will display the characteristics of second degree burns. We observed vesiculation and desquamation followed by re-epithelialization over several weeks, despite meticulous care. Such reactions occurred after single fractions of light as small as 9 joules/cm*. Doses up to 30 joules/cm2 appear to yield a transient complete response. Disease, however, promptly recurs, presumably from repopulation from the unaffected deep margin. Attempts to treat to higher light doses with 630 nm red light appear to cause full thickness necrosis and a consequent loss of potential for spontaneous re-
SPERDUTOet al.
445
epithelialization. Furthermore, such efforts produce only a marginal increase in dose at the deep margin because of the exponential fall off in light dose. In the future, this problem may be obviated by the use of other sensitizers, excitable at longer wavelengths and therefore capable of deeper penetration. In this series, we were unwilling to subject our patients to the attendent morbidity (necrosis, weeks of burn care, and possible plastic surgery) of higher doses, especially when the deep margin was suspect or when there was disseminated disease. Another limitation to PDT is the variable retention of the photosensitizer in comparable tissues. This is most manifest by the patient (Table 1, No. 8) who developed a full thickness necrosis in only part of the treatment field. The mechanism of DHE distribution and retention in tumor are not well characterized. There is no currently available method to determine accurately the DHE distribution or concentration in tissue. Although the tissue fluorescence pattern after hematoporphyrin injection is consonant with the pattern of gross disease (compare Figs. 1 and 5) it remains uncertain whether tumor fluorescence accurately reflects DHE concentration. As Schuh et al. note, fluorescence may simply reflect DHE in the blood and interstitial spaces and not necessarily the intracellular concentration. This may explain the variable response in comparable tissues. Similarly, there is no method at present to assess singlet oxygen, a more direct measure of tumoricidal potential. Clearly, oxygen delivery is essential to yield singlet oxygen and a photodynamic reaction. The relationship between tumor response and tumor oxygenation and blood flow remains to be characterized. All of the above need further illumination before optimal treatment parameters can be confirmed. The average total dose increased linearly with the degree
Fig. 5. An MRI of the chest demonstrates one of the limitations of photodynamic therapy; the tumor involvement of the left chest wall delineated by the bright signal extends approximately 4 cm in depth. The light currently used (630 nm wavelength) can only penetrate 0.5 to 1.O cm. Because of the depth of chest wall involvement and the parenchymal nodules seen in the right lung, this patient was not treated with photodynamic therapy.
446
I. J. Radiation Oncology 0 Biology 0 Physics
of response; however, the multiple variables discussed above preclude the simplistic deduction that more dose would yield better response. The dose/fraction and the power density do not show a clear relationship to response. The power density is inversely related to the size of the field. The power density could theoretically limit the therapeutic effect if it was inadequate to activate a photoreaction. Although power density has only been shown to be a limiting factor at power densities below 0.3 mW/ cm2 (12), these low power densities may prevail at the deep margins of the treatment field. Accordingly, we have listed the minimum, not average, power density in Table 1. On the other hand, there was no evidence to suggest that high power density led to more severe complications.
July 1991, Volume 21. Number 2 It would be of great value to be able to identify patients a priori who may achieve complete responses. Toward this end, we reviewed the pre-treatment characteristics common among our subset of complete response patients and found only that all were post-menopausal and 2 of 5 had not received prior chemotherapy. Extent of disease did not distinguish this group. Photodynamic therapy has the potential to be effective local treatment for recurrent breast cancer, but not until a sensitizer that absorbs light at longer, deeper penetrating wavelengths is used will durable complete responses be seen. Future work must address the above mentioned problems and seek to expand the spectrum of clinical applications.
REFERENCES 1. Dougherty, T. J. Hematoporphyrin derivative for detection and treatment of cancer. J. Surg. Oncol. 15:209-2 10; 1980. 2. Dougherty, T. J. Photosensitization of malignant tumors. Sem. Surg. Oncol. 2:24-37; 1986. 3. Dougherty, T. J.; Lawrence, G.; Kaufman, J.; Boyle, D.: Weishaupt, K.; Goldfarb, A. Photoradiation in the Treatment of Recurrent Breast Carcinoma. JNCI 62:231-237; 1979. 4. Dougherty, T. J.; Weishaupt, K. R.; Boyle, D. G. Photodynamic sensitizers. In: DeVita, V. T., Jr., Hellman, S., Rosenberg, S., eds. Cancer: principles and practice of oncology, Vol. 2, 2nd edition. Philadelphia: Lippincott; 1985:22722279. 5. Edell, E. S. Bronchoscopic phototherapy with hematoporphyrin derivative for treatment of localized bronchogenic carcinoma: a S-year experience. Mayo Clin. Proc. 62:8- 14; 1987. 6. Gomer, C. J.; Dougherty, T. J. Determination of (3H) and (14C) hematoporphyrin derivative distribution in malignant and normal tissue. Cancer Res. 39: 146- 15 1; 1979. 7. Hall, E. J. Chemical and pharmacologic modifiers. In: Hall, E. J., ed. Radiobiology for the radiologist. Hagerstown: Harper&Row; 1978:171-194. 8. Kaye, A. H.; Morstyn, G.; Brownbill, D. Adjuvant highdose photoradiation therapy in the treatment of cerebral glioma; a phase l-2 study. J. Neurosurg. 67:500-505; 1987. 9. Kessel, D.; Cheng, M. On the preparation of dihematoporphyrin ether, the tumor-localizing component of HpD. Photochem. Photobio. 41:277-282; 1985. 10. Moan, J. Porphyrin photosensitization and phototherapy. Photochem. Photobio. 43:681-690; 1986.
11. Pass, H. I.; DeLaney, T. F.; Smith, P. D.; Bonner, R.; Russo, A. Bronchoscopic phototherapy at comparable dose rates: Early results. Ann. Thoracic Surg. 47:693-99; 1989. 12. Matthews, W.; Cook, J.; Mitchell, J.; Perry, R.; Evans, S.; Pass, H. Photodynamic therapy with dihematoporphyrin ether (DHE) exhibits dose-rate effects. Cancer Research 49: 718-721: 1989. 13. Prout, G. R.; Lin, C. W.; Benson, R.; Nseyo, J. J.; Griffin, P. P.; Kinsey, J.; Tian, M.; Lao, Chen, X.; Ren, F.; Qiao, S. Photodynamic hematoporphyrin derivative in the treatment transitional cell carcinoma of the bladder. N. 531733-736: 1987.
U. 0.; Daly, Y.; Mian, Y.; therapy with of superficial Engl. J. Med.
14. Raab, 0. Uber die Wirkung fluorescirender fusorien: Z. Biol. 39:524-546; 1900.
Stoffe auf In-
15. Rettenmaier, M. A.; Berman, M. L.; DiSaia, P. J.; Burns, R. G.; Berns, M. W. Photoradiation therapy of gynecologic malignancies. Gynecol. Oncol. 17:200-206; 1984. 16. Schuh, M.; Nseyo, U. 0.; Potter, W. R.; Dao, T. L.; Dougherty, T. J. Photodynamic therapy for palliation of locally recurrent breast carcinoma. J. Clin. Oncol. 5:1766-1770; 1987. 17. Svaasand, 1. 0.; Dorran, D. R.; Dougherty, T. J. Temperature rise during photoradiation therapy of malignant tumors. Med. Phys. 1O:lO; 1983. 18. Tochner, Z.; Mitchell, J. B.; Smith, P.; Harrington, F.; Glatstein, E.; Russo, D.; Russo, A. Photodynamic therapy of ascites tumours within the peritoneal cavity. Br. J. Cancer 53:733-736; 1986.
