Cancer Therapy Vol 1, 283-291, 2003.

A Phase I safety and imaging study using radiofrequency ablation (RFA) followed by ¹³¹I-chTNT-1/B radioimmunotherapy adjuvant treatment of hepatic metastases

Research Article

Peter M. Anderson^1,2,3*, Gregory A. Wiseman^3,4, Bradley D. Lewis⁵, J. William Charboneau⁵, William L. Dunn⁴, Susan P. Carpenter⁶, Terrence Chew⁶

Dept. of Pediatrics¹, Internal Medicine (Hematology)², and Mayo Clinic Comprehensive Cancer Center³, Nuclear Medicine⁴, and Radiology⁵, Mayo Clinic, Rochester MN, Peregrine Pharmaceuticals⁶, Tustin CA

__________________________________________________________________________________

Correspondence: Peter M. Anderson MD, PhD, Mayo Clinic, 200 First St. SW, Rochester MN 55905, Phone: 507-284-3442; Fax 507-284-0727; e-mail: anderson.peter@mayo.edu

Key Words: cancer, monoclonal antibody, liver metastases, radioimmunotherapy, radiofrequency ablation

Abbreviations: Radiofrequency ablation (RFA); Human anti-chimeric antibody (HACA)

Received: 27 October 2003; Accepted: 1 December 2003; electronically published: December 2003

Summary

Biologic therapy of solid tumors with monoclonal antibodies has been difficult due to poor antibody localization and heterogeneous expression of target antigens. TNT-1/B is a murine-human monoclonal antibody that recognizes a DNA/histone 1 epitope and concentrates in necrotic tissue and is now available as genetically engineered murine-human chimeric (ch) construct. Thus, the binding of chTNT-1/B to areas of necrosis in tumors has potential to treat a very wide variety of cancers. Since radiofrequency ablation (RFA) of tumor nodules reliably produces 1-5 cm zones of >99% necrotic tissue, RFA may create abundant binding sites for chTNT-1/B, regardless of initial tumor histology. Study design and results: Hepatic distribution and safety of iv ¹³¹I-chTNT-1/B (Peregrine Pharmaceuticals, Tustin CA) given after RFA of hepatic metastases was evaluated in six patients. Five of 6 had metastatic disease confined to the liver. Diagnoses included carcinoid, leiomyosarcoma, colon adenocarcinoma and islet cell carcinoma. RFA of metastases was done in a standard manner under ultrasound guidance using the RITA device. Liver function tests were monitored sequentially after RFA Patients were eligible to receive the ¹³¹I radiolabeled antibody when AST and ALT were £ CTC grade 3 and also decreased on 2 successive days and performance was acceptable for the procedure. Patients that had the RFA procedure done percutaneously received the ¹³¹I -chTNT-1/B at 3, 3, 4, and 6 days after the procedure. The two patients having RFA intraoperatively received the ¹³¹I-chTNT-1/B somewhat later (6 and 10 days). Patients received 0.35 mCi/Kg or 0.71 mCi/Kg ¹³¹I-chTNT-1/B; total doses ranged between 22 and 55 mCi. Infusions were given over 30 minutes; no infusion toxicity was seen. Between 12 to 29% (Mean 28.1 +/- 4.0%) of an injected dose concentrated in the liver. Gamma camera imaging confirmed selective and avid targeting of radioisotope to areas of RFA within the liver. No significant adverse events were observed. Conclusion: The chTNT-1/B construct has excellent potential to become useful after RFA. Zones of necrosis that facilitate ¹³¹I-chTNT-1/B antibody binding were probably created after RFA. A further improvement in patient convenience and specific targeting with this promising immunoconjugate may also be possible using direct antibody injection at the end of the RFA procedure into the zone of necrosis using temperature monitoring.

I. Introduction

Successful biologic therapy of cancer using monoclonal antibodies against solid tumors has been difficult. In humans, only about 0.001-0.01% of an injected dose of antibody per gram of tumor is delivered, resulting in radiation doses inadequate for the task of elimination of commonly encountered adenocarcinomas (i.e. >5000 cGy; Goldenberg, 2002, 2003). Other obstacles to radioimmunotherapy of solid tumors include the need to spare radiosensitive normal organs such as the bone marrow, and heterogeneity of target antigen expression and density (Weiner, 1999; von Mehren et al, 2003). Nevertheless, the field has experienced revival since successful therapy for hematologic malignancies including non-Hodgkins lymphoma using monoclonal antibodies such as anti-CD20 (Rituxan) and ⁹⁰Y-anti-CD20 (Zevalin; Witzig et al, 2002).

Since target: non-target ratios of antibody binding determine the imaging and non-specific dose-limiting toxicities, a variety of strategies are being tested to improve specificity and efficacy of antibody therapy against non-hematologic cancer. These include affinity-enhancement systems with bi-specific antibodies to separate antibody targeting and the delivery of the radioactive payload to the site of neoplasia (Goldenberg, 2002, 2003). Other strategies to increase therapeutic index also include use chemical modification to improve pharmacokinetics (Sharifi et al, 1998) and intact or fragments of interleukin-2 cytokine-antibody fusion proteins to improve vascular permeability (Hu et al, 1996, 2003; Hornick et al, 1999; Carnemolla et al, 2002; Epstein et al, 2003). Although biotin is more commonly used in two and three step pre-targeting methods with streptavidin or avidin, this chemical modification procedure also lowers the ionic charge of the antibody to decrease non-specific binding in tissues and blood. Thus chTNT-1/B had better performance than chTNT-1 in vivo including better tumor uptake, less non-specific uptake in normal tissues, faster clearance profile (Sharifi et al, 1998).  At the present time treatment of metastatic colorectal cancer with radioimmunotherapy has been most successful in the adjuvant setting or in small volume disease (Behr et al, 2002).

Another means to increase therapeutic index is to use an antibody that spares normal tissue and binds necrotic tumor tissue (Epstein et al, 1988; 1991; Chen et al, 1989; 1990; Miller et al, 1993). TNT-1 is an antibody that binds a 22 kilodalton nuclear protein associated with the DNA/histone H1 (Epstein et al, 1988). In pre-clinical murine models, however, biodistribution of TNT-1 to human xenografts was similar to other antibodies against solid tumors- about 2% of an injected dose/gram tumor tissue. One means to overcome low accretion into tumor is to directly inject the antibody into a tumor cavity to reduce systemic toxicity. Results of a phase I trial of ¹³¹I- chTNT-1/B in brain tumors indicated that 20-40 mCi could deliver 700-13,000 cGy to the tumor with 34 +/- 9% dose retention at 24 hours, and a half-life of about 46 +/- 16 hours (Patel et al, 1999). A recently completed phase II trial in brain tumors with ¹³¹I-chTNT-1/B showed that after 8.6 to 52 mCi injected locally, the calculated tumor dose was 1641 to 11,171 cGy (Wessels et al, 2001). Because of the critical location of antibody infusion in the brain tumor trials of chTNT-1/B, it is not clear in many cases whether adverse events were related to underlying brain tumor, induction of necrosis, or radiolabeled antibody.

Since the target epitope of chTNT-1/B, histone H1/DNA, is available for antibody binding only in necrotic tissue (Epstein et al, 1988), physical means to increase necrosis within a tumor may possibly enhance chTNT-1/B targeting to areas of tumor. Percutaneous ethanol injection is one means to produce necrosis and destroy tumor nodules (Livraghi et al, 1995; Virag et al, 1997; Isozaki et al, 1999; Meloni et al, 2001; Lewis et al, 2002). Radiofrequency ablation (RFA) which uses thermal energy at the tip of a radiofrequency probe is another reliable means to induce 1-5 cm zones of necrosis in hepatic tumor nodules (Bilchik et al, 1999; Wood et al, 2000; Izzo et al, 2001; Charboneau et al, 2002; Curley and Izzo, 2002a, b; Dick et al, 2002; Livraghi and Meloni, 2002; Nordlinger and Rougier, 2002; Seidenfeld et al, 2002a, b; Shibata et al, 2002; Garcea et al, 2003; Lau et al, 2003; Numata et al, 2003). Thus, RFA should markedly increase sites of chTNT-1/B in tumor nodules compared to the amount of necrosis normally present in macroscopic tumor nodules. Therefore, we conducted a limited, phase I study to determine safety and hepatic distribution of ¹³¹I-chTNT-1/B given intravenously after RFA of hepatic metastases. ¹³¹I-chTNT-1/B is a binotinylated⁴⁰, chimeric antibody with radioiodine attached as shown in Figure 1.

II. Materials and methods

A. Materials

¹³¹I-chTNT-1/B was supplied by Peregrine, Inc (Tustin CA).

B. Methods

All patients were informed of indications, risks, and alternatives and signed informed consent. The protocol was approved by the Mayo Clinic Institutional Review Board (IRB O-368-01). Patients with any type of cancer that would have RFA of at least 1 hepatic metastasis were eligible. Other criteria included: no more than two prior chemotherapy regimens, no prior radiotherapy to the liver, > 14 years of age. Karnofsky Performance Status ³ 70%, adequate CBC as evidenced by absolute neutrophil count > 1,000/mm³ (£ Grade 2 toxicity), platelets > 75,000/mm³ (£ Grade 2 toxicity), hemoglobin > 8.0 g/dL (£ Grade 2 toxicity, and adequate renal function (Cr<3 x ULN; 3.6 mg/dL for males; 2.7 mg/dL for females; £ Grade 2 toxicity). Adequate liver function before RFA was total bilirubin £ 1.5 x ULN; AST, ALT £2.5 x ULN (i.e. <Grade 2 toxicity) and after RFA (within 3 to10 days): total bilirubin £ 3 x ULN; AST, ALT £20 x ULN (i.e. £Grade 3 toxicity).

Patients were offered percutaneous or intraoperative RFA as per recommendation of medical oncology and/or surgery consultants before study entry. RFA was done in a standard manner using the RITA device. Ablation of hepatic lesions was confirmed by CT scan. ¹³¹I-chTNT-1/B antibody infusion was done 3-10 days after RFA using standard radiation safety precautions when performance was >70%, patients had received at least 2 days of SSKI to protect the thyroid (4 gtt po TID x 14 days), and both AST and ALT were £ CTC grade 3 and had decreased on two consecutive days.

Pretreatment prior to ¹³¹I-chTNT-1/B antibody infusion consisted of 650 mg acetaminophen and diphenhydramine 50 mg and 250 cc normal saline over 1 hour. ¹³¹I-chTNT-1/B was diluted to 50 mL with normal saline and 5% human serum albumin and infused by nuclear medicine personnel over 30 minutes using a lead shielded syringe pump.

Total Body Retention Survey (G-M readings) were done daily x 3 and side effects were monitored. SPECT imaging was done on day 3, 4, or 5 after antibody infusion. CBC, liver function tests (AST, ALT, biliribin) were monitored twice weekly x 4 weeks (e.g. day 3, 7, 10, 14, 17, 21, 24, 28). Creatinine was monitored weekly.

Figure 1. ¹³¹I-chTNT-1/B

Table 1. Patient Characteristics Prior to RFA of Hepatic Metastases

Liver Metastasis Karnofsky
Age	Sex	Diagnosis	Size (cm^2)	Segment	Performance score
20	M	Carcinoid	1.0	8	100
			1.3	8
65	M	leiomyosarcoma	44.4	4A	90
			2.0	5/8
74	M	adenocarcinoma	1.6	8	100
81	M	adenocarcinoma	16.8	3	90
			2/4	17.2
			3.2	8
			5.6	4E
			6.2	6
65	F	islet cell	<5	--	100
49	F	islet cell	0.8	6/7	100
			1.4	6/7

Figure 2. RFA, then ¹³¹I-chTNT-1/B study schema.

Follow-up visit at 8 weeks included physical exam, discussion of side effects, determination of Karnofsky Performance Status, blood tests (TSH, CBC, chemistry panel), urinalysis, and imaging of the liver metastases.

III. Results

Table 1 details patient characteristics including segmental location of indicator lesions in the liver. Figure 1 depicts the structure of ¹³¹I-chTNT-1/B, the agent to be tested. Figure 2 shows a schematic diagram of the protocol design. RFA of hepatic lesions was done in a standard manner using ultrasound imaging (Figure 3, 4). CT scans confirmed ablation in all patients. Patients had close monitoring of liver function after RFA and all were eligible to receive ¹³¹I-chTNT-1/B when LFT had returned to grade 3 CTC and performance was adequate (Karnofsky >70%). Doses infused are detailed in Table 2.

Post procedural pain requiring overnight hospitalization occurred in one of the four patients having RFA percutaneously. Median duration from RFA to infusion of ¹³¹I-chTNT-1/B in the percutaneous RFA group was 5 days.

In the two patients that had RFA done during hepatic surgery, hepatic function as assessed by AST and ALT recovered more quickly than performance and it was 6 and 10 days before ¹³¹I-chTNT-1/B was infused. As expected AST, ALT progressively decreased after RFA and became CTC grade 2 (2-5 x ULN) or 3 (>5-20 x ULN) in all patients (Table 3).

Bilirubin increase did not ever exceed grade 1 CTC criteria (>1.5 x ULN) Peak elevation of AST and ALT was always on the first determination after RFA with subsequent steady decline. There were no side effects nor significant changes in liver function tests associated with the ¹³¹I-chTNT-1/B antibody infusion (Table 3). Targeting of ¹³¹I-to areas of the liver that previously had RFA was confirmed by radioscintigraphy (Figures 5, 6, 7).

Other parameters that remained normal during period of observation included renal function as determined by serum creatinine and urinalysis. Hematologic parameters did not change with exception of one very mild case of leukopenia (wbc 2000, ANC 1000; CTC grade 2) noted at week 8 which spontaneously resolved. One patient had increased TSH noted at follow-up visit but did not require hormone replacement. Human anti-chimeric antibody (HACA) titers were negative at both time points tested in 6 of 6 patients.

Patients had imaging done 1, 2 and 3-5 days after antibody infusion. Approximately 30% of an injected dose of ¹³¹I-TNT-1/B became localized to the liver (Table 4). SPECT confirmed relative selectivity of ¹³¹I -TNT-1/B localization to hepatic RFA sites (Figures 5, 6, 7). Examples of gamma camera imaging showing the focally increased hepatic localization of the ¹³¹I radioisotope to RFA sites are apparent in Figures 5, 6, and 7. This occurred for all histologies. Pixel analysis showed a RFA target: background liver ratio average of 2.9 3 days after ¹³¹I -TNT-1/B infusion of (Table 5). Because of volume imaging considerations it was not possible to perform dimetric calculations to estimate absorbed radiation dose of a nodule compared to surrounding normal tissue.

Figure 3. Radiofrequency ablation (RFA) catheter

Figure 4. Ultrasound guided Radiofrequency ablation (RFA) of hepatic metastases

Table 2. Doses of ¹³¹I-chTNT-I/B

	Dose level mCi/kg	Actual mCi/kg	Total Dose (mCi)	Days s/p RFA
Patient
1	0.35	0.35	34.3	6
2	0.35	0.35	22.1	6
3	0.35	0.35	31.6	4
4	0.71	0.50	35.2	3
5	0.71	0.56	44.6	10
6	0.71	0.71	55.0	6

Table 3. Liver Function Tests after RFA and ¹³¹I-chTNT-1/B

	Patient 1	Patient 2		Patient 3		Patient 4
	Patient 5	Patient 6
Visit*	Bili/AST/ALT	Bili/AST/ALT	Bili/AST/ALT	Bili/AST/ALT	Bili/AST/ALT	Bili/AST/ALT
2	0.3/108/216	0.6/241/ND	0.8/151/127	1.1/564/486	0.6/96/155	0.5/43/165
2.1	0.4/69/171	0.5/165/ND	0.7/51/91	1.0/280/318	0.3/60/109	0.5/38/133
2.2	0.3/47/131	0.9/103/ND
Post**	0.1/23/22	0.4/38/ND	0.4/20/26	0.1/43/69	0.3/19/23	0.5/20/39

chTNT-1/B

*visit 2 is 1 day after RFA, 2.1 is 1-2 days later, and 2.2 is 2-3 days after RFA.

** Post is 3 days s/p infusion of ¹³¹I-TNT-1/B

Figure 5. Metastatic carcinoid. CT pre and SPECT s/p RFA, then ¹³¹I-TNT-1/B

Figure 6. Hepatic metastases of colon adenocarcinoma. SPECT imaging s/p RFA, then ¹³¹I-TNT-1/B

Figure 7. Hepatic metastases of colon adenocarcinoma. SPECT imaging s/p RFA, then ¹³¹I-TNT-1/B. Dose level 2; SPECT imaging

Table 4. Hepatic Distribution of TNT-1/B after RFA

Patient	Total dose injected (mCi)	Hours s/p Infusion	% injected dose in liver
1	34.3	97.1	34.6
2