| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Cancer Therapy: Clinical |
Authors' Affiliations: 1 Departments of Radiation Oncology, 2 Surgery, 3 Biostatistics and Epidemiology, 4 Physics and Astronomy and 5 Division of Gynecologic Oncology, Department of Obstetrics and Gynecology, University of Pennsylvania, Philadelphia, Pennsylvania
Requests for reprints: Stephen M. Hahn, 3400 Spruce Street, 2 Donner, Philadelphia, PA 19104-4283. Phone: 215-662-7296; Fax: 215-349-5445; E-mail: hahn{at}xrt.upenn.edu.
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
Experimental Design: Patients received Photofrin 2.5 mg/kg i.v. 48 hours before debulking surgery. Intraoperative laser light was delivered to the peritoneal surfaces of the abdomen and pelvis. The outcomes of interest were (a) complete response, (b) failure-free survival time, and (c) overall survival time. Photosensitizer levels in tumor and normal tissues were measured.
Results: One hundred patients were enrolled into one of three strata (33 ovarian, 37 gastrointestinal, and 30 sarcoma). Twenty-nine patients did not receive light treatment. All 100 patients had progressed by the time of statistical analysis. The median failure-free survival and overall survival by strata were ovarian, 2.1 and 20.1 months; gastrointestinal cancers, 1.8 and 11.1 months; sarcoma, 3.7 and 21.9 months. Substantial fluid shifts were observed postoperatively, and the major toxicities were related to volume overload. Two patients died in the immediate postoperative period from bleeding, sepsis, adult respiratory distress syndrome, and cardiac ischemia.
Conclusions: Intraperitoneal Photofrin-mediated photodynamic therapy is feasible but does not lead to significant objective complete responses or long-term tumor control. Heterogeneity in photosensitizer uptake and tumor oxygenation, lack of tumor specificity for photosensitizer uptake, and the heterogeneity in tissue optical properties may account for the lack of efficacy observed.
Despite a large body of data exploring the basic biology of photodynamic therapy in animal models, photodynamic therapy has been used infrequently in the intraoperative setting. Early phase clinical studies are important both to determine whether the therapy shows sufficient efficacy to proceed to randomized trials, as well as to determine whether the results from animal models have relevance in a clinical setting. A phase I study of surgery in combination with photodynamic therapy with laser light and Photofrin (Axcan Pharma, Birmingham, AL) was conducted by the Surgery and Radiation Oncology Branches of the National Cancer Institute for disseminated i.p. malignancies (4, 5). Fifty-four patients with peritoneal carcinomatosis who underwent debulking surgery were treated on the study. The photodynamic therapy dose was sequentially escalated by increasing the sensitizer dose from 1.5 to 2.5 mg/kg, by shortening the interval between sensitizer injection and the surgery, and by increasing the light dose. With the use of 630 nm red light, photodynamic therapy induced small bowel edema and resulted in three small bowel perforations. Because of this transmural penetration by red light, less penetrating 514-nm green light was used for the large field exposures to the bowel and mesentery thereafter. This allowed further light dose escalation, and no additional small bowel complications were seen. Dose-limiting toxicity was encountered in two of three patients at the highest dose (5.0 J/cm2) of green light with boost. These two patients both developed pleural effusions that required thoracentesis and postoperative respiratory support for 7 to 9 days. The maximally tolerated dose of photodynamic therapy 48 hours after i.v. administration of Photofrin (2.5 mg/kg) was determined to be 3.75 J/cm2 of 514-nm green light to the entire peritoneal surface with boosts 5.0 to 7.5 J/cm2 of green light or 10 to 15 J/cm2 of red light to areas of gross disease. Subsequent modifications to the light doses have been made since the initial report was published.
The follow-up of these patients has been updated (5). An overall peritoneal cytologic response rate of 76% was reported in evaluable patients. More importantly, long-term disease-free survivors were reported in the subgroup of patients with ovarian cancer. Three of 25 ovarian cancer patients were disease-free >36 months after treatment. These results are especially impressive when one considers that this was a phase I trial. It should also be emphasized that many of the patients received what in retrospect was an inadequate light dose. Based upon the encouraging results of the phase I study and the lack of curative therapy for patients with peritoneal carcinomatosis and sarcomatosis, we initiated a phase II study of Photofrin-mediated i.p. photodynamic therapy.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
48 hours before the planned laparotomy. Surgical resection and light delivery were done as described below. The protocol permitted light administration to any patient whose peritoneal disease was resected to a thickness of 
5 mm. Toxicity was scored using the National Cancer Institute Cooperative Group Common Toxicity Criteria, version 1.0. Operative mortality was defined as death within 30 days of patients leaving the hospital. The patients were seen by the investigators or primary oncologists every 3 months after surgery until disease progression. Patient selection. Patients were selected based upon the following eligibility criteria: biopsy-proven peritoneal carcinomatosis or sarcomatosis who had exhausted curative therapies, age > 18 years, signed informed consent, and absence of distant metastatic disease. Patients were entered into one of three strata: ovarian cancer, sarcoma, and gastrointestinal cancers.
Patients were excluded based on the following criteria: borderline tumors of low malignant potential; ulcerative colitis; regional enteritis; inability to tolerate general anesthesia; HIV positivity; white count <2,000 per mm3 or platelet count <100,000 per mm3; serum creatinine 
2.5 mg/dL; severe liver disease, including cirrhosis, grade 3 to 4 elevations in liver function studies, or bilirubin in excess of 1.5 mg/dL, and pregnant or lactating patients. Patients who in the opinion of the attending surgeon could not be debulked to a thickness of 5 mm on preoperative evaluation were also excluded.
Patients underwent a preoperative evaluation that included a history and physical examination, imaging studies of the abdomen and pelvis (computerized tomography and/or magnetic imaging resonance), a chest radiograph and laboratory studies including an HIV antibody test. Other tests, including tests to exclude distant metastases, were done as clinically indicated.
The protocol was conducted under an investigator-sponsored Investigational New Drug Application with the U.S. Food and Drug Administration. This study was done in accordance with the Declaration of Helsinki and had approval from the Institutional Review Board of the University of Pennsylvania and the University of Pennsylvania Clinical Trials Scientific Review and Monitoring Committee. The trial opened in February 1997 and closed in March 2004.
Surgical procedure. Surgery was done at the Hospital of the University of Pennsylvania. The operating room lights and the surgeons' headlamps were covered with amber filter paper (Roscoe, Inc., Hollywood, CA) to reduce unwanted activation of the photosensitizer. The pulse oximeter, which uses a red light probe capable of activating the photosensitizer, was rotated to a different finger every 15 to 30 minutes to avoid burns to the nail bed. A laparotomy was done on all patients under the direction of the attending surgeon. The goal was to have all tumor resected to a thickness of 5 mm at the conclusion of the surgical resection. Normal tissues were resected only as necessary to debulk the tumor. The abdominal cavity was irrigated, hemostasis was achieved, and the light delivery portion of the procedure was then initiated.
Intraoperative photodynamic therapy. The light delivery and dosimetry system were similar to the technique previously described (6). Light doses were monitored with fixed peritoneal diodes located in the right upper quadrant, left upper quadrant, left peritoneal gutter, right peritoneal gutter, and in the midline pelvis anteriorly. One to two mobile photodiodes were also employed to measure dose in other regions. Laser light was generated using the KTP/532 Laser System manufactured by Laserscope, Inc. (San Jose, CA). The KTP laser pumped a model 630 XP Dye Module manufactured by Laserscope. The power density of the light did not exceed, on average, 150 mW/cm2. Eye protection in the form of goggles was used to attenuate the wavelength used to 10–4 of incident intensity.
The light doses were determined previously in a phase I trial (4, 5). The mesentery, the large bowel, and the small bowel were treated with 532-nm green light at a dose of 2.5 J/cm2 using a flat-cut optical fiber. The bowel was positioned under a circular beam of light and was treated in segments sequentially. After delivery of green light to the bowel and before 630-nm light delivery to the rest of the peritoneal surfaces with 630-nm red light, the abdomen was filled with a dilute solution of intralipid (0.01%) in sterile normal saline. This was used to enhance scattering of the light. The 630-nm light to the peritoneal cavity was delivered with an optical fiber sheathed within a modified endotracheal tube. The balloon cuff was inflated and filled with a 0.1% intralipid solution. The stomach received 5.0 J/cm2; the diaphragms, liver, and spleen received 7.5 J/cm2, whereas the pelvis and abdominal gutters received 10 J/cm2. Sites of gross disease on the right diaphragm, on the abdominal gutters, and in the pelvis were treated with a boost treatment up to 15 J/cm2. Patients received one course of light therapy at the time of surgery. Following the completion of light administration, the sterile photodiodes were removed from the abdomen. In the immediate postoperative period and after discharge, patients were instructed to avoid direct sunlight for 30 to 60 days after Photofrin administration.
Patients were transferred to the intensive care unit immediately after surgery and were, in general, intubated for at least 24 hours after the procedure. Significant fluid shifts were observed postoperatively, which necessitated massive fluid resuscitation as previously described (7).
Patient follow-up. Patients were seen 1 month after their discharge either in the outpatient clinics at the Hospital of the University of Pennsylvania or by the primary referring physician to assess treatment-related toxicities. The patients were then seen 3 months after the procedure and every 3 months for the first 24 months. A computerized tomography scan of the abdomen and pelvis was done every 3 months during the first year after treatment, every 6 months the second year, and then as clinically indicated thereafter. Other tests were done as clinically indicated. At the 6-month follow-up visit, if the patient had no clinical evidence of disease, a minilaparotomy or a laparoscopy was done for pathologic restaging. Biopsies, if clinically indicated, were taken from all regions that could be safely exposed at the time of laparoscopy/minilaparotomy. A laparoscopy was done in five patients, and a minilaparotomy was done in seven patients.
Photosensitizer concentration in tissue samples. Portions of tumor and resected normal tissues were placed in specimen containers, protected from visible light, and frozen at –70°C. The assay for porfimer sodium quantitation was based on a previous report (8). Tissue samples were thawed to room temperature and weighed and, depending on the amount available, 10 to 50 mg of tissue was placed in a vial with 0.150 to 0.500 mL, respectively, of the tissue solubilizer Solvable (Packard, Meriden, CT). The vial was capped and heated at 50°C overnight (20 ± 2 hours) in the dark. The solution was cooled to room temperature, an equal volume of distilled water was added and, after thorough mixing, it was transferred to a quartz cuvette of 0.2 mL (10-mg samples) or 0.6 mL (50-mg samples) capacity. The fluorescence of the solubilized sample was measured by a spectrofluorometer (FluoroMax-3, Jobin Yvon, Inc., Edison, NJ) with 
ex of 405 nm and em of 627 nm. Porfimer sodium concentration in the tissue was calculated based on the increase in fluorescence resulting from the addition of a known amount of porfimer sodium to each sample after its initial reading. Triplicates of each sample were run.
Photofrin concentrations were determined in the tumors of 48 patients with 26 patients having multiple tumor tissues sampled from various sites. Within each stratum, we described these concentrations by pooling all of the measurements and determining the median and range (minimum and maximum concentration). For each patient with multiple measurements, we determined an overall mean Photofrin concentration, a SD, and a coefficient of variation percent for each patient [CV% = (SD / mean) x 100]. We summarized the coefficient of variations within each strata using the median and the range. Stratum-specific means of tumor Photofrin concentration were calculated using individual measurements for patients with exactly one measurement and from within-patient means for patients with multiple samples. For consistency with analyses to be reported later, the mean for each patient was calculated by first averaging across each site with multiple measurements and then forming an overall mean. Statistical analyses were done using the software package SPSS (SPSS, Inc., Chicago, IL) or R 2.0 (http://www.r-project.org).
Summary statistics for tumor to normal ratios of Photofrin in large bowel without mucosa and small bowel without mucosa were determined using only those patients with samples from tumor as well as normal bowel tissues. We first averaged Photofrin concentrations across all tumor samples and across all samples of either small or large bowel within a patient. We then formed a ratio for each patient and summarized the results using medians and ranges.
Statistical considerations. This study was designed to evaluate toxicity and efficacy of photodynamic therapy for i.p. malignancies. Patients were enrolled into one of three disease strata: ovarian cancer, gastrointestinal cancers, and sarcoma. Toxicities were graded using Common Toxicity Criteria, version 1.0. Response was coded as either complete response (no evidence of abdominal disease) or treatment failure (new or recurrent abdominal disease) based on laparoscopy and/or radiographic findings at 6 months after surgery. The secondary objectives were to estimate failure-free and overall survival within each disease stratum and to characterize tissue Photofrin concentrations. Failure-free survival was measured from study entry (date of Photofrin injection) to first documented progression, death due to any cause, or last patient contact. For patients who were found ineligible to receive photodynamic therapy at the time of surgery, failure-free survival was measured from date of Photofrin injection to date of surgery, which was 2 days. Overall survival was measured from study entry to death due to any cause or last patient contact in all patients.
Within each stratum, we used a two-stage design under the premise that a true complete response rate of 20% would indicate an active treatment, whereas a true complete response rate of 5% would indicate an inactive treatment. In the first stage, enrollment continued until 14 patients received Photofrin injection and photodynamic light therapy. If at least one complete response was observed, then enrollment continued in the second stage until an additional 21 patients had received Photofrin injection and photodynamic light therapy. After the second stage, if at least four complete responses were seen, then the treatment would be declared active. At the end of the trial, the probabilities of either accepting an inactive treatment or rejecting an active treatment were each <10%. The accrual goal was 105 patients (35 per strata) treated with Photofrin injection and photodynamic light therapy. Accrual of sarcoma patients was terminated in January 2002 due to slowed enrollment after imatinib mesylate was approved for gastrointestinal stromal tumors. In consultation with the coinvestigators and the study biostatisticians in March 2004, it was determined after a review of the response data and the accrual rates that accrual to the ovarian cancer and gastrointestinal cohorts would be terminated.
Failure-time analyses were based on follow-up data available as of January 2005. The primary analyses of patient characteristics, toxicity, complete response rate failure-free survival, and overall survival were based on the intent-to-treat population (i.e., all patients who received Photofrin injection, regardless of whether photodynamic light therapy was administered). Patients who did not receive photodynamic light therapy were excluded from the decision to terminate the study early in the two-stage design but were included in the primary analyses. A subset analysis was also done based on patients who received both drug and light. Median failure-free and overall survival were estimated using the Kaplan and Meier method with 95% confidence intervals for medians based on Greenwood's formula (9, 10). Statistical analyses were done using the software package SPSS (SPSS) or R 2.0 (http://www.r-project.org).
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
|
50 years. The vast majority of patients were Eastern Cooperative Oncology Group performance status 0 or 1. All ovarian cancer patients had undergone surgery and chemotherapy before photodynamic therapy. All sarcoma patients had undergone surgery, and some had received chemotherapy (10 of 30) and radiotherapy (4 of 30). The majority of gastrointestinal cancer patients (32 of 37) had undergone surgery, and most had also received chemotherapy (29 of 37). Only two gastrointestinal patients had previously undergone radiotherapy, whereas four patients had received no preoperative therapies.
|
|
|
|
|
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Photodynamic therapy is a cancer treatment that employs the use of a photosensitizing agent and laser light of a specific wavelength to activate the photosensitizer in the presence of oxygen. The effective depth of cellular damage of photodynamic therapy in tissue is dependent upon several factors, including photosensitizer concentration, wavelength of laser light, amount of tissue oxygen, and the optical properties of the underlying tissues. In general, photodynamic therapy is a superficial treatment with an effective depth of cellular damage for Photofrin-mediated photodynamic therapy of 2 to 5 mm (1, 2). The superficial nature of Photofrin-mediated photodynamic therapy makes this treatment a potentially ideal therapy for surface malignancies, such as cancers, which have spread to the serosal surfaces of the peritoneum and pleura. The limited penetration of photodynamic therapy tissue effect would theoretically limit the potential for damaging underlying critical organs. We have recently reported our experience with treating non–small cell lung cancer, which has spread to the pleural surface (11). That study showed that excellent local control could be achieved with acceptable toxicity using Photofrin-mediated photodynamic therapy. Furthermore, we showed that the median survival of these patients exceeded what would normally be expected in patients with pleural carcinomatosis. This study provides some evidence that photodynamic therapy might be effective as a treatment for patients with malignancies that have spread to serosal surfaces.
Patients with malignancies that spread to the peritoneum (including patients with ovarian cancer, sarcoma, and gastrointestinal cancers) have few to no curative treatment options. These patients present with abdominal pain, altered bowel function, ascites, and recurrent small bowel obstruction. Chemotherapy is often administered to these patients, especially those with ovarian cancer and gastrointestinal malignancies. The subset of sarcoma patients with gastrointestinal stromal tumors is usually treated with imatinib mesylate (12, 13). Whole abdominal radiotherapy is not a curative treatment option for patients with peritoneal carcinomatosis and sarcomatosis because the underlying abdominal organs limit the ability to deliver a curative dose of radiation. Radical surgical approaches have been attempted in these patients but are unlikely to sterilize the peritoneum.
The phase I trial of Photofrin-mediated photodynamic therapy completed at the National Cancer Institute (4, 5) showed that this procedure could be done with acceptable toxicities. Encouraging preliminary responses, especially in ovarian cancer patients, provided the rationale for performing this phase II study. Our results show that Photofrin-mediated photodynamic therapy is feasible after surgical debulking but does not lead to control of peritoneal carcinomatosis or sarcomatosis after surgical debulking. Some of our patients had clinical benefit from i.p. photodynamic therapy that did not qualify as a complete response. Reduction in the number of peritoneal nodules and the elimination of the need for abdominal paracentesis were observed in selected individuals but could not be scored as a complete response. It should be kept in mind that the criteria used for efficacy in this study were rigorous and required the complete absence of clinical, radiographic, and, in some cases, pathologic disease within the abdomen. This is a standard of efficacy that is higher than is typically used in studies of chemotherapy in this patient population. Notwithstanding the strict criteria used, complete responses to Photofrin-mediated photodynamic therapy were uncommon, and failure-free survival was short in all disease cohorts. The toxicity caused by surgical debulking and Photofrin-mediated photodynamic therapy was substantial, including a capillary leak syndrome (7, 14) observed in many patients in the immediate postoperative period. It is likely that this systemic toxicity of i.p. photodynamic therapy is a result of tissue damage to the serosal surfaces of the peritoneum similar to that observed in burn victims. The presence of this toxicity and the absence of a significant therapeutic response suggest that i.p. photodynamic therapy has a narrow therapeutic ratio.
One of the reported theoretical advantages of photodynamic therapy as a cancer treatment is that systemically administered photosensitizers in animal models have greater retention in cancers compared with selected normal tissues (2, 3, 15). These studies, however, compared the uptake of photosensitizers in tumor to that in normal tissues, such as muscle or skin, which are not particularly relevant to normal tissues at risk for toxicity from i.p. photodynamic therapy. If confirmed in human clinical trials, the selectivity of photosensitizers in tumor compared with relevant normal tissues would provide an enhanced therapeutic ratio for photodynamic therapy. The need for tumor selectivity of photosensitizer retention is particularly important when considering treatment of large surface areas, such as the peritoneum, where a large number of sensitive normal tissues are at risk for toxicity and where activation of a photosensitizer in large surface areas has been associated with systemic inflammatory response syndrome or capillary leak syndrome (16). The photosensitizer uptake measurements in patients on this trial show that absolute tissue levels in human tumors are similar to those described for tumors in murine models (17, 18). However, tumor-to-normal tissue (T/N) ratios for relevant normal tissues, such as the large bowel, were very modest, with median values ranging between 1.1 and 2.1, depending on whether tissue contained mucosa (data not shown). Furthermore, substantial heterogeneity was found among patients, as could be expected from the variability among absolute tumor levels of drug within disease strata. Such heterogeneity would be expected to limit the photodynamic therapy dose that could be achieved before reaching clinical toxicity and likely contributes to the disappointing clinical findings described above. In a separate clinical trial, Photofrin measurements in several patients with non–small cell lung cancer with pleural spread (11) also identified modest drug selectivity between tumor and relevant normal tissues (normal pleura and lung). T/N ratios within specific tissue types were consistent among the first three patients evaluated on this study, but it is too early to assess the degree of intrapatient and interpatient heterogeneity in tissue Photofrin uptake on this trial in non–small cell lung cancer. A more expansive report with detailed statistical analysis of tumor and normal tissue levels of Photofrin uptake in the patients of the i.p. photodynamic therapy trial is currently being prepared. Overall, these data highlight the importance of assessing tumor to normal tissue ratios in human tumors and relevant normal tissues as part of the clinical evaluation of photodynamic therapy.
The photosensitizer measurements described in this study and in our previous reports (19) collectively shed some light on the delivery of drugs to tumor nodules on the peritoneal surface and are potentially relevant to the delivery of chemotherapy drugs to peritoneal tumors. We have previously shown that Photofrin is present in tumor nodules as small as 1 mm in size (17, 19). In addition, a preliminary evaluation of photosensitizer uptake and oxygen levels in these patients also suggested substantial interpatient and intrapatient heterogeneity in Photofrin and oxygen levels in tumor (19). The photodynamic process is dependent upon the presence of adequate photosensitizer and oxygen (2). Either of these factors may limit the efficacy of the treatment. An additional factor of importance is that a major mechanism of action for Photofrin-mediated photodynamic therapy is a vascular effect, which may not be optimal in a clinical setting where small volume disease is treated after debulking surgery.
The peritoneum is a complex cavity with a surface area similar to the external body surface area (4). There are regions of the peritoneal cavity that are difficult to access for light delivery, such as the diaphragmatic surfaces and deep within the pelvis. We have also previously shown that there is heterogeneity in the optical properties in peritoneal light delivery and light dose absorbed in various regions within a patient as well as in the same region among different patients (6, 20). These data suggest, therefore, that individualization of light doses based upon measured optical properties and measurement of both scattered and incident light will be necessary to adequately control light dose deposition in tissues.
In summary, we have shown that i.p. Photofrin-mediated photodynamic therapy is feasible but does not lead to significant objective responses or tumor control. Heterogeneity in photosensitizer uptake and tumor oxygenation, lack of significant tumor specificity for photosensitizer uptake, and the heterogeneity in tissue optical properties may account for the lack of efficacy observed. Photofrin-mediated i.p. photodynamic therapy delivered as a single treatment is unlikely to lead to a sustained complete response in patients with i.p. malignancies who have gross disease. This situation is similar to the early development of radiation therapy where it was discovered that single, large doses of ionizing radiation were not as effective as fractionated treatment. The use of methods to enhance the efficacy of photodynamic therapy against tumor compared with normal tissues, improved techniques for light delivery, and fractionated photodynamic therapy are reasonable approaches to consider in future studies of i.p. photodynamic therapy.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Received 7/27/05; revised 2/ 2/06; accepted 2/ 9/06.
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. M. Hahn, M. E. Putt, J. Metz, D. B. Shin, E. Rickter, C. Menon, D. Smith, E. Glatstein, D. L. Fraker, and T. M. Busch Photofrin uptake in the tumor and normal tissues of patients receiving intraperitoneal photodynamic therapy. Clin. Cancer Res., September 15, 2006; 12(18): 5464 - 5470. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
