Neoadjuvant Chemoradiation for Adenocarcinoma of the Pancreas

1055-3207 /98 $8.00

PANCREATIC CANCER

+ .00

NEOADJUVANT CHEMORADIATION FOR ADENOCARCINOMA OF THE PANCREAS Alexander R. Miller, MD, Emily K. Robinson, MD, Jeffrey E. Lee, MD, Peter W.T. Pisters, MD, Paul J. Chiao, PhD, Renato Lenzi, MD, James L. Abbruzzese, MD, and Douglas B. Evans, M D

Clinical research in pancreatic adenocarcinoma at the University of Texas M.D. Anderson Cancer Center (MDACC) has focused on the use of chemoradiation strategies as a mechanism to improve local-regional tumor control in patients with potentially resectable disease.14Critical to the accurate analysis of preoperative or postoperative adjuvant therapy is the incorporation of a standardized approach to patient selection (pretreatment staging), operative technique, and pathologic evaluation of surgical specimens. The inability of institutions to standardize these important variables is largely responsible for the small amount of data that currently exists on the use of multimodality therapy for pancreatic cancer Therefore, this chapter briefly outlines the critical aspects (Table 1).18,23,28,31,58,67.70 of radiographic staging, surgical technique, and pathologic evaluation of the resected specimen that are necessary to conduct multimodality clinical trials. This article focuses on current and future chemoradiation strategies for patients with potentially resectable adenocarcinoma of the pancreatic head and uncinate process, emphasizing the use of new radiation sensitizing agents and systemic therapies of low toxicity. PREOPERATIVE RADIOGRAPHIC STAGING

Tumors of the pancreas are unlike other solid tumors of the gastrointestinal tract because accurate diagnosis, clinical staging, and pathologic evaluation of From the Departments of Surgical Oncology, Gastrointestinal Oncology and Digestive Diseases, and Cell Biology, The University of Texas M. D. Anderson Cancer Center, Houston, Texas SURGICAL ONCOLOGY CLINICS OF NORTH AMERICA VOLUME 7. NUMBER I .JANUARY 1998

183

184

MILLER et a1

Table 1. RECENT CHEMORADIATION STUDIES IN PATIENTS WITH RESECTABLE

PANCREATIC CANCER First Author (Year) Postoperative (adjuvant) Kalser3' (1995)

surgery alone GITSG23(1987) Whittingtonw(I 991) Fool8(1993) yeo7~ (1 995) surgery alone

No. Patients*

EBRT Dose (Gy)

Chemotherapy Agent(s)

Median Survival (mo)

21 22 30 28 29 56 22

Preoperative (neoadjuvant) Hoffmanz8(1995) 23 Staleys8(1996) 39 EBRT = external-beam radiation therapy, 5-FU = 5-fluorouracil, Mito-C = mitornycin C. *All patients underwent a pancreatectomywith curative intent. Adapted from Spitz Fr, Abbruzzese J, Lee J, et al: Preoperative and postoperative chemoradiation strategies in patients treated with pancreatico-duodenectomyfor adenocarcinoma of the pancreas. J Clin Oncol 15:928, 1997, with permission.

resected specimens require extensive interaction and cooperation between physicians of different specialties.Accurate clinical staging requires high-quality (helical) CT to define accurately the relationship of the tumor to the celiac axis and superior mesenteric vessels. In the absence of extrapancreatic disease, the relationship of the low-density tumor mass to the superior mesenteric artery (SMA) and celiac axis is the main focus of preoperative imaging studies. The goal of both the pancreatic surgeon and the radiologist in assessing resectability is the accurate prediction of the likelihood of obtaining a negative retroperitoneal margin of resection. Local tumor resectability is most accurately assessed preoperatively; no role exists for exploratory surgery in patients with adenocarcinoma of the pan~ r e a s . We ~ ' use objective, reproducible radiologic criteria to define potentially resectable disease as (1) the absence of extrapancreatic disease, (2) the absence of direct tumor extension to the SMA and celiac axis as defined by the presence of a fat plane between the low-density tumor and these arterial structures, and (3) a patent superior mesenteric-portal vein (SMPV) confluence. The accuracy of this orm of radiographic staging is supported by our initial studyZ0and confirmed in a recent report by Spitz et alS6demonstrating a resectability rate of 80% (94/118) and a low rate of microscopic retroperitoneal margin positivity (17%).The accuand the inaccuracy of intraoperative racy of CT in predicting unre~ectability'~,~~ assessment of resectability51are both well established. Pretreatment staging to exclude patients with locally advanced disease is critical to allow accurate interpretation of results from studies that examine the value of multimodality therapy in patients with pancreatic cancer. CT Technique

At MDACC, helical CT is generally used in the evaluation of patients with presumed pancreatic neoplasms. The development of helical or spiral scanning has improved scan speed; the continuous rotation of the x-ray tube around the gantry allows the entire pancreas to be imaged during the bolus phase of contrast

185

NEOADJUVANT CHEMORADIATION FOR ADENOCARCINOMA OF THE PANCREAS

enhancement. In addition, scan data can be processed to display images in threedimensional and multiplanar formats. Dilute Gastrografin or 2% barium sulfate is used to opacify the stomach and small bowel before scanning. Water can be used as an oral contrast agent when it is necessary to evaluate the gastric wall or duodenum. Precontrast CT of the liver and pancreas is performed at 10-mm slice thickness to localize the pancreas. Nonionic contrast material (300 mg/dL) is then delivered intravenously by an automatic injector at a rate of 2 to 3 mL/sec for a total of 150 mL. Helical CT of the pancreas is performed 60 to 70 seconds after the start of the injection. A dynamic series of scans through the pancreas is completed at 3-mm slice thickness with a pitch factor of 1.5 to 2.0, depending on the anatomic extent of the tumor. The slice thickness can be increased to 5 mm in a large patient. The rest of the abdomen is then scanned at 7-mm slice thickness. SURGICAL TECHNIQUE

Surgical resection incorporates a six-step technique for pancreaticoduodenectomy, with specific attention to the retroperitoneal dissection along the right lateral border of the SMA.15This strategy is used because a survival benefit from surgical resection of the primary tumor is realized only in patients who undergo a negative-margin pancreaticoduodenectomy. The median survival of 8 to 12 months in patients who undergo pancreaticoduodenectomy and are found to have a positive is no different than the median survival margin of resection (Table 2)37,44,62,63,67,68,70 reported for patients with locally advanced disease who are treated with palliative chemoradiation without surgical resection of the pancreas.22Although most studies have not defined precisely the retroperitoneal margin, it is reasonable to assume that the margin most frequently reported as positive in patients who undergo pancreaticoduodenectomy is along the superior mesenteric vein (SMV) or proximal SMA.19Therefore, current data do not justify operation in patients with evidence of arterial encasement on pretreatment imaging studies because it is impossible to perform a margin-negative pancreaticoduodenectomy with tumor involvement of the SMA. In contrast to tumor extension to the SMA, isolated involvement of the SMV or SMPV confluence is treated with venous resection and reconstruction. Tradi-

Table 2. MEDIAN SURVIVAL FOR PATIENTS WHO UNDERWENT SURGICAL RESECTION FOR ADENOCARCINOMA OF THE PANCREAS AND WERE FOUND TO HAVE A POSITIVE MARGIN OF RESECTION Margin

Reference (Year) TepperQ (1976) Tredem (1990) Whittingtonm (I 991) Willett68(1993) N i t e ~ k(1i 995) ~~ y e o 7 (1 ~ 995) L i l l e r n ~ e(1~996) ~

17* 54 19 37 28 58 64

G/M G/M G G/M G G/M G/M

Median Survival (mo)

8 10

t

11 9 10 12

G = grossly positive margin; M = microscopically positive margin. *All patients also had positive regional lymph nodes. t T w o patients alive at 18 months of follow-up. Adapted from Staley C, Lee J, Cleary K, et al: Preoperative chemoradiation therapy for adenocarcinoma of the pancreatic head. Am J Surg 171: 118, 1996; with permission.

186

MILLER et a1

tionally, tumor involvement of the SMV or SMPV confluence has been considered a contraindication to pancreaticoduodenectomy. However, reports of venous resection at the time of pancreaticoduodenectomy often involved patients with retroperitoneal tumor extension that involved the SMA or celiac axis and resulted in ~ ,contrast, ~~ isolated involvement of the SMPV conincomplete tumor r e s e c t i ~ n .In fluence without radiographically evident involvement of the SMA can be managed intraoperatively with resection of the involved segment of vein and vascular reData from our institution demonstrate that resection of the SMV at constr~ction.~~ the time of pancreaticoduodenectomy can be performed safely and is not associated with retroperitoneal margin positivity (when high-quality preoperative imaging excludes patients with tumor extension to the SMA).21Detailed evaluation of patients who require venous resection and reconstruction reveals a long-term outcome that is comparable to that of similarly staged patients who do not require ~ ~ ,fundamental ~~ distinction between tumor involvement vascular r e s e c t i ~ n .The of the SMV or SMPV confluence and tumor involvement of the SMA is critical to the accurate interpretation of any study that reports results of extended pancreaticoduodenectomy to include vascular resection and reconstruction. STANDARDIZED PATHOLOGIC STAGING OF OPERATIVE SPECIMENS

Accurate pathologic assessment of surgical specimens is critical for both the evaluation of innovative preoperative treatment strategies and the development of reproducible prognostic predictors of patient survival and treatment failure. Retrospective pathologic analysis of archival material does not allow accurate assessment of margins of resection or number of lymph nodes retrieved. The standard pathologic evaluation of the pancreaticoduodenectomy specimen developed at MDACC57begins by performing frozen-section evaluations of the common bile duct transection margin, the pancreatic transection margin, and the retroperitoneal transection margin. The retroperitoneal margin is defined as the soft-tissuemargin directly adjacent to the proximal 3 to 4 cm of the SMA. This margin is evaluated by frozen-section microscopic examination of a 2- to 3-mm full-face (en-face)section of the margin. A positive bile duct or pancreatic transection margin is treated with re-resection; this is not possible in the retroperitoneum, where the aorta and SMA origin limit the extent of surgical resection. Samples of multiple areas of each tumor, including the interface between tumor and adjacent uninvolved tissue, are submitted for paraffin-embedded histologic examination (5-10 blocks). Four-micrometer-thick sections are cut and stained with hematoxylin and eosin. Final pathologic evaluation of permanent sections includes a description of tumor histology and differentiation, gross and microscopic evaluation of the tissue of origin (pancreas, bile duct, ampulla of Vater, or duodenum), and assessments of maximal transverse tumor diameter, the presence or absence of perineural, lymphatic, and vascular invasion, and lymph node status. When segmental resection of the superior mesenteric vein is required, the area of presumed tumor invasion of the vein wall is sectioned serially and examined in an attempt to discriminate benign fibrous attachment from direct tumor invasion. In patients who receive preoperative chemoradiation, the grade of treatment effect is assessed on permanent sections. In 1991, a method for classifying subsets of regional lymph nodes in pancreaticoduodenectomy specimens was developed.57The soft fibrofatty tissue that contains regional lymph nodes was divided into six regions outlined on an anatomic pathology dissection board.57We previously reported on the prospective analysis of pancreaticoduodenectomy specimens from 40 consecutive patients with histo-

NEOADIUVANT CHEMORADIATION FOR ADENOCARCINOMA OF THE PANCREAS

187

logically confirmed adenocarcinoma of the pancreatic head. Fifteen consecutive specimens were examined before the introduction of the standardized pathologic evaluation, and 25 consecutive specimens underwent standardized pathologic evaluation. With the introduction of standardized pathologic evaluation, a statistically significant increase in the number of lymph nodes was identified in the surgical specimens. As the use of multimodality treatment strategies for patients with pancreatic cancer becomes more common, standardization of pathologic assessment of tumor specimens will be critical. Our systematic approach should serve as a model for others engaged in protocol-based clinical research that involves the surgical management of patients with pancreatic cancers. MULTIMODALITY TREATMENT STRATEGIES

External-beam radiation therapy and concomitant 5-fluorouracil (FU) chemotherapy (chemoradiation) were shown to prolong survival in patients with locally advanced adenocarcinoma of the pancreas.22Those data were the foundation for a prospective, randomized study of adjuvant chemoradiation (500 mg/ m2/day of 5-FU for 6 days and 40 Gy of radiation) after pancreaticoduodenectomy conducted by the Gastrointestinal Tumor Study Group (GITSG). That trial also demonstrated a survival advantage from multimodality therapy compared with ~ ~ , ~ ~ owing to a prolonged recovery, 5 (24%) of the 21 resection a l ~ n e .However, patients in the adjuvant chemoradiation arm could not begin chemoradiation until more than 10 weeks after pancreaticoduodenectorny. This was true despite the fact that the patients likely to be considered for protocol entry were those who recovered rapidly from surgery and had a good performance status. The number of patients who experienced delayed recovery from surgery and therefore were not considered for postoperative adjuvant therapy was unknown. Recent data from MDACC and Johns Hopkins University suggest that at least 25% of patients who undergo pancreaticoduodenectorny do not receive intended adjuvant therapy if a postoperative strategy is The magnitude of the operation and its associated morbidity often results in a lengthy period of recovery that presents a major obstacle to the routine use of postoperative chernoradiation. However, in those patients who successfully receive adjuvant therapy, recent data support results from the GITSG study. Yeo et a170reviewed the outcome of all patients who underwent pancreaticoduodenectorny for adenocarcinoma of the pancreatic head during a 4-year period. Fifty-six patients received adjuvant chemoradiation, and 22 underwent pancreaticoduodenectorny alone. Despite the chemoradiation group containing a larger percentage of patients with aneuploid tumors, median survival for that group was 20 months compared with 12 months for the group who received surgery alone. The risk of delaying adjuvant therapy, combined with small published experiences of successful pancreatic resection after radiation therapy alone, prompted many institutions to initiate studies in which chemoradiation was given before pancreaticoduodenectorny for patients with potentially resectable (or locally advanced) adenocarcinoma of the pancreas.I4The preoperative use of chemoradiation is supported by the following considerations14:(1)The high frequency of positive-margin resections recently reported supports the concern that the retroperitoneal margin of excision, even when negative, may be only a few millimeters; therefore, surgery alone may be an inadequate strategy for local tumor control. (2) Peritoneal tumor cell implantation due to the manipulation of surgery may be prevented by preoperative chernoradiation. (3) Patients with disseminated disease that is evident on restaging studies after chemoradiation are not subjected to laparotomy. (4) Because radiation therapy and chemotherapy are given first,

188

MILLER et a1

delayed postoperative recovery has no effect on the delivery of multimodality therapy, a frequent problem in adjuvant therapy studies. (5) The preoperative deliverv of chemoradiation does not increase perioverative morbiditv or mortality in pati&ts who undergo pancreaticoduodenecto&y, and recent data suggest that preoperative chemoradiation may decrease the incidence of pancreaticojejunal . . anasiomotic fistula.38 In patients who receive chemoradiation before surgery, a repeat staging CT If these patients had scan after chemoradiation reveals liver metastases in 25%.13,56 undergone pancreaticoduodenectomy at the time of diagnosis, it is probable that the liver metastases would have been subclinical; therefore, these patients would have undergone a major surgical procedure only to have liver metastases found soon after surgery. In the MDACC trial, patients who were found to have disease progression at the time of restaging had a median survival of only 6.7 months.13 The avoidance of a lengthy recovery period and the potential morbidity of pancreaticoduodenectomy in patients with such a short expected survival duration represents a distinct advantage of preoperative over postoperative chemoradiation. When delivering multimodality therapy for any disease, it is beneficial, when possible, to deliver the most toxic therapy last, thereby avoiding morbidity in patients who experience rapid disease progression not amenable to currently available therapies. The survival advantage for the combination of chemoradiation and surgery compared with surgery alone likely results from improved local-regional tumor control. Because of the poor rates of response to 5-FU-based systemic therapy in patients with measurable metastatic disease, it is unlikely that previously reported chemoradiation regimens significantly impacted the development of distant metastatic disease. Recent data from Staley et alS8at MDACC support this belief. Thirty-nine patients received preoperative 5-FU-based chernoradiation, pancreaticoduodenectomy, and electron-beam intraoperative radiation therapy for adenocarcinoma of the pancreatic head. The liver was the most frequent site of tumor recurrence, and liver metastases were a component of treatment failure in 69% of all patients who had recurrences. Forty-eight percent of patients who had recurrences had liver metastases as their only site of recurrence. Isolated local or peritoneal recurrences were documented in only four patients (11%). In contrast, previous reports of pancreaticoduodenectomy for adenocarcinoma of the pancreas have documented local recurrence in 50% to 80% of patients.14This improvement in local-regional control was seen despite the fact that 14 of 38 evaluable patients had undergone laparotomy with tumor manipulation and biopsy before referral for chemoradiation and reoperation. If these 14 patients were excluded, only 2 patients (8%)would have experienced local or peritoneal recurrence as any component of treatment failure. However, because of the large percentage of patients who developed distant metastatic disease, predominantly in the liver, improved local-regional tumor control translated into only a small improvement in median survival compared with that in other recently published studies. Therefore, in the absence of more effective systemic therapy, the goal of chemoradiation (preoperative or postoperative) and pancreatectomy should be to maximize local-regional tumor control while minimizing treatment time, treatment-related toxicity, and cost. This basic principle has guided the evolution of multimodality therapy for potentially resectable pancreatic cancer at our institution. The first report of standard-fractionation chemoradiation (50.4 Gy over 5.5 weeks with concomitant 5-FU) from MDACC documented gastrointestinal toxic effects (nausea, vomiting, and dehydration) that required hospital admission in one third of patients.16A similar experience was reported from the multicenter

NEOADJUVANT CHEMORADIATION FOR ADENOCARCINOMA OF THE PANCREAS

189

Eastern Cooperative Oncology Group (ECOG) triaLZ8This finding caused us to change the delivery of radiation therapy and 5-FU to a rapid-fractionation program of chemoradiation designed to avoid the gastrointestinal toxicity seen with our standard 5.5-week program while attempting to maintain the excellent local tumor control achieved with multimodality therapy.46Rapid-fractionation chemoradiation was delivered over 2 weeks with 18-MeV photons using a four-field technique to a total dose of 30 Gy, prescribed to the 95% isodose, at 3 Gy/fraction (10 fractions), 5 d/wk. 5-FU was given concurrently by continuous infusion at a dosage of 300 mg/mZ/day, five d/wk. This program was based on the principle that the total radiation dose required to obtain a given biologic effect decreases as the dose per fraction increases.12Restaging with chest radiography and abdominal CT was performed 4 weeks after completion of chemoradiation in preparation for pancreaticoduodenectomy. Thirty-five patients received this treatment, 27 were taken to surgery, and 20 (74%) underwent successful pancreaticoduodenectorny. Local tumor control and patient survival were equal to our results with standardfractionation (5.5 weeks) ~hemoradiation.~~ In an effort to compare preoperative and postoperative chemoradiation strategies, we recently reported on the multimodality treatment of 142 consecutive patients with localized adenocarcinoma of the pancreatic head who were deemed able to undergo resection on the basis of pretreatment radiographic images.56The subset of 41 patients who completed protocol-based preoperative chemoradiation and pancreaticoduodenectomy [27 patients received standard-fractionation chemoradiation (50.4Gy) and 14 patients received rapid-fractionationchemoradiation (30 Gy)] were compared with 19 patients who received pancreaticoduodenectomy and postoperative adjuvant chernoradiation. Overall median follow-up for these 60 patients was 19 months. No patient who received preoperative chemoradiation experienced a delay in surgery because of chemoradiation toxicity, but 6 (24%)of 25 eligible patients did not receive intended postoperative chemoradiationbecause of delayed recovery after pancreaticoduodenectorny. Patients who were treated with rapid-fractionation preoperative chemoradiation had a significantly ( P < 0.01) shorter duration of treatment (median, 62.5 days) compared with patients who received postoperative chemoradiation (median, 98.5 days) or standardfractionation preoperative chemoradiation (median, 91 days) (Fig. 1). No patient who received preoperative chemoradiation and pancreaticoduodenectorny experienced a local recurrence; peritoneal (regional) recurrence occurred in 10% of these patients. Local or regional recurrence occurred in 21% of patients who received pancreaticoduodenectomy and postoperative chernoradiation. Despite surgeons' ability to perform pancreaticoduodenectomy safely, this procedure remains too extensive and complex to enable the consistent postoperative delivery of standard-fractionation adjuvant chernoradiation. In the absence of compelling data that demonstrate superior survival results with either a preoperative or postoperative treatment approach, our data suggest that a greater proportion of patients receive potentially beneficial adjuvant therapy when chemoradiation is administered before pancreaticoduodenectomy. Moreover, preoperative treatment strategies spare a significant number of patients the morbidity and mortality associated with laparotomy, because up to one fourth of patients show evidence of metastatic disease at the time of preoperative restaging after chernoradiation. Therefore, in an effort to build on our experience with 5-FU-based chemoradiation, current (and future) therapies combine improved radiation sensitizing agents, radiation therapy, and surgery with the systemic or regional delivery of novel agents that inhibit essential steps in tumor cell growth (Fig. 2).

190

I

MILLER et a1

I

I

ChernoXRT

bjj

Surgery

1

A

ChernoXRT

I

I

I

I

ChernoXRT

P

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

restaging evaluation

Week

Figure 1. The three multimodality treatment schemas used in this study and the median duration of treatment for each component of therapy. A = standard-fractionationpreoperative chemoradiation (ChernoXRT); B = rapid-fractionation preoperative chemoradiation; and C = standard-fractionation postoperative chemoradiation: standard-fractionation chemoradiation (A and C) was delivered over 5.5 weeks, and rapid-fractionation chemoradiation (61was delivered over 2 weeks. Surgery defines the length'of hospitalization required for p&&eaticoduodenectomy (A and B,15 days; C, 14 days). In the preoperative chemoradiationschemas (A and B), surgery was performed 4 to 5 weeks following the completion of chernoradiation. In the postoperative chemoradiation schema (C) chemoradiation was initiated following satisfactory recovery from pancreaticoduodenectomy(median 6.5 weeks). (Adapted from Spitz FR, Abbruzzese J, Lee J, et al: Preoperative and postoperative chemoradiation strategies in patients treated with pancreaticoduodenectomyfor adenocarcinoma of the pancreas. J Clin Oncol 15; 933, 1997; with permission.)

NEW RADIATION SENSITIZING AGENTS Paclitaxel

Paclitaxel is a plant product isolated from the stem bark of Taxus brevifolia, the western yew, a small evergreen indigenous to the Pacific N ~ r t h w e s tPacli.~~ taxel enhances the polymerization of tubulin to stable microtubules and inhibits spindle cell function during mitosis, thereby preventing normal cell replication. Cells exposed to paclitaxel experience growth arrest in the G2/M phase of the cell cycle-a state during which they are especially sensitive to irradiation. In addition to its effect on the cell cycle, paclitaxel may enhance the effect of radiation therapy through reoxygenation of hypoxic tumor cells.40Tumor cell loss due to paclitaxelinduced apoptosis results in improved oxygenation of the remaining neoplastic cells within a tumor, probably due to improved blood flow to the remaining cells; oxygenated cells are more sensitive to the toxic effects of radiation than are hypoxic cells. In clinical trials, patients with a variety of solid tumors, including ovary, breast, and metastatic pancreatic adenocarcinoma, have demonstrated objective responses to taxanes (paclitaxel and docetaxel) despite significant tumor burdens that failed to respond to conventional the rap^.^,^^ Recently, Safran et a155from the Brown University Oncology Group performed a phase I study using paclitaxel and concurrent radiation therapy in patients with locally advanced pancreatic and

NEOADJUVANT CHEMORADIATION FOR ADENOCARCINOMA OF THE PANCREAS

0

Pretreatment Evaluation

191

2 3 4 5 6 7 0 9 10 11 12 13 1 4 0

Restaging Evaluation

systemic agent:

.

Paclitaxel Antiangiogenic agents Matrix metalloproteinase inhibitors Agents directed at the cell cycle

Figure 2. The future of multimodality therapy for patients with potentially resectable adenocarcinoma of the pancreatic head. Treatment schemas emphasize the importance of minimizing toxicity and treatment duration, while attempting to improve therapeutic efficacy. Cytotoxicity is enhanced by combining radiation therapy with more potent radiation-sensitizing agents. Systemic therapy is continued after chemoradiation and surgery with systemicagents of low toxicity directed at specific molecular events involved in pancreatic tumorigenesis (i.e., inhibition of angiogenesis,induction of apoptosis, or arrest of the cell cycle).

gastric adenocarcinoma. Dose-limiting toxicity was due to abdominal pain, nausea, and anorexia and occurred at a dose of 60 mg/m2/wk. Four objective (radiographic) partial responses were observed in 13 patients with pancreatic cancer. These data provide the rationale for the current pilot trial being conducted at MDACC of preoperative paclitaxel and rapid-fractionation radiation therapy for patients with potentially resectable adenocarcinoma of the pancreatic head. After chemoradiation, patients are restaged and, in the absence of extrapancreatic disease, undergo pancreaticoduodenectomy. The resected tumor is then available for pathologic assessment of the degree of treatment response due to the combination of paclitaxel and irradiation. Preliminary experience with this regimen has demonstrated minimal toxicity and improved histologic response in the resected pancreatic tumor compared with previous studies with 5-FU-based preoperative chemoradiation.

Gemcitabine

Gemcitabine (2',2'-difluorodeoxycytidine,Gemzar) is a deoxycytidine analogue capable of inhibiting DNA replication and repair. After a phase I study,' gemcitabine was evaluated in a multicenter trial of 44 patients with advanced .~ only five objective responses were documented, the pancreatic ~ a n c e rAlthough investigators noted frequent subjective symptomatic benefit, often in the absence of an objective tumor response. Toxicity appeared minor and included myelosuppression, particularly thrombocytopenia, as well as a flu-like syndrome and mild hemolytic-uremic syndrome. Based on these observations, two subsequent trials of gemcitabine in patients with advanced pancreatic cancer have been comtrial, gemcitabine was compared with 5-FU in pre~ l e t e d .In~ one ~ , ~randomized ~ viously untreated patients.42Patients treated with gemcitabine achieved modest

192

MILLER et a1

but statistically significant improvements in response rate and median survival compared with patients treated with 5-FU. In addition, more clinically meaningful effects on disease-related symptoms (pain control, performance status, weight gain) were observed with gemcitabine than with 5-FU. Similar systemic effects and demonstrable disease resvonses were documented in vatients who were treated with gemcitabine after experiencing disease progression while receiving 5-FU.53 Preliminary data demonstrate that gemcitabine is also a potent radiation sensitizer of human pancreatic cancer cells in vitro, supporting studies examining its .~~ studies suggest that the inhibitory effect of gemcitabine use in ~ i v oLaboratory on DNA synthesis (when combined with irradiation) is prolonged in tumor compared with normal tissues." This effect may provide a window of opportunity for the combination of gemcitabine and radiation therapy when delivered in a fractionated schedule. Such data provide the basis for the current phase I study in progress at MDACC. SYSTEMIC AGENTS OF LOW TOXICITY

Multimodality therapy has improved significantly local-regional tumor control, shifting the dominant pattern of treatment failure from local to distant (liver). This shift provides the rationale for combining systemic therapy with multimodality local-regional treatment schemas (Fig. 2). Selected agents of low toxicity that are currently entering clinical study are reviewed next. Inhibitors of Angiogenesis (TNP-470)

Folkman17has established that tumor growth is angiogenesis dependent. Angiogenesis is essential for tumors to grow larger than 1mm3,the maximal distance of oxygen transport and nutrient diffusion. In addition, angiogenesis must occur for metastasis formation and growth. As with many other tumor types, the growth and metastatic potential of pancreatic adenocarcinoma has been shown to depend on neovascularization, which is correlated with the expression of vascular endothelial growth factor (VEGF). Experimental data suggest that VEGF acts via a paracrine mechanism on specific receptors present on the surface of endothelial cells," thereby stimulating their proliferation, migration, and organization in vitro and hence angiogenesis in vivo. Recent data suggest that upregulation of VEGF may occur due to activating mutations of the uas oncogene48;specific point mutations at codon 12 of the K-uas oncogene occur in 75% to 90% of pancreatic adenocarcinoma specimens3 TNP-470, an angiogenesis inhibitor, is a synthetic analogue of fumagillin, a natural product secreted by the fungus Aspeugillus fumigatus fresenius. In vitro, TNP-470 inhibits endothelial cell proliferation, migration, and capillary tube formation in response to basic fibroblast growth factor (bFGF) at doses far less than those required for inhibition of these activities in tumor cells.30TNP-470 has been shown to inhibit (in vivo) the growth of Lewis lung carcinoma, B16 melanoma, M5076 reticulum cell sarcoma, colon 26 carcinoma, and Engelbreth-Holm sarcoma.30Inhibition of tumor growth was seen both when mice were treated early after innoculation and when tumor volume exceeded 100 mm3. Animals treated with effective doses of TNP-470 for prolonged periods of time (>I00 days) did not exhibit hair loss, intestinal disturbance, or infection. In a recent report of the use of TNP-470 in an orthotopic nude mouse model of human colon cancer, TNP-470 virtually eliminated the development of liver

NEOADJUVANT CHEMORADIATION FOR ADENOCARCINOMA OF THE PANCREAS

193

m e t a ~ t a s e sIn . ~a~rabbit liver metastasis model, infusion of TNP-470 into the portal vein after partial hepatectomy decreased tumor growth in the liver without impairing liver regeneration or wound healing61 Treatment of patients with TNP470 represents an attractive therapeutic alternative of low systemic toxicity. A pilot trial of preoperative and postoperative TNP-470 combined with chcmoradiation has been proposed for patients with potentially resectable adenocarcinoma of the pancreatic head. Matrix Metalloproteinase Inhibitors

Matrix metalloproteinases (MMP) are a family of enzymes that collectively are capable of degrading all of the components of the extracellular matrix. They are termed metallopvoteinases because they require a zinc molecule in their catalytic binding site. These enzymes are present in normal tissue and play important roles in such processes as wound healing, pregnancy and parturition, and bone resorpt i ~ nHigh . ~ ~concentrations of metalloproteinases have been detected in tumors in which they are believed to be important in invasion and metastasis by degrading the extracellular matrix.24A class of endogenous inhibitors of MMP termed tissue inhibitors of metallopvoteinases (TIMPs) bind to and inactivate both precursor and active MMP.43The extent of matrix degradation depends on the balance between active and inactive MMP as well as the ratio of the MMP to their inhibitors. Studies that use TIMPs demonstrated the utility of matrix metalloproteinase inhibitors (MMPI) in reducing tumor growth and metastases in animal model^.^,^^ However, TIMPs cannot be produced as an oral formulation, and mass production of the compounds would be prohibitively expensive. Therefore, interest has focused on the production and testing of synthetic inhibitors of MMP. The first synthetic MMPI to be developed was Batimastat. Batimastat has no effect on cancer cells in vitro; therefore, it is unlikely to be acting by direct cytotoxic mechanism^.^^ In animal studies, Batimastat has been able to reduce significantly ~,~~ to its low primary tumor growth, local-regional invasion, and m e t a s t a s e ~ .Due bioavailability when administered orally, clinical trials with Batimastat were limited to intraperitoneal or intrapleural injections in patients with malignant pleural effusion or ascites. In contrast, Marimastat, the first of a newer class of MMPI, has good bioavailability when administered orally. Marimastat is a low-molecular weight peptide that inhibits MMP by chelating the zinc molecule in the catalytic site of the enzyme. Because Marimastat is not a cytoxic agent, early clinical trials defined dose response as disease stabilization using changes in tumor antigen levels as surrogate markers of biologic activity. Marimastat has been well tolerated with dose-limiting toxicities that consist of musculoskeletal symptoms such as pain and tenderness. Clinical trials of Marimastat are currently underway in pancreatic cancer patients, and we are participating in a multicenter study that examines the use of postoperative adjuvant Marimastat after pancreaticoduodenectomy. Perillyl Alcohol

The monoterpene perillyl alcohol is a nontoxic, naturally occurring substance derived from lavender and other plants. In animal studies, perillyl alcohol has been shown to be an effective chemotherapeutic and chemopreventive agent.25,41 Perillyl alcohol was believed to inhibit tumor cell growth by decreasing total cellular vas levels.z9However, recent cell culture studies involving human pancreatic cancer failed to show any inhibition of ras function despite induction of growth

194

MILLER et a1

i n h i b i t i ~ nStudies .~~ using a chemically induced liver tumor model suggest that perillyl alcohol may enhance tumor cell apoptosis by increasing the ability of the tumor cell to activate and respond to transforming growth factor-P.41In vitro, treatment of human pancreatic cancer cells with perillyl alcohol results in a dosedependent decrease in proliferati~n.~~ Perillyl alcohol also inhibits anchorage independent growth in human pancreatic cell lines and restores contact inhibition, as demonstrated by a decrease in the cell density of cultures grown to s a t ~ r a t i o n . ~ ~ Recent studies using hamster and human pancreatic cancer cells demonstrated induction of apoptosis by perillyl alcohol associated with an increase in the expression of the apoptosis-promoting protein Bak.lo In vivo studies demonstrated a significant growth inhibition of heterotopic pancreatic tumors with no observable toxicity to the animals.I0Preclinical studies have shown perillyl alcohol to be an effective, nontoxic alternative to current chemotherapeutic regimens for pancreatic cancer establishing the basis for ongoing phase I clinical trials. Tamoxifen Tamoxifen was first described as an estrogen antagonist, competing with estrogen for binding sites in target tissues. The demonstration of estrogen binding by normal and neoplastic pancreatic cells, as well as evidence that pancreatic tumors may be hormonally sensitive, led to the clinical use of tamoxifen in patients with pancreatic cancer. Two studies have demonstrated a survival advantage for ~,~~ cells contain patients with pancreatic cancer who received t a m ~ x i f e n .Pancreatic a 4s protein that binds estradiol with high capacity but low affinity.47Boctor et a16 subsequently identified an accessory coligand that is necessary for high affinity binding of estrogen to the pancreatic receptor. Discrepant estrogen binding characteristics of pancreatic cancer cells may be due to the variable expression of the coligand. More recent studies have shown that at high doses (>5 yM), tamoxifen has activity that is not mediated by the estrogen receptor.49In vitro studies with pancreatic cancer cell lines demonstrated a dose-dependent decrease in proliferation with tamoxifen treatment.50This treatment effect was seen even in cell lines that do not contain high affinity estrogen receptors and was due to the induction of Go/G, cell cycle arrest. However, the inhibition of proliferation did not become statistically significant until tamoxifen concentrations reached the 4 to 6 pM range. In a recent phase I study of high-dose tamoxifen given with cisplatin in patients with metastatic melanoma, the maximum tolerated dose of tamoxifen was 240 mg/d, which corresponded to a serum level of 1 pM.39Another study administered tamoxifen at 480 mg/d for 6 days and achieved serum levels in the 3 to 3.5 PM range with acceptable toxicity.60What remains to be determined is the concentration of tamoxifen achieved in pancreatic tumors with standard or high-dose therapy. In summary, a subgroup of patients do benefit from tamoxifen therapy at the standard dose, possibly due to the expression of high affinity estrogen receptors. High-dose tamoxifen is effective against pancreatic cancer cells in vitro, and perhaps with higher dose therapy more patients may benefit from tamoxifen treatment. References 1. Abbruzzese J, Grunewald R, Weeks E, et al: A phase I clinical, plasma and cellular pharmacology study of gemcitabine. J Clin Oncol9:491,1991

NEOADJUVANT CHEMORADIATION FOR ADENOCARCINOMA OF THE PANCREAS

195

2. Allema J, Reinders M, van Gulik T, et al: Portal vein resection in patients undergoing pancreatoduodenectomy for carcinoma of the pancreatic head. Br J Surg 81:1642, 1994 3. Almoguera C, Shibata D, Forrester K, et al: Most human carcinomas of the exocrine pancreas contain mutant c-K-uas genes. Cell 53:549,1988 4. Alvarez 0, Carmichael D, DeClerk Y: Inhibition of collagenolytic activity and metastasis of tumor cells by a recombinant human tissue inhibitor of metalloproteinases. J Natl Cancer Inst 82:589, 1990 5. Bakkevold K, Petterson A, Arnesjo B, et al: Tamoxifen therapy in unresectable adenocarcinoma of the pancreas and the papilla of Vater. Br J Surg 77:725, 1990 6. Boctor A, Band P, Grossman A: Requirement for an accessory factor for binding [3H]estradiolto protein in the cytosol fraction of rat pancreas. Proc Natl Acad Sci U S A 78:5648, 1981 7. Casper E, Green M, Kelsen D, et al: Phase I1 trial of gemcitabine (2',2'-difluorodeoxycytidine) in patients with adenocarcinoma of the pancreas. Invest New Drugs 12:29,1994 8. Chirivi R, Garofalo A, Crimmin M: Inhibition of the metastatic spread and growth of B16-BL6 murine melanoma by a synthetic matrix metalloproteinase inhibitor. Int J Cancer 58:460, 1994 9. Cortes J, Pazdur R: Docetaxel. J Clin Oncol13:2643,1995 10. Crowell P, Stayrook K, McKinzie J, et al: Induction of the apoptosis-promoting protein Bak by perillyl alcohol in pancreatic ductal adenocarcinoma relative to the untransformed ductal cells [abstract 6471. Proceedings of the American Association for Cancer Research 38:97,1997 11. deVries C, Escobido J, Ueno J, et al: The fms-like tyrosine kinase, a receptor for vascular endothelial growth factor. Science 255:989,1992 12. Evans DB, Abbruzzese J, Cleary K, et al: Rapid-fractionation preoperative chemoradiation for malignant periampullary neoplasms. J R Coll Surg Edinb 40:319, 1995 13. Evans DB, Abbruzzese JL, Lee JE, et al: Preoperative chemoradiation and pancreaticoduodenectomy for adenocarcinoma of the pancreas [abstract 131. h Programs and Abstracts of the American Society of Clinical Oncology, Dallas, TX, 1994, p 226 14. Evans DB, Abbruzzese JL, Rich TA: Cancer of the pancreas. In DeVita VT Jr, Hellman S, Rosenberg SA (eds): Cancer: Principles and Practice of Oncology, ed 5. Philadelphia, JB Lippincott, 1977, p 1054 15. Evans DB, Lee JR, Pisters PWT: Pancreaticoduodenectomy (Whipple operation) and total pancreatectomy for cancer. In Nyhus LM, Baker RJ, Fischer JF (eds):Mastery of Surgery, ed 3. Boston, Little, Brown, 1996, p 1233 16. Evans DB, Rich T, Byrd D, et al: Preoperative chemoradiation and pancreaticoduodenectomy for adenocarcinoma of the pancreas. Arch Surg 127:1335,1992 17. Folkman J: What is the evidence that tumors are angiogenesis dependent? J Natl Cancer Inst 82:4, 1989 18. Foo M, Gunderson L, Nagorney D, et al: Patterns of failure in grossly resected pancreatic ductal adenocarcinoma treated with adjuvant irradiation rt 5 fluorouracil. Int J Radiat Oncol Biol Phys 26:483, 1993 19. Freeny P, Traverso L, Ryan J: Diagnosis and staging of pancreatic adenocarcinoma with dynamic computed tomography. Am J Surg 165:600,1993 20. Fuhrman G, Charnsangavej C, Abbruzzese J, et al: Thin-section contrast-enhanced computed tomography accurately predicts the resectability of malignant pancreatic neoplasms. Am J Surg 167:104,1994 21. Fuhrman G, Leach S, Staley C, et al: Rationale for en bloc vein resection in the treatment of pancreatic adenocarcinoma adherent to the superior mesenteric-portal vein confluence. Ann Surg 223:154,1996 22. Gastrointestinal Tumor Study Group: A multi-institutional comparative trial of radiation therapy alone and in combination with 5-fluorouracil for locally unresectable pancreatic carcinoma. Ann Surg 189:205,1979 23. Gastrointestinal Tumor Study Group: Further evidence of effective adjuvant combined radiation and chemotherapy following curative resection of pancreatic cancer. Cancer 59:2006, 1987 24. Gress T, Muller-Pillasch F, Lerch M, et al: Expression and in-situ localization of genes coding for extracellular matrix proteins and extracellular degrading proteases in pancreatic cancer. Int J Cancer 62:407, 1995

196

MILLER et a1

25. Haag J, Gould M: Mammary carcinoma regression induced by perillyl alcohol, a hydroxylated analog of limonene. Cancer Chemother Pharmacol34:477,1994 26. Harrison L, Klimstra D, Brennan M: Isolated portal vein involvement in pancreatic adenocarcinoma: A contraindication to resection? Ann Surg 224:342, 1996 27. Hittelman W, Fujii W, Hunter N, et al: Identification of a window of therapeutic opportunity for the combination of gemcitabine and radiation. Proceedings of the American Association for Cancer Research 321973,1996 28. Hoffman JP, Weese JL, Solin LJ, et al: Preoperative chemoradiation for patients with resectable pancreatic adenocarcinoma: An Eastern Cooperative Oncology Group (ECOG) phase I1 study [abstract 141. In Programs and Abstracts of the American Society of Clinical Oncology. Los Angeles, CA, 1995, p 201 29. Hohl R, Lewis K: Differential effects of monoterpenes and lovastatin on RAS processing. J Biol Chem 29:17508,1995 30. Ingber D, Fujita T, Kishimoto S, et al: Synthetic analogues of fumagillin that inhibit angiogenesis and suppress tumor growth. Nature 348:555,1990 31. Kalser M, Ellenberg S: Pancreatic cancer: Adjuvant combined radiation and chemotherapy following curative resection. Arch Surg 120:899, 1985 32. Karison J, Borg-Karlson A, Unelius R, et al: Inhibition of tumor cell growth by monoterpenes in vitro: Evidence of a RAS-independent mechanism of action. Anticancer Drugs 7422,1996 33. Khokha R: Suppression of the tumorigenic and metastatic abilities of murine B16-F10 melanoma cells in vivo by the overexpression of the tissue inhibitor of the metalloproteinase-1. J Natl Cancer Inst 86:299,1994 34. Konno H, Tanaka T, Kanai T, et al: Efficacy of an angiogenesis inhibitor, TNP-470, in xenotransplanted human colorectal cancer with high metastatic potential. Cancer (suppl) 77:1736, 1996 35. Lawrence T, Chang E, Hahn T, et al: Radiosensitization of pancreatic cancer cells by 2',2'-difluoro-2'-deoxycytidine.Int J Radio1 Oncol Biol Physics 342367,1996 36. Leach SD, Lee JE, Charnsangavej C, et al: Patient survival following pancreaticoduodenectomy with resection of the superior mesenteric-portal vein confluence for adenocarcinoma of the pancreatic head. Brit J Surg, in press 37. Lillemoe K, Cameron J, Yeo C, et al: Pancreaticoduodenectomy: Does it have a role in the palliation of pancreatic cancer? Ann Surg 223:718,1996 38. Lowy AM, Lee JE, Pisters PWT, et al: Prospective, randomized trial of octreotide to prevent pancreatic fistula following pancreaticoduodenectomy for malignant disease. Ann Surg, in press 39. McClay E, McClay M, Jones J, et al: A phase I study of high dose tamoxifen and weekly cisplatin in patients with metastatic melanoma. Cancer 79:1037,1997 40. Milas L, Hunter N, Mason K, et al: Role of reoxygenation in induction of enhancement of tumor radioresponse by paclitaxel. Cancer Res 55:3564,1995 41. Mills J, Chari R, Boyer I, et al: Induction of apoptosis by the monoterpene perillyl alcohol. Cancer Res 55:979,1995 42. Moore M, Andersen J, Burris H, et al: A randomized trial of gemcitabine versus 5-FU as first-line therapy in advanced pancreatic cancer [abstract 141. In Programs and Abstracts of the American Society Clinical Oncology, 1995, p 199, Los Angeles, CA 43. Murphy G, Willenbrock F: Tissue inhibitors of matrix metalloproteinases. Methods Enzymol248:496,1995 44. Nitecki S, Sarr M, Colby T, et al: Long-term survival after resection for ductal adenocarcinoma of the pancreas: Is it really improving? Ann Surg 221:59,1995 45. Parsons S, Watson S, Brown P, et al: Matrix metalloproteinases. Br J Surg 84:160, 1997 46. Pisters PWT, Abbruzzese JL, Cleary KR, et al: Pilot study of rapid-fractionation preoperative chemoradiation pancreaticoduodenectomy (PD) and intraoperative radiation therapy (IORT) for resectable pancreatic adenocarcinoma [abstract 161. In Programs and Abstracts of the American Society of Clinical Oncology, 1997, p 1076, Los Angeles, CA 47. Pousette A, Carlstrom K, Skoldefors H, et al: Purification and partial characterization of a 17b-estradiol binding macromolecule in the human pancreas. Cancer Res 42:633,1982 48. Rak J, Mitsuhashi Y, Bayko L: Mutant vas oncogenes upregulate VEGF/VPF expression: Implications for induction and inhibition of tumor angiogenesis. Cancer Res 55:4575, 1995

NEOADJWANT CHEMORADIATION FOR ADENOCARCINOMA OF THE PANCREAS

197

49. Reddel RR, Murphy LC, Hall RE, et a1: Differential sensitivity of human breast cancer cell lines to the growth-inhibitory effects of tamoxifen. Cancer Res 45;1525,1985 50. Robinson E, Evans D, Chiao P, et al: Mechanism of tamoxifen induced growth inhibition in human pancreatic cancer [abstract 40351. Proceedings of the American Association for Cancer Research 38:601, 1997 51. Robinson E, Lee J, Pisters P, et al: Reoperative pancreaticoduodenectomy for periampullary carcinoma. Am J Surg 172:432,1996 52. Roder J, Stein H, Siewert J: Carcinoma of the periampullary region: Who benefits from portal vein resection? Am J Surg 171:170,1996 53. Rothenberg M, Burris H 111, Andersen J, et al: Gemcitabine: Effective palliative therapy for pancreas cancer patients failing 5-FU [abstract 141. In Programs and Abstracts of the American Society Clinical Oncology, 1995, p 198, Los Angeles, CA 54. Rowinsky E, Donehower R: Paclitaxel. N Engl J Med 332:1004,1995 55. Safran H, King T, Choy H, et al: Paclitaxel and concurrent radiation for locally advanced pancreatic and gastric cancer: A phase I study. J Clin Oncol 15:901,1997 56. Spitz F, Abbruzzese J, Lee J, et al: Preoperative and postoperative chemoradiation strategies in patients treated with pancreaticoduodenectomy for adenocarcinoma of the pancreas. J Clin Oncol15:928,1997 57. Staley C, Cleary K, Abbruzzese J, et al: Need for standardized pathologic staging of pancreaticoduodenectomy specimens. Pancreas 12:373,1996 58. Staley C, Lee J, Cleary K, et al: Preoperative chemoradiation, pancreaticoduodenectomy, and intraoperative radiation therapy for adenocarcinoma of the pancreatic head. Am J Surg 171:118,1996 59. Stark M, Burke Y, McKinzie J, et al: Chemotherapy of pancreatic cancer with the monoterpene perillyl alcohol. Cancer Lett 96:15,1995 60. Stuart N, Philip P, Harris A, et al: High-dose tamoxifen as an effector of etoposide cytotoxicity: Clinical effects and in vitro assessment in p-glycoprotein expressing cell lines. Br J Cancer 66:833,1992 61. Tanaka H, Taniguchi H, Mugitani T, et al: Angiogenesis inhibitor TNP-40 prevents implanted liver metastases after partial hepatectomy in an experimental model without impairing wound healing. Br J Surg 83:1444, 1996 62. Tepper J, Nardi G, Sutt H: Carcinoma of the pancreas: Review of MGH experience from 1963 to 1973: Analysis of surgical failure and implications for radiation therapy. Cancer 37:1519, 1976 63. Trede M, Schwa11 G, Saeger H: Survival after pancreatoduodenectomy: 118 consecutive resections without an operative mortality. Ann Surg 211:447, 1990 64. Wang X, Fu X, Brown P, et al: Matrix metalloproteinase inhibitor BB-94 (Batimastat) inhibits human colon tumor growth and spread in a patient-like orthotopic model in nude mice. Cancer Res 54:4726,1994 65. Warshaw A, Gu Z, Wittenberg J, et al: Preoperative staging and assessment of resectability of pancreatic cancer. Arch Surg 125:230, 1990 66. Watson S, Morris T, Robinson G, et al: Inhibition of organ invasion by the matrixmetalloproteinase inhibitor Batimastat (BB-94) in two human colon carcinoma metastasis models. Cancer Res 55:3629, 1995 67. Whittington R, Bryer M, Haller D, et al: Adjuvant therapy of resected adenocarcinoma of the pancreas. Int J Radiat Oncol Biol Phys 21:1137,1991 68. Willett C, Lewandrowski K, Warshaw A, et al: Resection margins in carcinoma of the head of the pancreas: Implications for radiation therapy. Ann Surg 217:144, 1993 69. Wong A, Chan A: Survival benefit of tamoxifen therapy in adenocarcinoma of the pancreas: A case-control study. Cancer 71:2200,1993 70. Yeo C, Cameron J, Lillemoe K, et al: Pancreaticoduodenectomy for cancer of the head of the pancreas: 201 patients. Ann Surg 221:721,1995

Adclress reprint requests to Douglas B. Evans, MD Department of Surgical Oncology Box 106 The University of Texas M. D. Anderson Cancer Center 1515 Holcombe Boulevard Houston, TX 77030