Purpose: Humanized monoclonal antibody A33 (huA33) targets the A33 antigen which is expressed on 95% of colorectal cancers. A previous study has shown excellent tumor-targeting of iodine-131 labeled huA33 (131I-huA33). Therefore, we did a phase I dose escalation trial of 131I-huA33 radioimmunotherapy.

Experimental Designs: Fifteen patients with pretreated metastatic colorectal carcinoma each received two i.v. doses of 131I-huA33. The first was an outpatient trace-labeled “scout” dose for biodistribution assessment, followed by a second “therapy” dose. Three patients were treated at 20, 30, and 40 mCi/m2 dose levels, and six patients at 50 mCi/m2 to define the maximum tolerated dose.

Results: Hematologic toxicity was 131I dose-dependent, with one episode of grade 4 neutropenia and two episodes of grade 3 thrombocytopenia observed at 50 mCi/m2. The maximum tolerated dose was determined to be 40 mCi/m2. There were no acute infusion-related adverse events, and gastrointestinal toxicity was not observed despite uptake of 131I-huA33 in bowel. Seven patients developed pruritus or rash, which was not related to 131I dose. There was excellent tumor-targeting of 131I-huA33 shown in all patients. The serum T1/2β of 131I-huA33 was (mean ± SD) 135.2 ± 46.9 hours. The mean absorbed tumor dose was 6.49 ± 2.47 Gy/GBq. Four patients developed human anti-human antibodies. At restaging, 4 patients had stable disease, whereas 11 patients had progressive disease.

Conclusion: Radioimmunotherapy using 131I-huA33 shows promise in targeting colorectal tumors, and is deliverable at a maximum tolerated dose of 40 mCi/m2. Further studies of 131I-huA33 in combination with chemotherapy are planned.

Radioimmunotherapy is coming of age as a treatment modality for cancer. Patients with hematologic malignancies have been the first to benefit from treatment with radiolabeled monoclonal antibodies (mAb; refs. 1, 2). Yet, there is much potential for the application of radioimmunotherapy to epithelial malignancies as well. Colorectal cancer is a significant public health problem and is the leading cancer by incidence in Australia and the U.S. (3). Unfortunately, ∼50% of all patients develop metastatic disease, for which curative treatment is not yet available (4).

The A33 antigen is a 43 kDa membrane bound glycoprotein present on the basolateral surfaces of normal colon and small bowel epithelial cells in mice and humans (57). It is also homogeneously expressed in >95% of human colorectal cancers and ∼50% of gastric cancers (5). Approximately 50% of pancreatic cancers show heterogeneous expression. The A33 antigen is not detectably shed into the extracellular space or the bloodstream, in contrast to other tumor antigens such as carcinoembryonic antigen.

The A33 antigen has been sequenced and comprises three structural domains: an extracellular region containing two immunoglobulin-like domains, a hydrophobic transmembrane domain, and a highly polar intracellular tail (69). It is therefore regarded as part of a subfamily within the immunoglobulin superfamily, together with the transmembrane proteins CTX (corticothymocyte marker in Xenopus), CTM/CTH (mouse and human homologues of CTX) and CAR (Coxsackie and adenovirus receptor; ref. 7). Its structure is consistent with a putative role as a cell adhesion molecule or a novel cell surface receptor.

Previous radioimmunotherapy studies using iodinated murine A33 mAb (muA33) have shown the therapeutic potential of the A33 antigen-antibody system in patients with metastatic colorectal cancer (1012). A phase I/II study of treatment doses of 131I-muA33 enrolled 23 patients with advanced unresectable colon cancer. Reversible hematologic toxicity was dose-limiting and the maximum tolerated dose was determined to be 75 mCi/m2. No significant bowel toxicity was observed. Minor evidence of tumor response was seen, with three patients demonstrating mixed responses to treatment. Another phase I trial examined the effects of escalating doses of 125I-muA33 in patients with advanced colon cancer (12). Twenty-one patients were enrolled and treated with 125I-muA33 at doses ranging from 50 to 350 mCi/m2. Dose-limiting toxicity was not observed and maximum tolerated dose was therefore not reached. One patient had a mixed tumor response and 12 patients had stable disease. This series of muA33 radioimmunotherapy trials showed safety, tolerability, and specific tumor-targeting in patients with advanced colorectal cancer. Hints of antitumor efficacy were also seen, justifying further clinical studies; however, the development of human anti-mouse antibodies in all patients prevented repeat treatment. In order to reduce immunogenicity, a CDR grafted, humanized mAb (huA33) was therefore developed (13). Clinical trials using multiple dose schedules of huA33 have shown safety and possible efficacy of huA33 mAb alone and combined with chemotherapy (14, 15).

As a prelude to 131I-huA33 radioimmunotherapy, a phase I pharmacokinetic and tumor-targeting study accrued patients with colorectal carcinoma who were scheduled for surgical resection of one or more tumors. One week prior to surgery, patients were infused with 131I and 125I labeled huA33. Surgical tumor biopsies were analyzed for 131I-huA33 uptake and microdistribution. Favorable biodistribution and tumor-targeting data were obtained,4

4

Scott AM, Lee FT, Jones R, Hopkins W, MacGregor D, Cebon JS Hannah A, Chong G, U P, Papenfuss A, Rigopoulos A, Sturrock S, Murphy R, Wirth V, Murone C, Smyth FE, Knight S, Welt S, Ritter G, Richards E, Nice EC, Burgess AW, Old LJ. Phase I biopsy-based study of humanised monoclonal antibody A33 in patients with colorectal carcinoma. Clin Can Res, In press.

thus a phase I trial using radioimmunotherapy doses of 131I-huA33 was conducted and is reported here.

Patients. Patients with metastatic colorectal cancer who had received at least one line of chemotherapy for metastatic disease were eligible for this trial if they fulfilled all of the following inclusion criteria: histologically proven colorectal cancer; measurable or evaluable disease, Karnofsky performance status ≥70%; expected survival ≥4 months; serum creatinine <0.15 mmol/L; serum bilirubin <34 μmol/L; neutrophil count >1.5 × 109/L; platelet count >150 × 109/L; prothrombin time <1.3 × control; age ≥18 years; ability to give written informed consent. Patients who fulfilled any of the following exclusion criteria were ineligible for study treatment: active central nervous system metastases; exposure to chemotherapy, radiotherapy or immunotherapy within 4 weeks of study entry; not fully recovered from surgery; metastatic disease involving >50% of liver mass; positive huA33 human anti-human antibodies (HAHA) titer in the context of previous huA33 treatment; serious concurrent illness; pregnancy or lactation; concurrent treatment with systemic corticosteroids/immunosuppressants. All patients gave witnessed written informed consent to participate in this study. The protocol was approved by the Human Research Ethics Committee of the Austin Hospital, Melbourne, Australia.

A phase I dose escalation design was used where only the therapy 131I dose was escalated between dose levels; the huA33 protein dose was held constant at 10 mg/m2. Three patients were initially treated per 131I dose level. Toxicity was defined as adverse events which were possibly, probably, or definitely related to the study agent, i.e., 131I-huA33. Dose-limiting toxicity was defined as grade 3 nonhematologic toxicity, or grade 4 hematologic toxicity as defined by the National Cancer Institute Common Toxicity Criteria version 2.0, 1999. If dose-limiting toxicity was experienced by any patient at a particular dose level, three more patients were to be accrued to that dose level. The maximum tolerated dose was defined as the 131I dose below that where two or more patients out of six experienced dose-limiting toxicities.

Within 2 weeks of study treatment, pretreatment assessments were done to confirm eligibility. Included were complete medical history and physical examination, documentation of measurable or evaluable tumor by computerized tomography, serum biochemistry, and hematology. A baseline serum sample for HAHA determination was taken in all patients. Assessments of cardiorespiratory function were also done as a baseline for monitoring of possible delayed radiotoxicity.

All patients commenced treatment with an i.v. trace-labeled “scout” dose of 131I-huA33 in order to assess tumor-targeting, dosimetry, and biodistribution. All patients received 5 mg (fixed dose) huA33 conjugated to 5 to 8 mCi (200-320 MBq) 131I, using an established labeling technique as previously described (16). Each patient was directly observed for evidence of anaphylactoid reactions during the entire 1-hour infusion period and for 1 hour after completion of the infusion. Vital signs were measured once during and immediately following the infusion, and 2, 3, and 4 hours after the completion of infusion. All patients received saturated solution of potassium iodide prior to the scout 131I-huA33 infusion and continued for 3 weeks in total at 10 drops orally thrice daily.

Each patient received an i.v. therapy dose of 131I-huA33 7 ± 2 days following the scout infusion. Patients were hospitalized in a self-contained radiation-shielded room for this dose. All patients received 10 mg/m2 huA33 conjugated to 20, 30, 40, or 50 mCi/m2131I depending on assigned dose level. Automatic monitoring of blood pressure, pulse, and oxygen saturation was done during this time using a programmable patient monitoring device (Agilent Technologies, Andover, MA). Patients remained confined to this room until emitted radiation at 2 m fell to <25 μSv/hour or <25% of the initial reading. Geiger counter measurements were done daily while patients were hospitalized. All radiation safety precautions were fully compliant with Australian Radiation Protection and Nuclear Safety Agency guidelines and were supervised by the Radiation Safety Officer of the Austin Hospital.

For 6 weeks following the therapy dose, patients were reviewed weekly in order to conduct assessments of adverse events and to determine their relationship, if any, to the study agent. Blood was drawn at each weekly visit for biochemistry, hematology, and HAHA measurement. Clinical and investigational adverse events were recorded in case report forms and were coded according to the Common Toxicity Criteria.

Six weeks after administration of the therapy 131I-huA33 dose, an end of study evaluation for each patient was done. This included clinical assessment of adverse events, as well as clinical and radiological tumor reassessment. Tumor response or progression was defined according to WHO criteria.

In order to detect possible delayed toxicity from 131I-huA33 therapy, all patients were followed, where possible, at 3, 6, and 12 months following completion of the initial study period. Follow up assessments included clinical assessment, biochemical assessment of renal and thyroid function, chest X-rays, respiratory function tests, and gated cardiac blood pool scans.

Pharmacokinetics. Blood samples for analysis of pharmacokinetics of 131I-huA33 (scout infusion) were drawn immediately prior to the infusion (day 0), 5 minutes after completion of 131I-huA33 infusion, and 1, 2, 4, and 20 ± 2 hours (days 2 or 3, 4 or 5, and prior to the therapy infusion). The serum obtained was aliquoted in duplicate and counted in a gamma scintillation counter (Packard Instruments, Melbourne, Australia). Triplicate standards prepared from the injected material were counted for 131I at each time point with serum samples to enable calculations to be corrected for the isotope's physical decay. The results of the serum were expressed as the percentage of injected dose per liter (%ID/L) and converted to micrograms per milliliter based on actual protein dose infused. Population pharmacokinetics used normalized %ID/L (expressing subsequent %ID/L values as a percentage of the initial value for each patient) in order to compare clearance patterns between different patients.

Pharmacokinetic calculations were done for serum data using a curve-fitting program (WinNonlin, Pharsight Co., Mountain View, CA). A two-compartment model was used to calculate pharmacokinetic results. The following variables were calculated from the model: Cmax (maximum serum concentration); AUC (area under the serum concentration curve extrapolated to infinite time); CL (total serum clearance); T1/2α and T1/2β (half lives of the initial and terminal phases of disposition); and V1 (central compartment distribution).

Biodistribution of 131I-huA33. Whole-body gamma camera scans were done 1 to 4 hours post-scout 131I-huA33 infusion, and then on days 1, 2 or 3, and 4 or 5. A standard of known activity was included in the field of view to allow quantitation. Single-photon emission computed tomography imaging of areas of known tumor was done on day 4 or 5. Analysis of biodistribution images was done following the scout 131I-huA33 dose to ensure that no abnormal distribution of 131I-huA33 had occurred prior to administration of the therapy 131I-huA33 dose.

Whole-body clearance and dosimetry of 131I-huA33. Dosimetric analysis was done on the series of gamma camera whole-body planar images acquired in all patients following scout infusion. Whole body, lungs, liver, spleen, kidney, and thyroid regions of interest were defined for each time point on both anterior and posterior images. Organ radioactivity content was estimated from the geometric mean of anterior and posterior regions of interest counts. The counts for each organ were corrected for background using regions of interest drawn adjacent to each organ. Correction for attenuation of individual organs was estimated using an analytic technique as described by Liu et al. (17). Resultant counts were converted to activity using a camera sensitivity factor calculated from a gamma camera standard of known activity which was scanned at the same time.

A time-activity curve was generated from the derived activity of each imaging time point for each organ. The time-activity curves were decay-corrected to time of injection and fitted with a single component exponential clearance expression. These fits were used to determine biological clearance half-life, Tbio, and initial fraction of injected activity, A0, for the whole body and various normal organs. The AUC was analytically determined and expressed as the residence time of the organ and remainder tissue (18). Residence times for the lungs, liver, spleen, kidney, thyroid, and remainder tissue were entered into the MIRDOSE3 computer software program (19) to calculate the estimated radiation absorbed dose to these specific organs per unit of injected activity (mGy/MBq). This value was multiplied by the total activity delivered to the patient to determine the absorbed radiation dose for each organ.

For tumor dosimetry calculations, a region of interest was drawn around all of or part of the tumor and a time-activity curve derived using similar methodology to normal organ dose. Tumor volumes were calculated from computerized tomography scans. Due to prolonged retention of 131I-huA33 in tumor, additional imaging was done up to 28 days following therapy infusion in 5 of 15 patients to more accurately determine tumor uptake and dose. Tumor mean absorbed dose was calculated as follows:

\[{\breve{D}}={\tilde{A}}S\]

where , mean absorbed dose to target; , cumulated activity in target; S, absorbed dose in target per unit cumulated activity.

For 131I, S values were obtained from predefined tables corresponding to the calculated tumor mass [MIRDOSE3, Nodule Module; ref. (19)].

To extrapolate therapy absorbed dose to tumor, the scout absorbed dose was normalized by the injected activity and then multiplied by the therapy injected activity.

Red marrow dosimetry calculations. Calculation of red marrow dose for 131I-huA33 was done using a patient-specific marrow dosimetry methodology (20). (A) Serum clearance calculations of 131I-huA33 were used to calculate serum cumulated activity concentration, and marrow cumulated activity concentration (μCi − h/gm). (B) Whole body cumulated activity concentration (μCi − h/gm) was calculated as previously described. The calculation of whole-body clearance for scout doses for red marrow dose was based on imaging data. As there were fewer imaging time points following the therapy infusion, red marrow–absorbed doses for therapy infusions were calculated using the whole body clearance determined from handheld dose rate monitor readings (UMo LB 123, EG&G Berthold, Germany).

Red marrow absorbed dose (rad/mCi) was then calculated with the formula (ref. 20):

\[\frac{\mathrm{Red\ marrow\ absorbed\ dose\ (rad/mCi)}=(0.313{\times}A+0.456{\times}B)}{\mathrm{injected\ dose\ (mCi)}}\]

Biacore HAHA analyses. Blood samples for HAHA assessment were taken prior to each 131I-huA33 infusion, then at weekly intervals until the end-of-study visit (6 weeks after therapy dose). Antibody responses against humanized antibodies (HAHA) induced after treatment of patients with huA33 mAb were analyzed by surface plasmon resonance technology using a Biacore 2000 instrument as previously described (2123). Patient serum was considered HAHA-positive if the RU value at a serum dilution of 1:100 exceeded a cutoff value, defined as the mean interpatient baseline RU value +3 SD of pretreatment sera at a serum dilution of 1:100, provided that such an increase was absent in the isotype-matched control antibody channel (24).

Fifteen patients were entered and treated according to the study protocol. Table 1 shows patient age, gender, assigned dose level, Karnofsky performance status and site of primary tumor. The median age was 63 (range, 39-77). The male/female ratio was 11:4. Median performance status was 90 (range, 80-100). Sites of metastatic disease ranged from liver, lung, retroperitoneal/mediastinal lymph nodes, pelvic, and peritoneum (Table 2). All patients had received previous therapy for metastatic colorectal cancer (Table 2). The median number of lines of previous chemotherapy was two (range, 1-5). Three patients had been previously exposed to humanized antibodies as part of clinical trials, but there was no evidence of HAHA to huA33 prior to study entry.

Table 1.

Baseline patient demographics

Patient no.Dose level (mCi/m2)Therapy 131I dose administered
Age at study entry (years)SexKarnofsky performance statusSite of primary tumor
(mCi)(GBq)
20 32.7 1.21 65 90 rectum 
20 34.8 1.29 65 100 colon 
20 37.3 1.38 63 90 colon 
30 55.4 2.05 50 90 rectum 
30 68.0 2.52 63 80 colon 
30 63.4 2.35 39 90 rectum 
40 77.0 2.85 53 90 colon 
40 72.2 2.67 68 100 rectum 
40 69.0 2.55 73 100 rectum 
10 50 75.5 2.80 67 90 colon 
11 50 94.7 3.50 56 100 rectum 
12 50 85.5 3.16 77 90 colon 
13 50 81.6 3.02 62 90 colon 
14 50 70.8 2.62 54 100 colon 
15 50 84.1 3.11 74 100 colon 
Patient no.Dose level (mCi/m2)Therapy 131I dose administered
Age at study entry (years)SexKarnofsky performance statusSite of primary tumor
(mCi)(GBq)
20 32.7 1.21 65 90 rectum 
20 34.8 1.29 65 100 colon 
20 37.3 1.38 63 90 colon 
30 55.4 2.05 50 90 rectum 
30 68.0 2.52 63 80 colon 
30 63.4 2.35 39 90 rectum 
40 77.0 2.85 53 90 colon 
40 72.2 2.67 68 100 rectum 
40 69.0 2.55 73 100 rectum 
10 50 75.5 2.80 67 90 colon 
11 50 94.7 3.50 56 100 rectum 
12 50 85.5 3.16 77 90 colon 
13 50 81.6 3.02 62 90 colon 
14 50 70.8 2.62 54 100 colon 
15 50 84.1 3.11 74 100 colon 
Table 2.

Patient treatment history and response to 131I-huA33

Patient no.Adjuvant radiotherapyChemotherapy for metastatic diseaseOther treatment for metastatic diseaseSites of tumor at study entryTumor response to 131I-huA33
yes 5-FU, I, O, R RFA of liver metastases, huA33 liver, lung, LN PD 
no 5-FU — liver SD 
no IHA 5-FU, 5-FU resection of liver metastases, alcohol injection, huA33 liver, lung, LN, sacrum PD 
no O/5-FU, I, R — liver, LN, presacral PD 
no 5-FU, I — liver, lung, LN PD 
yes 5-FU, I — liver, lung, LN SD 
no 5-FU, I resection of pelvic and left iliac fossa metastases, pelvic radiotherapy liver, lung, pelvic, LN SD 
yes 5-FU, I resection of liver metastases liver, lung, LN PD 
yes O/5-FU Sibrotuzumab liver SD 
10 no 5-FU, I right lung lower lobectomy liver, lung PD 
11 yes IHA 5-FU, I/5-FU — liver, lung PD 
12 no — liver, abdominal wall PD 
13 no O/5-FU — liver, lung PD 
14 no O/5-FU — liver PD 
15 no I,O — lung, LN, ascites PD 
Patient no.Adjuvant radiotherapyChemotherapy for metastatic diseaseOther treatment for metastatic diseaseSites of tumor at study entryTumor response to 131I-huA33
yes 5-FU, I, O, R RFA of liver metastases, huA33 liver, lung, LN PD 
no 5-FU — liver SD 
no IHA 5-FU, 5-FU resection of liver metastases, alcohol injection, huA33 liver, lung, LN, sacrum PD 
no O/5-FU, I, R — liver, LN, presacral PD 
no 5-FU, I — liver, lung, LN PD 
yes 5-FU, I — liver, lung, LN SD 
no 5-FU, I resection of pelvic and left iliac fossa metastases, pelvic radiotherapy liver, lung, pelvic, LN SD 
yes 5-FU, I resection of liver metastases liver, lung, LN PD 
yes O/5-FU Sibrotuzumab liver SD 
10 no 5-FU, I right lung lower lobectomy liver, lung PD 
11 yes IHA 5-FU, I/5-FU — liver, lung PD 
12 no — liver, abdominal wall PD 
13 no O/5-FU — liver, lung PD 
14 no O/5-FU — liver PD 
15 no I,O — lung, LN, ascites PD 

Abbreviations: I, irinotecan; O, oxaliplatin; R, raltitrexed; IHA, intrahepatic artery; RFA, radiofrequency ablation; sibrotuzumab, clinical trial monoclonal antibody; LN, lymph nodes; PD, progressive disease; SD, stable disease.

Toxicity. Overall, 131I-huA33 was well-tolerated at dose levels up to the maximum tolerated dose. No acute infusion-related toxicities were observed. Hematologic toxicity was observed at all dose levels and increased with increasing dose level. Table 3 shows the maximum grade of hematologic toxicity by Common Toxicity Criteria experienced by each patient. Because of concerns about hematologic toxicity, in particular, thrombocytopenia, the 50 mCi/m2 dose level was expanded to six patients. There was one episode of hematologic dose-limiting toxicity observed, which was grade 4 neutropenia in patient 14. This patient also experienced grade 3 thrombocytopenia and the longest time to complete hematologic recovery (10 weeks). No episodes of febrile neutropenia or thrombocytopenia requiring platelet transfusion were observed at any dose level.

Table 3.

Study agent–related toxicity (maximum Common Toxicity Criteria grade)

Patient no.Dose level (mCi/m2)HemoglobinLeukocytesNeutrophilsPlatelets
20 
20 
20 
30 
30 
30 
40 
40 
40 
10 50 
11 50 
12 50 
13 50 
14 50 
15 50 
Patient no.Dose level (mCi/m2)HemoglobinLeukocytesNeutrophilsPlatelets
20 
20 
20 
30 
30 
30 
40 
40 
40 
10 50 
11 50 
12 50 
13 50 
14 50 
15 50 

Unexpected, but clinically significant, cutaneous toxicity was observed in a total of seven patients. The frequency of cutaneous events was similar at all 131I dose levels. Two patients (nos. 7 and 14) experienced grade 3 extensive pruritic, erythematous, urticaria-like rashes requiring treatment with systemic corticosteroids. There was no evidence that the rash was associated with 131I-dose; it was most likely due to A33 protein. In the other patients, symptoms related to rashes were treated with oral antihistamines or no specific treatment. There was no clear relationship between development of rash and/or pruritis and the development of HAHA. Table 4 shows the occurrence of cutaneous events, the presence of HAHA, and 131I-huA33 dose levels. Skin biopsies from patient 7 were taken 3 weeks post-therapy 131I-huA33 infusion, and histologic features were consistent with a nonspecific spongiform dermatitis and further immunostains for immune complexes/complement were negative.

Table 4.

Occurrence of cutaneous events, presence of HAHA, and 131I-huA33 dose level

Patient no.Dose level (mCi/m2)Rash (Common Toxicity Criteria grade)Pruritus (Common Toxicity Criteria grade)HAHA
20 no 
20 no 
20 no 
30 yes 
30 yes 
30 no 
40 yes 
40 no 
40 no 
10 50 no 
11 50 no 
12 50 no 
13 50 yes 
14 50 no 
15 50 no 
Patient no.Dose level (mCi/m2)Rash (Common Toxicity Criteria grade)Pruritus (Common Toxicity Criteria grade)HAHA
20 no 
20 no 
20 no 
30 yes 
30 yes 
30 no 
40 yes 
40 no 
40 no 
10 50 no 
11 50 no 
12 50 no 
13 50 yes 
14 50 no 
15 50 no 

NOTE: All listed cutaneous events were considered to be related to the study agent.

No clinically significant gastrointestinal toxicity was observed, either in the initial study follow-up or in the long-term follow-up phases. Patient 13 experienced a “serum sickness”–like clinical syndrome in association with Biacore evidence of HAHA formation. Symptoms of myalgias, chills, and fatigue were prominent and commenced 1 week following the 131I-huA33 therapy dose. Biodistribution following the scout infusion was normal and excellent tumor-targeting was evident.

Maximum tolerated dose. The maximum tolerated dose was determined to be 40 mCi/m2 of 131I. This was based on the occurrence of two episodes of dose-limiting toxicity (grade 4 neutropenia and grade 3 liver function test abnormalities) across six patients treated at the 50 mCi/m2 dose level.

Evaluation of delayed toxicity.Eight patients were available for at least one follow-up assessment. The other seven patients were not available because of death due to tumor progression, other treatment, or withdrawal of consent for further follow-up. One patient developed biochemical hypothyroidism on long-term follow-up in the context of previous treatment with trace 131I-labeled mAb as part of a previous clinical trial. The patient remained clinically euthyroid and oral thyroid hormone replacement was commenced with subsequent improvement in biochemical thyroid function. No significant deterioration in cardiac, respiratory, renal, or liver function attributable to 131I-huA33 was detected on long-term follow-up on available patients.

Pharmacokinetics.Pharmacokinetic analysis of scout infusion data showed consistent results across all dose levels, with T1/2α (mean ± SD) 23.0 ± 12.5 hours and T1/2β 135.2 ± 46.9 hours. The calculated (mean ± SD) Cmax was 1.40 ± 0.23 mg/L, CL was 0.035 ± 0.010 L/hour, AUC was 144.4 ± 41.8 hour × mg/L, and V1 3.48 ± 0.64 L.

Biodistribution and dosimetry of 131I-huA33: Biodistribution. All patients had normal 131I-huA33 biodistribution following the scout infusion, with initial images demonstrating 131I-huA33 blood pool distribution, followed by tumor uptake and bowel localization (Fig. 1). Colon activity gradually resolved with blood clearance of 131I-huA33. Excellent uptake of 131I-huA33 in tumor was seen in all patients, with persisting high uptake visualized up to 4 weeks after the 131I-huA33 therapy infusion (Fig. 2). Post-therapy images showed essentially identical biodistribution and tumor uptake of 131I-huA33 compared with scout infusions in all patients.

Fig. 1.

Anterior whole body gamma camera images following infusion of 131I-huA33 in patient 4 (30 mCi/m2 dose level) are shown: (A) day 0, (B) day 1, (C) day 2, and (D) day 5 post-scout infusion, and (E) day 6 post-therapy infusion. A standard for quantitation of 131I-huA33 uptake is present adjacent to the left shoulder. A, Initial (day 0) images show blood pool appearance only, with a metastatic lesion in the liver demonstrating an initial hypovascular appearance. B, excellent targeting of the metastatic lesion in the liver by 131I-huA33 is clearly seen (arrow), as early as day 1, and (C-D) increasing rapidly with time. D, some central necrosis in the tumor is also evident (arrow). Gradual bowel uptake (double arrow) of 131I-huA33 is also seen. E, the post-therapy image shows identical biodistribution and tumor uptake of 131I-huA33 compared to the scout dose. F, computerized tomography scan of the metastatic lesion in the liver in this patient.

Fig. 1.

Anterior whole body gamma camera images following infusion of 131I-huA33 in patient 4 (30 mCi/m2 dose level) are shown: (A) day 0, (B) day 1, (C) day 2, and (D) day 5 post-scout infusion, and (E) day 6 post-therapy infusion. A standard for quantitation of 131I-huA33 uptake is present adjacent to the left shoulder. A, Initial (day 0) images show blood pool appearance only, with a metastatic lesion in the liver demonstrating an initial hypovascular appearance. B, excellent targeting of the metastatic lesion in the liver by 131I-huA33 is clearly seen (arrow), as early as day 1, and (C-D) increasing rapidly with time. D, some central necrosis in the tumor is also evident (arrow). Gradual bowel uptake (double arrow) of 131I-huA33 is also seen. E, the post-therapy image shows identical biodistribution and tumor uptake of 131I-huA33 compared to the scout dose. F, computerized tomography scan of the metastatic lesion in the liver in this patient.

Close modal
Fig. 2.

Prolonged tumor localization following therapy 131I-huA33 infusion in patient 12 (50 mCi/m2). Gamma camera images of the upper abdomen (A) anterior and (B) posterior (day 7 post-therapy infusion); (C) anterior (D) posterior (day 21 post-therapy infusion); (E) anterior and (F) posterior (day 28 post-therapy infusion) are shown. Excellent uptake of 131I-huA33 in a metastatic lesion in the posterior right lobe of liver (arrow) is shown up to day 28 post-therapy infusion. Bowel activity (double arrow) reduces with time, in contrast to the prolonged tumor retention of 131I-huA33.

Fig. 2.

Prolonged tumor localization following therapy 131I-huA33 infusion in patient 12 (50 mCi/m2). Gamma camera images of the upper abdomen (A) anterior and (B) posterior (day 7 post-therapy infusion); (C) anterior (D) posterior (day 21 post-therapy infusion); (E) anterior and (F) posterior (day 28 post-therapy infusion) are shown. Excellent uptake of 131I-huA33 in a metastatic lesion in the posterior right lobe of liver (arrow) is shown up to day 28 post-therapy infusion. Bowel activity (double arrow) reduces with time, in contrast to the prolonged tumor retention of 131I-huA33.

Close modal

Whole body clearance. Whole body clearance [biologic half-life (T1/2)] of 131I-huA33 scout dose was calculated to be 227.52 ± 46.15 hours (mean ± SD).

Whole body and normal organ dosimetry. The results for individual organs and whole-body doses were similar between patients at the four dose levels. Target organs with the highest doses were lungs, liver, spleen, kidney, and thyroid, consistent with the blood pool appearance on gamma camera images and minor thyroid uptake from free 131I (Fig. 1). Mean ± SD values were for kidney (0.23 ± 0.07 cGy/MBq), liver (0.17 ± 0.06 cGy/MBq), lungs (0.13 ± 0.04 cGy/MBq), spleen (0.26 ± 0.06 cGy/MBq), and thyroid (0.39 ± 0.16 cGy/MBq). The whole-body dose (effective dose equivalent) was calculated to be 0.80 ± 0.11 mSv/MBq (mean ± SD).

Red marrow absorbed dose. The total red marrow radiation absorbed doses was similar between scout and therapy doses across all patients (0.056 ± 0.012 versus 0.054 ± 0.012 cGy/MBq). Total red marrow absorbed dose increased in proportion to therapy dose levels, with the mean ± SD values at the 20 mCi/m2 dose level (69.00 ± 5.05 cGy), 30 mCi/m2 dose level (96.68 ± 12.19 cGy), 40 mCi/m2 dose level (122.46 ± 13.12 cGy), and 50 mCi/m2 dose level (153.49 ± 46.80 cGy).

Tumor dosimetry. The mean specific absorbed tumor dose across all dose levels was 6.49 ± 2.47 Gy/GBq (mean ± SD). Total tumor dose increased with dose level, and was 19.26 ± 11.68 Gy at the 40 mCi/m2 dose level (range 12.04-32.73 Gy), and was 21.19 ± 8.17 Gy at the 50 mCi/m2 dose level (range 11.73-31.81 Gy).

Immunogenicity.There was Biacore evidence of HAHA formation in four patients (nos. 4, 5, 7, and 13). Only one patient had clinical manifestations probably due to HAHA (see Toxicity). Onset of Biacore-detected HAHA was generally 2 to 3 weeks following the scout 131I-huA33 dose, with peak values within a week of onset. The exception was patient 13, whose HAHA was Biacore-detectable 1 week after the scout dose, peaking 1 week later. Biacore responses gradually declined after peaking, although in almost all affected patients, HAHA was still detectable 7 weeks after the scout infusion.

Tumor response.On WHO computerized tomography-based criteria, 4 patients had stable disease at the end of study visit (6 weeks after 131I-huA33 therapy dose) and 11 patients had progressive disease (Table 1). Patients with more bulky disease tended to develop progressive disease while in the study, compared with patients with relatively low volume tumors who were more likely to have stable disease 6 weeks after treatment. In most cases, patients had clinical symptoms or signs of tumor progression in addition to computerized tomography evidence.

This phase I radioimmunotherapy study reinforces our current understanding of the A33-huA33 mAb system and improves our ability to apply this to patients with advanced colorectal cancer. Like muA33, huA33 has been shown to be well-tolerated when given by i.v. infusion. Tumor-targeting and biodistribution with 131I-huA33 is comparable to radiolabeled muA33. HuA33 has a longer serum half-life compared with muA33 with a T1/2β of 135 versus 38.5 hours (10). The incidence of HAHA is clearly less than the frequency of human anti-mouse antibody generated with the murine mAb.

The temporal pattern and magnitude of hematologic toxicity observed was consistent with previous radioimmunotherapy trials, including studies with 131I-muA33 (11). In our study, two episodes of dose-limiting toxicity were observed at 50 mCi/m2 dose level: one episode of grade 4 neutropenia was clearly due to 131I-huA33, and it was not possible to exclude an acute radiation-induced hepatitis as the cause of the grade 3 liver function test abnormalities in an additional patient. The maximum tolerated dose in the 131I-muA33 study was 75 mCi/m2, which is somewhat higher than the maximum tolerated dose of 40 mCi/m2 observed for 131I-huA33. This can be readily explained by the longer half-life of the humanized mAb labeled with a moderately long-lived radioisotope, and the resultant red marrow dose.

Non–infusion-related cutaneous toxicity has not been shown in any of the preceding muA33 or huA33 clinical studies. Thus, the 47% incidence of rash and/or pruritis in this huA33 radioimmunotherapy study was unexpected. The etiology of these symptoms remains unclear; however, they were almost certainly related to the huA33 protein rather than the radioisotope. The lack of correlation between incidence and severity of cutaneous reactions and 131I dose level supports this hypothesis. One might expect the mechanism to relate to HAHA formation, however, there was no correlation between skin toxicity and HAHA based on Biacore data.

Clinically significant gastrointestinal toxicity was not observed in any of the muA33 or huA33 radioimmunotherapy trials. Despite the presence of A33 antigen on normal colonic epithelium, and showed targeting of this in patients based on autoradiography and gamma imaging, clinical sequelae have not been observed. There are two potential reasons for this: the most likely explanation is that the radiation dose given was too low to cause radiation colitis. The other plausible explanation (which may act in tandem with the first) is that the normal physiologic turnover of colonic epithelial cells was protective against effects of radiation-induced apoptosis. This is consistent with the calculation of the average colon elimination (k30) half-life of 131I-huA33 to be 32.4 ± 8.1 hours from a previous biopsy-based huA33 trial.4 The selection of 131I rather than other β emitters (e.g., 90Y or 177Lu) labeled to huA33 was based in part on the bowel uptake of huA33, and our concern regarding potential bowel toxicity such as has been reported with other 90Y-labeled mAbs that also target an antigen expressed in normal bowel (25). The choice of 131I has been vindicated by the absence of clinically significant bowel toxicity in this trial. The lack of bowel toxicity, combined with the excellent uptake in tumor of 131I-huA33, indicates that radioimmunotherapy with huA33 is a highly attractive strategy for therapy of colorectal cancer.

Hypothyroidism is a known potential complication of 131I-based radioimmunotherapy (26). The standard preventive procedure of oral iodine administration has been shown to reduce thyroid uptake of 131I to <1% of the injected activity in patients receiving radioimmunotherapy with 131I-labeled mAbs (27). Interestingly, the only patient to become biochemically hypothyroid only received 5.2 Gy to the thyroid gland—well below the mean experienced by the study population. It is possible that preexisting mild thyroid dysfunction at study entry progressed despite seemingly adequate oral iodine blockade.

In this study, four patients had stable disease at restaging in response to 131I-huA33, all of whom had previously received two or fewer agents for advanced colon cancer. Eleven patients had progressive disease at restaging, of whom eight patients had previously received two or fewer agents for advanced colon cancer. Although this might suggest an inverse relationship between the number of prior therapies and response to 131I-huA33, the numbers are too small to draw any definitive conclusions.

Other 131I labeled humanized mAbs have been tested in trials and reported in humans with colorectal cancer. hMN-14 is a humanized mAb targeting carcinoembryonic antigen (28) and a phase II radioimmunotherapy trial using 131I-hMN-14 has been recently reported (29). A dose of 60 mCi/m2 was studied in 30 patients with small volume metastatic colorectal cancer. Moderate and reversible myelosuppression was the dominant toxicity, although one patient developed grade 4 thrombocytopenia. The pattern of myelotoxicity reported in this 131I-hMN-14 trial was very consistent with that observed in our 131I-huA33 radioimmunotherapy study. Normal organ and tumor dosimetry for 131I-hMN-14 has been shown to be similar to that for 131I-huA33, considering the differences in mAb specificity and radioisotope dose (28, 29). The prolonged retention of 131I-huA33 in tumor (at least 4 weeks) is unique, however, and is likely due to the cellular location of the A33 antigen in tumor cells and lack of trafficking of A33 antigen/antibody complex to intracellular lysosomes. Our estimates of tumor retention and dose are therefore likely to be underestimates of the true radiation dose to tumor with this therapeutic approach.

Radioimmunotherapy has yet to establish itself as a standard part of therapeutic strategy for epithelial malignancies. In this respect, its application is not nearly as advanced as radioimmunotherapy for hematologic malignancies. The role of 131I-huA33 mAb is also not as yet clearly defined, despite clear evidence of antitumor efficacy in animal models, and the excellent tumor-targeting properties and stable disease in 4 of 15 patients shown in this trial. The dose to tumor at maximum tolerated dose in this trial ranged from 12 to 33 Gy, indicating that a single dose of 131I-huA33 may not be sufficient to achieve tumoricidal effects alone. Improved dose to tumor could be achieved with higher LET β emitters (e.g., 90Y or 177Lu); however, the problem of potential bowel toxicity will require careful evaluation in future trials.

Combined modality therapy is a currently favored route of drug and therapeutic development. Preclinical data have shown enhanced radiation sensitivity of tumor cells pretreated with cytotoxics such as paclitaxel (30). As a result, the combination of chemotherapy and radiotherapy has become standard treatment for a number of epithelial tumors over the last 10 years. The synergy of trastuzumab with cisplatin or paclitaxel observed in animal models of breast cancer has also translated into clinical practice, and Bevacizumab in combination with 5-fluorouracil (5-FU)/folinic acid has shown greater tumor response rates than 5-FU/folinic acid alone (31). There is also animal model data suggesting synergy between 131I-huA33 and 5-FU for colorectal tumors (32), and a trial of 90Y-anti-carcinoembryonic antigen chimeric T84.66 with 5-FU has shown the tolerability of this approach (33). Importantly, an orally bioavailable 5-FU prodrug, capecitabine, has been found to produce similar tumor response rates to i.v.-administered 5-FU, and also has been shown to improve response rates when given in conjunction with irradiation (34). The rationale for the combination of 131I-huA33 radioimmuotherapy and capecitabine is therefore compelling, and we intend to evaluate this therapeutic approach in future trials.

1
DeNardo GL, O'Donnell RT, Kroger LA, et al. Strategies for developing effective radioimmunotherapy for solid tumors.
Clin Cancer Res
1999
;
5
:
3219s
–23s.
2
Scott AM, Welt S. Antibody-based immunological therapies.
Curr Opin Immunol
1997
;
9
:
717
–22.
3
Greenlee RT, Hill-Harmon MB, Murray T, Thun M. Cancer statistics, 2001.
CA Cancer J Clin
2001
;
51
:
15
–36.
4
Grothey A SH. New chemotherapy approaches in colorectal cancer.
Curr Opin Oncol
2001
;
13
:
275
–86.
5
Sakamoto J, Kojima H, Kato J, Hamashima H, Suzuki H. Organ-specific expression of the intestinal epithelium-related antigen A33, a cell surface target for antibody-based imaging and treatment in gastrointestinal cancer.
Cancer Chemother Pharmacol
2000
;
46
Suppl:
S27
–32.
6
Heath JK, White SJ, Johnstone CN, et al. The human A33 antigen is a transmembrane glycoprotein and a novel member of the immunoglobulin superfamily.
Proc Natl Acad Sci U S A
1997
;
94
:
469
–74.
7
Johnstone CN, Tebbutt NC, Abud HE, et al. Characterization of mouse A33 antigen, a definitive marker for basolateral surfaces of intestinal epithelial cells.
Am J Physiol Gastrointest Liver Physiol
2000
;
279
:
G500
–10.
8
Moritz RL, Ritter G, Catimel B, et al. Micro-sequencing strategies for the human A33 antigen, a novel surface glycoprotein of human gastrointestinal epithelium.
J Chromatogr A
1998
;
798
:
91
–101.
9
Ritter G, Cohen LS, Nice EC, et al. Characterization of posttranslational modifications of human A33 antigen, a novel palmitoylated surface glycoprotein of human gastrointestinal epithelium.
Biochem Biophys Res Commun
1997
;
236
:
682
–6.
10
Welt S, Divgi CR, Real FX, et al. Quantitative analysis of antibody localization in human metastatic colon cancer: a phase I study of monoclonal antibody A33.
J Clin Oncol
1990
;
8
:
1894
–906.
11
Welt S, Divgi CR, Kemeny N, et al. Phase I/II study of iodine 131-labeled monoclonal antibody A33 in patients with advanced colon cancer.
J Clin Oncol
1994
;
12
:
1561
–71.
12
Welt S, Scott AM, Divgi CR, et al. Phase I/II study of iodine 125-labeled monoclonal antibody A33 in patients with advanced colon cancer.
J Clin Oncol
1996
;
14
:
1787
–97.
13
King DJ, Antoniw P, Owens RJ, et al. Preparation and preclinical evaluation of humanised A33 immunoconjugates for radioimmunotherapy.
Br J Cancer
1995
;
72
:
1364
–72.
14
Welt S, Ritter G, Williams C Jr, et al. Phase I study of anti-colon cancer humanised antibody A33.
Clin Cancer Res
2003
;
9
:
1338
–46.
15
Welt S, Ritter G, Williams C Jr, et al. Preliminary report of a phase I study of combination chemotherapy and huA33 immunotherapy.
Clin Cancer Res
2003
;
9
:
1347
–53.
16
Lee FT, Hall C, Rigopoulos A, et al. Immuno-PET of human colon xenograft- bearing BALB/c nude mice using 124I-CDR-grafted humanized A33 monoclonal antibody.
J Nucl Med
2001
;
42
:
764
–9.
17
Liu A, Williams LE, Raubitschek AA. A CT assisted method for absolute quantitation of internal radioactivity.
Med Phys
1996
;
23
:
1919
–28.
18
Loevinger R, Budinger TF, Watson EE. MIRD primer for absorbed dose calculations. New York: The Society of Nuclear Medicine; 1991.
19
Stabin MG. MIRDOSE: personal computer software for internal dose assessment in nuclear medicine.
J Nucl Med
1996
;
37
:
538
–46.
20
Shen S, DeNardo GL, Sgouros G, O'Donnell RT, DeNardo SJ. Practical determination of patient-specific marrow dose using radioactivity concentration in blood and body.
J Nucl Med
1999
;
40
:
2102
–6.
21
Ritter G, Cohen LS, Williams C Jr, Richards EC, Old LJ, Welt S. Serological analysis of human anti-human antibody responses in colon cancer patients treated with repeated doses of humanized monoclonal antibody A33.
Cancer Res
2001
;
61
:
6851
–9.
22
Scott AM, Lee FT, Hopkins W, et al. Specific targeting, biodistribution, and lack of immunogenicity of chimeric anti-GD3 monoclonal antibody KM871 in patients with metastatic melanoma: results of a phase I trial.
J Clin Oncol
2001
;
19
:
3976
–87.
23
Scott AM, Wiseman G, Welt S, et al. A phase I dose-escalation study of sibrotuzumab in patients with advanced or metastatic fibroblast activation protein-positive cancer.
Clin Cancer Res
2003
;
9
:
1639
–47.
24
Scott AM, Geleick D, Rubira M, et al. Construction, production, and characterization of humanized anti-Lewis Y monoclonal antibody 3S193 for targeted immunotherapy of solid tumors.
Cancer Res
2000
;
60
:
3254
–61.
25
Knox SJ, Goris ML, Tempero M, et al. Phase II trial of yttrium-90-DOTA-biotin pretargeted by NR-LU-10 antibody/streptavidin in patients with metastatic colon cancer.
Clin Cancer Res
2000
;
6
:
406
–14.
26
Press OW, Eary JF, Badger CC, et al. Treatment of refractory non-Hodgkin's lymphoma with radiolabeled MB-1 (anti-CD37) antibody.
J Clin Oncol
1989
;
7
:
1027
–38.
27
Behr TM, Juweid ME, Sharkey RM, et al. Thyroid radiation doses during radioimmunotherapy of CEA-expressing tumours with 131I-labelled monoclonal antibodies.
Nucl Med Commun
1996
;
17
:
767
–80.
28
Hajjar G, Sharkey RM, Burton J, et al. Phase I radioimmunotherapy trial with iodine-131–labeled humanized MN-14 anti-carcinoembryonic antigen monoclonal antibody in patients with metastatic gastrointestinal and colorectal cancer.
Clin Colorectal Cancer
2002
;
2
:
31
–42.
29
Behr TM, Liersch T, Greiner-Bechert L, et al. Radioimmunotherapy of small-volume disease of metastatic colorectal cancer.
Cancer
2002
;
94
:
1373
–81.
30
Liebmann J, Cook JA, Fisher J, Teague D, Mitchell JB. Changes in radiation survival curve parameters in human tumor and rodent cells exposed to paclitaxel (Taxol).
Int J Radiat Oncol Biol Phys
1994
;
29
:
559
–64.
31
Kabbinavar F, Hurwitz HI, Fehrenbacher L, et al. Phase II, randomized trial comparing bevacizumab plus fluorouracil (FU)/leucovorin (LV) with FU/LV alone in patients with metastatic colorectal cancer.
J Clin Oncol
2003
;
21
:
60
–5.
32
Tschmelitsch J, Barendswaard E, Williams C Jr, et al. Enhanced antitumor activity of combination radioimmunotherapy (131I-labeled monoclonal antibody A33) with chemotherapy (fluorouracil).
Cancer Res
1997
;
57
:
2181
–6.
33
Wong JY, Shibata S, Williams LE, et al. A phase I trial of 90Y-anti-carcinoembryonic antigen chimeric T84.66 radioimmunotherapy with 5-fluorouracil in patients with metastatic colorectal cancer.
Clin Cancer Res
2003
;
9
:
5842
–52.
34
Van Cutsem E, Twelves C, Cassidy J, et al. Oral capecitabine compared with intravenous fluorouracil plus leucovorin in patients with metastatic colorectal cancer: results of a large phase III study.
J Clin Oncol
2001
;
19
:
4097
–106.