| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Article |
, for Patients With Advanced Solid Tumors: Surgical Experience and Long-Term Follow-Up
1 Department of Surgery, Sammons Cancer Center of Baylor University Medical Center, Worth Street #420, Dallas, Texas 75246
2 Mary Crowley Medical Research Center, Texas Oncology PA, 3500 Gaston Avenue, Dallas, Texas 75246
Correspondence: Address correspondence and reprint requests to: Todd M. McCarty, MD; E-mail: toddmc{at}baylorhealth.edu.
| ABSTRACT |
|---|
|
|
|---|
to treat cancer. Many of these studies demonstrated dose-limiting systemic side effects from high concentrations of TNF-
. The recent focus has been on developing a local delivery system for TNF-
to minimize the systemic response.
Methods: This study was part of a phase I open-label multi-institutional trial using TNFerade. We focus on the patients treated at Baylor University Medical Center and provide postoperative and long-term follow-up. TNFerade uses a second-generation nonreplicating adenovirus as the vector for delivery of the human transgene TNF-
. An early growth response 1 promoter was placed upstream from the TNF-
gene. This promoter is activated by ionizing radiation, thus allowing for temporal and spatial control of TNF-
release. Tumors were injected over 5 weeks with ionizing radiation given 3 days after injections for 6 weeks. Tumor response was measured by computed tomographic imaging and physical examination.
Results: As described in our original experience, no patients experienced dose-limiting toxicities up to doses of 4 x 1011 particles per injection. Tumors injected demonstrated a response independently of histology. Four patients had complete regression of the tumor injected. Three patients with complete regression have survived
2 years from the time of treatment.
Conclusions: Both short-term and long-term safety are observed with TNFerade. These data demonstrate the need for phase II trials.
Key Words: TNFerade Tumor necrosis factor
Adenovirus Phase I
| INTRODUCTION |
|---|
|
|
|---|
as a cytokine involved in tumor necrosis. Since then, multiple studies have been performed to uncover the extent of its function and potential uses in human tumors. It has now been elucidated that TNF-
is a multifunctional, potent, tumoricidal cytokine released by macrophages.25
TNF-
can induce tumor cell death by several mechanisms. First, TNF-
binds to a number of possible receptors that activate a cascade of events that ultimately activate important caspases, cysteine proteases, thus leading to an ordered cell death or apoptosis.68 The mechanism is complex but includes TNF-
binding to a receptor, which then undergoes a conformational change that activates its death domain. The death domain complex then initiates the activation of several caspases, including caspases 3, 7, and 8. Through the action of these caspases, a cascade of events unfolds, including stimulation of mitochondria to release oxygen free radicals and cytochrome c.9 The activated caspases, cytochrome c, and the oxygen free radicals contribute to the eventual dissolution of the nuclear membrane and, ultimately, apoptosis.7,1012 Activation of TNFR55 also activates the sphingomyelinase pathway, which forms ceramide.8,13 Ceramide seems to activate the caspase and contribute to apoptosis.8,14 Further pathways and alternate mechanisms are currently being studied that suggest the ability for TNF-
to overcome tumor cells antiapoptotic defenses (i.e., bcl-2 and nuclear factor-
B), which contribute to chemotherapy and radiation resistance.9,10,15 Second, TNF-
contributes to the obliteration of tumor neovasculature. TNF-
increases the production of platelet-activating factor, tissue factor, angiotensin, and oxygen free radicals, all of which contribute to vascular thrombosis and subsequent necrosis.1,16,17 Finally, TNF-
increases the immune response to tumor cells by increasing T-cell activity, natural killer cell activity, and major histocompatibility I expression.1821
TNF-
has, in addition, been demonstrated to be synergistic with radiation, thus priming previously radiation-resistant tumors to be more responsive to the effect of radiation.10,22,23 Evidence also suggests that TNF-
may selectively destroy tumor cells, leaving normal cells and normal vasculature intact.5,16,18,24,25
Using these physiologic concepts, multiple animal studies and clinical trials have been performed to incorporate TNF-
into the oncological treatment of solid tumors. Initial systemic administration was attempted. Although documentable decreases in tumor size were noted, systemic toxicity, including severe hypotension and shocklike symptoms, limited this approach.2,2628 Further attempts using TNF-
, radiation, and melphalan in isolated limb perfusion therapy for extremity tumors have resulted in improved tumor responses. Occasional systemic toxicity resulting in notable hypotension has required lower doses than levels assumed to be most effective according to several animal studies.2931 Recently, gene therapy studies have focused on local administration to provide high-dose TNF-
and minimize systemic symptoms. Unfortunately, these studies also noted a systemic leak of TNF-
, presumably from the uncontrolled activation of TNF-
.32 TNFerade, developed by GenVec, Inc. (Gaithersburg, MD), is a second-generation nonreplicating adenovirus with the human TNF-
gene inserted downstream from an early growth response 1 promoter.33,34 The early growth response 1 promoter is induced by ionizing radiation; thus, the timing and spatial constraint of TNF-
release can be controlled. The proposed advantage of this construct is to minimize systemic symptoms from TNF-
.32,33,35,36 Furthermore, because the adenovirus does not incorporate its genome into the host DNA, the presence of the TNF-
gene is only temporary; studies demonstrate the absence of TNF-
approximately 3 weeks after the last injection.22,33,37,38
We thus investigated the use of TNFerade in combination with radiotherapy to determine safety and evidence of activity. Long-term follow-up indicates no late toxic effects.
| METHODS |
|---|
|
|
|---|
18 years of age, (4) Karnofsky score of
60, (5) life span estimated >3 months, (6) negative pregnancy test and effective use of contraception, (7) measurable disease, (8) liver enzymes less than five times higher than the normal range, (9) no recent acute infections, and (10) no chemotherapy or experimental medication in the previous 4 weeks. Accrual began in August 2001 and ended in June 2002. Each patient provided signed informed consent. The institutional review board and the Institutional BioSafety Committee reviewed and approved the protocol. The Recombinant Advisory Committee to the National Institutes of Health Director reviewed the protocol in accordance with the National Institutes of Health guidelines for gene and cell therapy products. The Food and Drug Administration also reviewed the protocol as part of an Investigational New Drug application.
Intratumoral injections to fully encompass the tumor were performed two times a week for weeks 1 and 2 as monomodal therapy and then one time per week during weeks 3 through 6 concurrently with 1.8 to 2.0 Gy/day five times per week for up to 5 weeks. On the days when both treatments were delivered, radiation was given 4 hours after the TNFerade injection. Each dose of TNFerade was divided into 2 to 5 aliquots with a volume-dependent injection for a total volume of 2 to 5 mL. These injections were given in a rotating clockwise manner, with the first injection at the 12-3-6-9-oclock position, the following injections at 1-4-7-10-oclock, and so on. Patients were entered in cohorts of three. Doses ranged from 4 x 107 particle units increasing by half-log intervals to 4 x 1011 particle units or until a maximum tolerated dose was encountered. Patients were followed up for tumor response per physical examination and by computed tomographic (CT) imaging. Biopsies or resection were performed for complete responders, and samples were examined microscopically for confirmation of response.
Patients were also evaluated for toxicities and side effects related to therapy. Dose-limiting toxicity (DLT) was defined by using the National Cancer Institute Common Toxicity Criteria as (1) any grade
3 toxicity related to TNFerade and radiation up to 2 weeks after treatment; (2) any grade 4 thrombocytopenia, neutropenia, or anemia; and (3) any grade
3 nausea, vomiting, or diarrhea up to 2 weeks after treatment.
The target tumor cross-sectional area for each patient was measured on baseline CT imaging and was remeasured after completion of the 6-week course. The change in tumor size was calculated, and patients were categorized by the extent of response. Complete response (CR) was defined as a 100% tumor response, partial response was a response between 50% and 99%, minimal response was between 25% and 49%, and stable disease was defined as a 0% to 24% response. The patients who completed the 6-week course were then followed up to determine response.
Neutralizing antibody titers against adenovirus were measured by using a standard cell infiltration assay at baseline, day 5 of week 1, at the end of treatment, and 4 weeks after treatment.34,35 Virus cultures were also performed baseline and at the end of the study.
TNF-
levels were measured at baseline; days 1, 2, 4, and 5 of week 1; days 2 and 5 of week 2; and day 2 of weeks 3 to 6. These measurements were quantified with an enzyme-linked immunosorbent assay kit (Quantikine HS, R&D Systems, Inc., Minneapolis, MN), which is able to detect down to a level of .18 pg/mL.34,35
| RESULTS |
|---|
|
|
|---|
|
Adenovirus antibodies were measured before therapy and throughout treatment. No significant side effects were associated with preexisting adenoviral antibodies. Of note, patients with preexisting antibodies still had a significant tumor response. TNF-
levels were measured during the study with a median level of 8 pg/mL (range, 350 pg/mL). These levels were well below the known maximum systemically tolerated dose of 5000 pg/mL.39,40
Follow-up ranged from 1 to 25 months, with a median follow-up of 12 months, including both physical examination and CT imaging. No long-term adverse events have been observed. None of the six patients with either a CR or partial response after treatment had a reoccurrence at follow-up (Table 2
). In the CR group, surgical resection was performed, and wound healing and tissue integrity were observed along with tumor response. No surgical complication or wound healing difficulty was observed. In the minimal response group, three of five patients had no change in tumor size at follow-up evaluation. One patient with cholangiocarcinoma had local progression of disease, and one patient died from a malignant pleural effusion shortly after completion of treatment. In the stable disease group, all four patients had no evidence of tumor growth at follow-up.
|
24 months. | DISCUSSION |
|---|
|
|
|---|
via a radiation-inducible promoter is a clinically and technically feasible treatment for continued investigation. TNFerade has a large therapeutic window, and doses up to 4 x 1011 can be used without significant adverse effects. It is notable that 15 of the 16 patients in this study demonstrated some degree of objective response to therapy. These patients had each been treated with and failed the current and standard therapy for their tumors. The results seen in tumors that had recurred after prior chemotherapy and radiation suggest a potential role in palliative treatment regimens.30 In reference to the study performed by Senzer et al.,34 which compared injected versus noninjected lesions, these results seem to be induced by the presence of TNFerade.
Similar to data with isolated limb perfusion, this study suggests that melanomas may be particularly sensitive to the combination of radiation and TNFerade.30,41 The three patients with melanoma had recurrent disease after prior systemic therapy that included combinations of cisplatin, interferon
, and interleukin 2, yet they have remained disease free up to 12 months since completing TNFerade treatment. A recently completed phase I trial by Mundt et al.35 using TNFerade in inoperable and radiation-resistant sarcomas demonstrated dramatic results; some patients had a complete tumor response when excision was performed. Again, in this phase I trial, no DLTs or maximum tolerated doses were encountered. Two phase II studies are currently under way at our institution evaluating combined-modality therapy (radiation and chemotherapy) with TNFerade as either neoadjuvant (esophageal cancer) or definitive (pancreatic cancer) therapy.
TNF-
selectively disrupts angiogenic endothelial cells and may interfere with wound healing. Previous animal studies have described erythema and ulceration at injection sites with doses >4 x 1012 PFUs.33 In our study, no patients experienced wound closure difficulties or grade 3 or 4 skin-related toxicities. Two patients in the previously mentioned phase I study for sarcomas had significant wound-closure difficulties on follow-up. Both of these patients had proximal extremity sarcomas with a diameter >18 cm. Local infection was documented in one patient that may have resulted in local synergistic inhibition with endotoxin. Although preclinical studies have demonstrated the specificity of TNF-
for tumor cells and vessels, wound healing may be a problem in specific patient subsets.2,16,19,42 This data point will need to be closely followed up in future studies. The effect of TNF-
on anastomoses is not known and will need to be prospectively evaluated in local resection candidates. TNFerade may potentially allow for significantly decreased preoperative radiation requirements, thus improving the ease of surgical resection and decreasing postoperative complications related to operating in a radiated field. At this point, the results are limited to regions that can be locally injected and irradiated. However, the addition of TNFerade may contribute to preoperative downsizing of a tumor, thus allowing for unresectable tumors to then become operative candidates.
Of additional interest in this study was the association of CT imaging and physical examination findings compared with the pathologic results. All of the melanomas were noted to have dramatic yet incomplete resolution of the mass, but pathologic evaluation revealed no viable tumor cells. Two patients had positron emission tomography scans performed on these tumors, and they revealed no activity, thus suggesting a legitimate preoperative test to evaluate tumor response before resection.
In conclusion, TNFerade injected intratumorally in combination with ionizing radiation is safe, tolerable, and technically feasible, with no significant side effects and no maximum tolerated dose. The promising response rates of TNFerade seem to be independent of the primary tumor site. Phase II studies will need to be performed to elucidate its role in neoadjuvant therapy for pancreatic, esophageal, and breast cancers; nonsmall-cell lung carcinomas; sarcomas; and colorectal carcinomas. Further studies will be needed to determine the role for TNF-
in cancers unresponsive to current radiation and chemotherapy regimens.
| ACKNOWLEDGMENTS |
|---|
Received for publication March 7, 2003. Accepted for publication April 21, 2005.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
W. Weber, W. Bacchus, M. Daoud-El Baba, and M. Fussenegger Vitamin H-regulated transgene expression in mammalian cells Nucleic Acids Res., September 27, 2007; 35(17): e116 - e116. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. Cross and J. K. Burmester Gene therapy for cancer treatment: past, present and future. Clin. Med. Res., September 1, 2006; 4(3): 218 - 227. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |