Phase I Imaging and Pharmacodynamic Trial of CS-1008 in Patients With Metastatic Colorectal Cancer

Abstract

Purpose

CS-1008 (tigatuzumab) is a humanized, monoclonal immunoglobulin G1 (IgG1) agonistic antibody to human death receptor 5. The purpose of this study was to investigate the impact of CS-1008 dose on the biodistribution, quantitative tumor uptake, and antitumor response in patients with metastatic colorectal cancer (mCRC).

Patients and Methods

Patients with mCRC who had received at least one course of chemotherapy were assigned to one of five dosage cohorts and infused with a weekly dose of CS-1008. Day 1 and day 36 doses were trace-labeled with indium-111 (¹¹¹In), followed by whole-body planar and regional single-photon emission computed tomography (SPECT) imaging at several time points over the course of 10 days.

Results

Nineteen patients were enrolled. ¹¹¹In-CS-1008 uptake in tumor was observed in only 12 patients (63%). ¹¹¹In-CS-1008 uptake and pharmacokinetics were not affected by dose or repeated drug administration. ¹¹¹In-CS-1008 biodistribution showed gradual blood-pool clearance and no abnormal uptake in normal tissue. No anti–CS-1008 antibody development was detected. One patient achieved partial response (3.7 months duration), eight patients had stable disease, and 10 patients had progressive disease. Clinical benefit rate (stable disease + partial response) in patients with ¹¹¹In-CS-1008 uptake in tumor was 58% versus 28% in patients with no uptake. An analysis of individual lesions showed that lesions with antibody uptake were one third as likely to progress as those without antibody uptake (P = .07). Death-receptor–5 expression in archived tumor samples did not correlate with ¹¹¹In-CS-1008 uptake (P = .5) or tumor response (P = .6).

Conclusion

Death-receptor–5 imaging with ¹¹¹In-CS-1008 reveals interpatient and intrapatient heterogeneity of uptake in tumor, is not dose dependent, and is predictive of clinical benefit in the treatment of patients who have mCRC.

INTRODUCTION

Death receptor 5 (DR5), also known as tumor necrosis factor-related apoptosis-inducing ligand receptor 2 (TRAIL-R2), is a cell surface receptor with a cytoplasmic death domain that, when activated by its ligand (apoptosis ligand 2 [Apo2L/TRAIL]), triggers the extrinsic apoptotic pathway by activating caspases.¹ DR5 is overexpressed in a variety of tumor types, including colon, gastric, pancreatic, lung, and cervical cancer, but with limited expression in normal tissues.²

CS-1008 (tigatuzumab) is a humanized, monoclonal immunoglobulin G1 (IgG1) antibody to human DR5 created by complementarity determining region–grafting the murine antibody TRA-8 (mTRA-8).^3,4 Both mTRA-8 and CS-1008 showed potent in vitro cytotoxicity⁴ and significant in vivo antitumor activity against solid tumor xenografts.^3,5 Preclinical studies demonstrated a direct correlation of CS-1008 uptake in tumor, receptor occupancy, and tumor growth inhibition, and receptor saturation in vivo was also associated with a threshold level of therapeutic effect.⁶

Clinically, CS-1008, similar to other DR5 agonists,^7–14 showed a favorable toxicity profile with no dose-limiting toxicity at doses as high as 8 mg/kg/wk, and long-term disease stabilization was observed.⁹ However, none of the combination studies with DR5 agonists^15–22 (Appendix Table A1, online only) achieved their end points of improving disease outcomes, thereby highlighting the potential importance of patient selection and/or rational therapeutic combinations.

Tumors are genetically unstable, as this is the most efficient way for them to evolve,^23–25 and this may lead to significant heterogeneity between tumors as well as within a single tumor for receptor expression. In addition, antibody penetration into tumors may be nonuniform²⁶ as a result of a variety of biophysical factors²⁷ such as intervessel distance, interstitial pressure, receptor density, and internalization rate. Thus, heterogeneity of receptor expression or of antibody biodistribution may affect results of phase II studies by diluting the therapeutic benefit seen in a subset of patients or tumor masses.

In view of the linkage of DR5 activation and therapeutic efficacy,⁶ and the lack of clinical data on the relationship of dose to receptor occupancy and saturation, this study aimed to determine the biodistribution and tumor uptake of CS-1008, and to correlate these results with antitumor response in patients with metastatic colorectal cancer (mCRC).

PATIENTS AND METHODS

Eligibility Criteria

Patients with histologically proven mCRC who had received at least one course of chemotherapy for metastatic disease, with one target lesion ≥ 2 cm evaluable by gamma camera imaging and with an Eastern Cooperative Oncology Group performance status ≤ 2, were eligible. Other inclusion criteria were an age of ≥ 18 years; a life expectancy of at least 3 months; and adequate bone marrow, liver, and renal function. Patients on regular corticosteroid, nonsteroidal anti-inflammatory drug, or other immunosuppressive treatment within 3 weeks before first drug administration were excluded. Written informed consent from all patients and approval from the appropriate independent ethics committee were obtained.

Overall Study Design and Drug Administration

The trial was an open-label, single-site, phase I study. The primary objectives were to determine the impact of different loading doses on initial biodistribution, pharmacokinetics, and tumor uptake of indium-111 labeled CS-1008 (¹¹¹In-CS-1008) and changes in biodistribution, pharmacokinetics, and tumor uptake following continuous sequential doses of CS-1008. Secondary objectives were to determine changes in tumor metabolism, antitumor response, changes in serum apoptosis biomarkers, and serum tumor response markers. Two to five patients were assigned to five nonsequential cohorts (Table 1) to facilitate optimal data acquisition for analysis of biodistribution, pharmacokinetics, and imaging characteristics across dosage levels.

Table 1.

Study Dose Cohorts

Cohort No.	Day 1 111In-CS-1008 (mg/kg)	Day 8 CS-1008 (mg/kg)	Days 15, 22, 29 CS-1008 (mg/kg)	Day 36 111In-CS-1008 (mg/kg)	Day 43 CS-1008 (mg/kg)	Additional Cycles CS-1008 (mg/kg)
1	0.2	6	2	2	2	2
2	1	6	2	2	2	2
3	2	6	2	2	2	2
4	4	4	2	2	2	2
5	6	2	2	2	2	2

Abbreviation: ¹¹¹In-CS-1008, indium-111 labeled to CS-1008.

As outlined in Table 1, different CS-1008 loading doses were administered on day 1 and day 8 in each cohort, followed by an intravenous weekly dose of 2 mg/kg. These loading doses were selected on the basis of previous phase I data.⁹ Day 1 and day 36 doses were trace-labeled with ¹¹¹In (2 mg of CS-1008 radiolabeled with 185 to 259 MBq [5-7 mCi]). The duration of the first cycle was 7 weeks, and patients with partial response (PR) or stable disease (SD) could receive additional CS-1008 until the occurrence of progressive disease (PD), unacceptable toxicity, or study withdrawal at the request of the patient or the treating physician. Additional cycles were scheduled as 4-week cycles and were administered weekly at a dose of 2 mg/kg.

Radiolabeling of CS-1008

The antibody CS-1008 was labeled with ¹¹¹In (Nordion; Ottawa, Ontario, Canada) via the bifunctional metal ion chelate, CHX-A″-DTPA, according to methods previously described.^28,29

Biodistribution and Tumor Uptake of ¹¹¹In-CS-1008

Gamma camera imaging.

Gamma camera imaging with anterior and posterior whole-body sweep scans and single-photon emission computed tomography (SPECT) imaging of relevant regions with known tumor(s) were performed at five time points (day 1, day 2, day 4 or 5, day 7 or 8, and day 11 or 12), after the completion of the initial infusion of ¹¹¹In-CS-1008. The whole-body gamma camera imaging and SPECT imaging after the day-36 infusion of ¹¹¹In-CS-1008 (256.80 ± 13.29 MBq) were acquired on four time points (day 36, day 37, day 39 or 40, and day 42 or 43). All gamma camera imaging was performed on a dual-head gamma camera (SKYLight, Philips Medical Systems, North Milpitas, CA).

Quantitative Tumor Uptake

SPECT images acquired at different time points were coregistered with computed tomography (CT) images. Nonuniform CT attenuation correction of the coregistered SPECT images was performed with use of a simplified Chang algorithm.^30–32 Volumes of interest (VOIs) were drawn around the whole tumor mass on the transverse slices of SPECT image at the time points at which the tumors were most clearly identified. The tumor VOI was then transposed onto all aligned images for a particular patient. Resultant counts in the tumor VOIs were then background corrected and converted to activity by calibrating counts to a standard of known activity that was in the field of view of the patient image.

Efficacy

Tumor response was assessed according to the response evaluation criteria in solid tumors (RECIST) version 1.1 guidelines.³³ Disease assessment was based on CT and other appropriate imaging obtained at the time of screening, at the end of cycle 1 (EOC1) (day 44 to day 50), and, for patients receiving additional cycles, at the end of odd-numbered cycles and at the end of the study. The duration of overall response was measured according to RECIST guidelines.³³

Tumor metabolic response to CS-1008 was assessed by [¹⁸F]fluorodeoxyglucose ([¹⁸F] FDG) positron emission tomography (PET)/CT scan performed at screening, at day 15, and at EOC1 according to the European Organisation for Research and Treatment of Cancer guidelines.³⁴

Pharmacokinetics

Serum obtained from patients following ¹¹¹In-CS-1008 infusion was aliquoted and counted in a gamma scintillation counter (Packard Instruments, Canberra, Australia). The results were expressed as the percent of injected dose per liter and μg/mL. A two-compartment intravenous bolus model (WNL model 8) was fitted to individual labeled infusions for each patient with use of unweighted, nonlinear least squares with WinNonlin (Scientific Consultant, Apex, NC) version 5.2 (Pharsight, Mountain View, CA). A validated sandwich enzyme-linked-immunosorbent–assay (ELISA) method was also used to measure CS-1008 concentrations in sera.

Human Anti–CS-1008 Antibody

Human anti–CS-1008 antibodies (HAHAs) were measured by Medpace Reference Laboratories (Cincinnati, OH) with use of a validated ELISA protocol.

DR5 Expression in Archived Tumor-Tissue Sample

DR5 (goat polyclonal) immunohistochemical testing was performed in formalin-fixed paraffin-embedded (FFPE) human cancer tissues in accordance with Mosaic Laboratories' standard operating procedures. DR5 immunohistochemistry staining was evaluated by a pathologist who assigned an H-score as follows: percentage of cells staining 0 (unstained), 1+ (weak staining), 2+ (moderate staining) and 3+ (strong staining) were recorded; the H-score was then calculated on the basis of the summation of the product of percent of cells stained at each intensity with use of the following equation: (3 × percentage of cells staining at 3+) + (2 × percentage of cells staining at 2+) + (1 × percentage of cells staining at 1+).

Serum Apoptotic Markers and Serum Tumor Response Biomarkers

Serum samples for biomarkers of apoptosis (caspase 3/7, 8, and M30) and tumor biomarkers (carcinoembryonic antigen [CEA]) were drawn at screening, before ¹¹¹In-CS-1008 infusions on day 1 and day 36; and at 4 hours, 24 hours, and day 3 or 4 after these infusions. Blood samples for the measurement of CEA were also drawn on day 1 of additional cycles and at EOC1.

Statistical Considerations

All comparisons across cohorts were performed using a one-way analysis of variance. Comparison of paired data was performed by means of paired t test. The statistical significance of any correlation between tumor uptake, tumor response and DR5 expression was examined using Fisher's exact test. Simple least square linear regression was used to calculate the correlation coefficient between dose and uptake. A repeated measures analysis of variance was used to assess changes in serum biomarkers across all time points. All statistical tests were conducted using a two-sided alpha level of .05.

RESULTS

Patient Characteristics and Treatment

Nineteen patients (11 male, eight female) with a mean age of 64 years (range, 50 to 83 years) were entered into the trial between October 2010 and March 2012. Patient characteristics are summarized in Table 2.

Table 2.

Patient Characteristics

Characteristic	All Patients (N = 19)
Age, years
Mean	64
Range	50-83
Sex, no. of patients (%)
Male	11 (58%)
Female	8 (42%)
ECOG performance status
0	6
1	12
2	1
No. of prior chemotherapy regimens
Median	4
Range	2-6
Primary site of disease
Colon	11
Rectum	8
Histologic type of primary tumor
Adenocarcinoma	18
Mucinous carcinoma (> 50% mucinous carcinoma)	1
Histologic grade (G)/differentiation
G2: Moderately differentiated	15
G3: Poorly differentiated	3
Unknown	1

Abbreviation: ECOG, Eastern Cooperative Oncology Group.

Eighteen patients completed cycle 1, and one patient was prematurely withdrawn on day 36 as the result of symptomatic deterioration secondary to PD. Nine patients received from one to five additional cycles of CS-1008. The number of CS-1008 infusions ranged from five to 27 (median, seven) with a cumulative dose per patient ranging from 899 to 5563 mg (mean, 2101 mg).

Biodistribution and Dosimetry Analyses

A similar biodistribution pattern was observed in all patients following all ¹¹¹In-CS-1008 infusions. Evaluation of gamma camera imaging showed initial blood pooling, followed by some hepatic uptake by day 4 and gradual blood-pool clearance. Hepatic uptake was consistent with excretion of catabolized ¹¹¹In-chelate, rather than specific CS-1008 uptake (Appendix Table A2, online only). There was also no discernible uptake of ¹¹¹In-CS-1008 in any other normal tissue (Fig 1A). High, specific uptake of ¹¹¹In-CS-1008 in tumor was visualized in the target lesions of 12 patients (Table 3) and was observed to peak on day 7 or 8 after each labeled infusion (Fig 1B).

Fig 1.

(A) Whole-body biodistribution of indium-111 labeled to CS-1008 (¹¹¹In-CS-1008) in patient 014, showing gradual blood-pool clearance and no specific normal tissue uptake. (B) ¹¹¹In-CS-1008 single-photon emission computed tomography and computed tomography (SPECT/CT) in patient 014 (left, SPECT; middle, CT; right, merged SPECT/CT), showing excellent uptake of ¹¹¹In-CS-1008 in tumor (arrow) in right lung by day 7.

Table 3.

Patient Outcome and Disease Response

Patient ID	Cohort	Age (years)	Sex	Best Response by RECIST (months on study)	Metabolic Response	111In-CS-1008 Tumor Uptake
001	1	68	Male	PD (1.6)	PMD	No
002	1	64	Male	SD (6.9)	SMD	No
004	3	50	Male	SD (2.9)	SMD	No
006	2	52	Male	PD (1.7)	PMD	No
009	5	80	Male	PD (1.7)	PMD	No
015	5	63	Female	PD (1.5)	PMD	No
019	3	63	Male	PD (1.6)	SMD	No
003	3	68	Female	SD (3.7)	SMD	Yes
005	2	59	Female	SD (3.7)	SMD	Yes
007	4	83	Male	SD (3.7)	SMD	Yes
008	4	57	Female	SD (5.5)	SMD	Yes
010	5	66	Male	SD (3.8)	PMD	Yes
011	2	65	Female	PD (1.8)	PMD	Yes
012	3	65	Male	PD (1.8)	PMD	Yes
013	5	51	Female	PD (1.6)	PMD	Yes
014	4	65	Female	PR (6.3)	PMR	Yes
016	3	74	Female	PD (1.4)	PMD	Yes
017	5	54	Male	PD (1.7)	PMD	Yes
018	2	62	Male	SD (3.2)	SMD	Yes

Abbreviations: ID, identification; ¹¹¹In-CS-1008, indium-111 labeled to CS-1008; PD, progressive disease; PMD, progressive metabolic disease; PMR, partial metabolic response; SD, stable disease; SMD, stable metabolic disease.

Five cohorts received different doses of CS-1008 as shown in Table 1. There was a strong, positive correlation between protein dose and quantitative tumor uptake (r = .95, P < .001). As shown in Figure 2A, increasing total CS-1008 protein dose on day 1 resulted in a corresponding proportional increase in tumor concentration of ¹¹¹In-CS-1008. Similarly, in the comparison between day 1 and day 36 (Fig 2A and 2B), tumor concentration varied proportionally with the dose, without alteration in biodistribution, which was maintained and consistent with repeat infusions. Intervening cold doses did not change the amount or degree of biodistribution within the same patients. No receptor saturation was seen at the doses examined.

Fig 2.

Quantitative tumor uptake of indium-111 labeled to CS-1008 (¹¹¹In-CS-1008; μg/mL) uptake after (A) week-1 and (B) week-6 infusions. Week-1 infusion showed linear increase in μg/mL ¹¹¹In-CS-1008 uptake in tumors across all dose levels. Week-6 infusion, at 2 mg/kg in all patients, showed consistent CS-1008 uptake compared with week-1 infusion, indicating no death receptor 5 saturation with repeat infusions.

Pharmacokinetics

Pharmacokinetic results (Appendix Table A3, online only) demonstrated a terminal (β) half-life of 6 to 12 days (6 to 8.8 days by ELISA), with total serum clearance ranging from 12 to 18.5 mL/h. Maximum serum concentration (C_max) and area under the serum concentration-time curve (AUC) values were dose proportional, showing a dose-dependent increase. The pharmacokinetic parameters were not significantly affected by repeated drug administration. The ELISA measurements were comparable with the results obtained with ¹¹¹In-CS-1008 (Appendix Table A4, online only) analysis.

Antitumor Activity

All patients were evaluable for antitumor response by RECIST criteria (Table 3). One patient achieved PR lasting 3.7 months (Fig 3). Eight patients had SD lasting a median of 4 months (range, 2.6 to 6.7 months). Ten patients had PD.

Fig 3.

Whole-body [¹⁸F]fluorodeoxyglucose ([¹⁸F]FDG) positron emission tomography (PET) and computed tomography (CT) in patient 014 with a partial response on CT and partial metabolic response on PET. Axial (top row) and coronal (bottom row) images of maximum-intensity projection CT and [¹⁸F]FDG PET images are displayed. Metastatic lesions in the right and left lungs show substantial shrinkage after treatment (reduction in maximum standardized update value, 43% and 38%, respectively), with the shrinkage identified as early as 2 weeks after commencement of treatment with CS-1008.

Metabolic response assessed by [¹⁸F]FDG PET and RECIST response at EOC1 were concordant in 17 patients. Patient 010 had SD on CT scan; however, this patient had a partial metabolic response on FDG PET.

¹¹¹In-CS-1008 Uptake in Tumor

Of the 19 patients entered into the study, seven patients showed no uptake in any tumor site, while 12 patients showed some degree of tumor uptake: three at each of the 1, 2, 4 and 6 mg/kg dose levels. Interestingly, liver metastatic lesions showed generally poor ¹¹¹In-CS-1008 uptake, with only one patient showing definite uptake equivalent or above liver background activity in any hepatic metastatic lesion. No significant differences were observed in visual tumor uptake between day 1 and day 36, and comparison between day-1 infusions also showed no effect of different loading doses on tumor uptake. Importantly, outcomes were notably different in patients whose tumors showed no ¹¹¹In-CS-1008 uptake. Of the 12 patients with ¹¹¹In-CS-1008 tumor uptake, seven patients had overall SD (n = 6) or PR (n = 1), for a clinical benefit rate of 58% (SD + PR). In contrast, five of the seven patients without CS-1008 tumor uptake experienced PD, for a clinical benefit rate of 28% (Fisher's exact test, P = .37).

¹¹¹In-CS-1008 Tumor Uptake in Individual Target Lesions

We also examined the ¹¹¹In-CS-1008 uptake in all reference lesions of each patient and correlated the uptake with the tumor response in those specific lesions. In the 12 patients for whom some uptake was seen, there were 24 evaluable lesions located in lung (n = 13), lymph nodes (n = 5), liver (n = 2), abdominal soft tissue (n = 2), thyroid (n = 1), and rectum (n = 1). Twenty-two of these lesions showed mild, moderate, or markedly increased ¹¹¹In-CS-1008 scans uptake. The seven patients whose tumors showed no uptake had 20 evaluable lesions distributed between liver (n = 13), lung (n = 4), lymph nodes (n = 2), and soft tissue (n = 1).

For the 12 patients whose tumors showed ¹¹¹In-CS-1008 uptake, we found that the lesions with uptake had an 88% probability of being stable or responding to treatment, even in patients with overall PD on restaging. The risk of PD in the lesions with no uptake was approximately three times higher than those lesions with uptake (40% v 12.5%, respectively). Lesions with uptake were one third as likely to progress compared with lesions without uptake, although the correlation between tumor response and CS-1008 uptake did not reach statistical significance (Fisher's exact test, P = .07).

HAHAs

There was no serologic evidence of HAHAs in any patient.

Pharmacodynamic Biomarkers

There was a trend toward an increase in caspase 8 on day 4 and an increase in M30 after sequential infusions. Caspase 3/7 tended to decrease in the postinfusion time points (Appendix Table A5, online only). Serum apoptotic markers were not significantly different between the two groups with and without ¹¹¹In-CS-1008 tumor uptake and between the two groups with and without clinical benefit. CEA levels increased significantly (P = .027) at EOC1 in patients showing PD as compared with patients who had SD or PR.

DR5 Expression

Overall positive expression of DR5 was observed in all tumor specimens (range, 5% to 100%; average, 60%). Five patients had both primary and metastatic tumor available for the analyses of DR5. Metastatic tumor sites had a significantly higher mean H-score than that assigned to primary tumor sites for both membrane (68.6 ± 51.5 v 17.8 ± 17.2, respectively; P = .048) and cytoplasmic (81.8 ± 55.2 v 39.2 ± 38.6, respectively; P = .047) compartments. DR5 expression in archived tumor samples (with use of both H-score and percentage of positive cells) did not correlate with ¹¹¹In-CS-1008 uptake nor with clinical outcome (Appendix Table A6, online only).

Toxicity

CS-1008 was well tolerated. Only one adverse event (grade 1 nausea) was considered possibly related to ¹¹¹In-CS-1008. No serious adverse events were related to study treatment.

DISCUSSION

DR agonists represent a new class of therapeutics that selectively target apoptosis. Several monotherapeutic studies using DR4-DR5 agonist antibodies^7–14 have demonstrated a favorable safety profile at the doses tested, with occasional but sustained responses and prolonged stable disease seen in isolated patients. However, when used in combination with established cancer therapeutics,^15–22 none of the trials involving DR5 agonist antibodies reported to date have achieved their primary end points of improving response rate or progression-free survival. This indicates that agonistic antibodies against DR5 are only active in a subset of patients and that, therefore, it is crucial to identify accurate ways to preselect patients. Moreover, intrapatient molecular heterogeneity within tumors and evolutionary dynamics are critical obstacles for all targeted therapies, and the ability to monitor these factors noninvasively in real time is essential to address this issue optimally.³⁵

In the current study, a novel molecular imaging signature predictive of DR5 agonist efficacy has been identified. We demonstrated that the reference lesions with ¹¹¹In-CS-1008 uptake were less likely to progress even in patients with overall PD at restaging. Furthermore, we observed intra- and interpatient variability in DR5 uptake, with only 12 patients showing ¹¹¹In-CS-1008 uptake in tumor. This variability is likely related to heterogeneous DR5 expression in tumor, although variable antibody penetration due to biophysical properties of individual tumor masses may also influence the concentration of antibody in tumor. A preclinical study⁶ of DR5 occupancy in vivo has shown that ¹¹¹In-CS-1008 uptake correlated both with DR5 expression on tumor cells and the degree of antitumor activity. We also used SPECT imaging in this study, and a PET imaging approach may have allowed greater sensitivity in lesion detection.

In this clinical study, ¹¹¹In-CS-1008 uptake in tumor demonstrated trends toward predicting clinical benefit (SD or PR) on both a per-patient and per-lesion basis, suggesting that lesions that progressed were likely DR5 negative or low DR5 expressers. It is noted that patients were not required to have PD at study entry, and therefore SD may be due to the biologic nature of disease in individual patients.

The ¹¹¹In-CS-1008 biodistribution and dosimetry analyses showed that doses up to 6 mg/kg did not result in DR5 receptor saturation. Repeat ¹¹¹In-CS-1008 infusions also demonstrated no saturation of DR5 receptors, with results that were similar to those observed after the first ¹¹¹In-CS-1008 infusion, indicating a dynamic turnover of receptors on the tumor cell surface. Pharmacokinetic analysis showed proportional increases in C_max and AUC with higher doses, and minor differences in T_1/2α and elimination clearance between dose levels is likely a reflection of patient numbers in each cohort.

In agreement with previous studies,^9,15,16 CS-1008 was well tolerated at doses as high as 6 mg/kg in heavily pretreated patients who had mCRC. Antitumor activity of CS-1008 was observed, with one patient achieving a PR and 8 patients having SD for a median duration of of 4 months.

Although several factors that predict sensitivity and resistance to DR5 agonists have been described in vitro,^36–43 biomarkers that aid in patient selection or predict response to these agents in the clinic remain to be determined. In agreement with a prior study,¹⁴ a relationship between DR5 expression and antitumor activity could not be established in our study, and no relationship was found between DR5 expression and degree of ¹¹¹In-CS-1008 tumor uptake. These findings suggest that it would be premature to use DR5 staining for screening, especially with archival tissue.

Taken together, our data suggest that tumor DR5 expression, assessed with use of molecular imaging of DR5 receptor occupancy (¹¹¹In-CS-1008 imaging) reveals real-time heterogeneous DR5 expression and appears to be a promising predictive imaging biomarker of clinical benefit in patients with mCRC treated with DR5 agonist antibody.

Appendix

Table A1.

Published Clinical Trials of DR5 Agonistss

Agonist	Site	Drug	Outcome	Safety
Conatumumab (C) AMG 655 (fully human monoclonal antibody DR5 agonist)	Advanced solid tumors7	Phase I: single agent	1 PR	No DLTs
	Advanced solid tumors12	Phase I: single agent	No antitumor responses	No DLTs
	NSCLC17	Randomized phase II: paclitaxel and carboplatin + C (3 mg/kg, 15 mg/kg or placebo)	Did not improve PFS	Well tolerated
	Sarcoma18	Phase I/randomized phase II: doxorubicin + C (15 mg/kg or placebo)	Did not improve PFS	Well tolerated
	CRC19	Phase Ib/randomized phase II: mFOLFOX/bevacizumab + C (2 mg/kg, 10 mg/kg or placebo)	Did not improve PFS	Well tolerated
	CRC20	Randomized phase II: FOLFIRI + C (10 mg/kg), ganitumab 12 mg/kg or placebo	Trend toward improved PFS	Well tolerated
	Pancreatic cancer21	Randomized phase II: gemcitabine + C (10 mg/kg), ganitumab 12 mg/kg or placebo	Trend toward improved 6-mo survival rate	Well tolerated
Tigatuzumab (T) CS-1008 (humanized monoclonal antibody DR5 agonist)	Advanced solid tumors9	Phase I: single agent	No antitumor responses	No DLTs
	NSCLC16	Randomized phase II: paclitaxel/carboplatin + T versus placebo	Did not improve PFS	Well tolerated
	Pancreatic cancer15	Single arm phase II: gemcitabine + T (single arm)	PFS similar to historical gemcitabine data	Well tolerated
Lexatumumab (fully human monoclonal antibody DR5 agonist)	Advanced solid tumors14	Phase I: single agent	No antitumor responses	4 DLTs
	Advanced solid tumors11	Phase I: single agent	No antitumor responses	1 DLT
Drozitumab (D) PRO95780 (fully human monoclonal antibody DR5 agonist)	Advanced solid tumors13	Phase I: single agent	No antitumor responses	2 DLTs
	CRC22	Phase Ib: FOLFOX/bevacizumab + D (two dose cohorts)	2 PRs	No DLTs

Abbreviations: CRC, colorectal cancer; DLT, dose-limiting toxicity; DR5, death receptor 5; FOLFOX, fluorouracil, leucovorin, and oxaliplatin; NSCLC, non–small-cell lung cancer; PFS, progression-free survival; PR, partial response.

Table A2.

Quantitative Uptake of ¹¹¹In-CS-1008 in Liver

111In-CS-1008 Infusion	Dose Cohort Group (mg/kg)	Liver Uptake
		%ID/g		μg/mL
		Mean	SD	Mean	SD
Week-1 infusion day 7 or 8 scan	1	.0077	.0024	5.22	.99
	2	.0059	.0000	9.50	.00
	4	.0084	.0034	22.40	7.39
	6	.0068	.0001	34.63	4.29
Week-6 infusion day 42 or 43 scan	1	.0075	.0026	9.85	1.55
	2	.0067	.0000	10.81	.00
	4	.0102	.0024	13.87	.97
	6	.0062	.0006	11.00	2.34

Abbreviations: ¹¹¹In-CS-1008, indium-111 labeled to CS-1008; %ID/g, percent of injected dose per gram.

Table A3.

Mean Pharmacokinetic Parameters After Initial Dose of ¹¹¹In-CS-1008

No. of Patients*	Cohort No. (day-1 dose, mg/kg)	t½α Mean ± SD (hr)	t½β Mean ± SD (hr)	V1 Mean ± SD (mL)	AUC Mean ± SD, (hr × μg/mL)	CL Mean ± SD (mL/h)	Cmax Mean ± SD (μg/mL)
2	1 (0.2)	14.51 ± 5.95	284.76 ± 0.38	3,209.83 ± 321.96	986.58 ± 125.11	14.92 ± 2.50	4.59 ± 0.89
4	2 (1)	21.52 ± 5.76	264.89 ± 122.12	2,592.36 ± 303.21	5,706.38 ± 2,632.72	12.31 ± 3.38	24.60 ± 3.75
5	3 (2)	5.29 ± 4.76	163.08 ± 39.86	3,329.11 ± 624.49	8,386.68 ± 855.70	18.23 ± 0.51	47.29 ± 10.19
2	4 (4)	10.45 ± 6.85	243.39 ± 52.21	2,658.70 ± 112.60	18,714.05 ± 1,511.25	12.00 ± 2.02	84.14 ± 11.00
4	5 (6)	14.73 ± 4.72	247.50 ± 52.90	4,036.84 ± 424.72	28,492.39 ± 1,598.03	18.52 ± 2.61	131.32 ± 20.05
One-way ANOVA, P		.01	.219	.007	< .001	.006	< .001

Abbreviations: ANOVA, analysis of variance; AUC, area under the serum concentration time-curve; CL, total serum clearance; C_max, maximum serum concentration; ¹¹¹In-CS-1008, indium-111 labeled to CS-1008; t_1/2α, half-life of distribution phase of drug; t_1/2β, half-life of elimination phase of drug; V1, volume of central compartment.

^*Patients 007 and 009 were not included in the determination of mean parameter values because of curve fit solution instability.

Table A4.

¹¹¹In-CS-1008 AUC, Clearance, and C_max Compared With CS-1008 Protein (ELISA)

Parameter	Cohort	111In-CS-1008		Serum CS-1008 (ELISA)		P (paired t test)
		Mean	SD	Mean	SD
AUC, hr × μg/mL	1	986.58	125.11	872.03	39.27	.43
	2	5,706.38	2,632.72	7,434.69	3,083.27	.49
	3	8,386.68	855.70	14,445.42	5,160.21	.34
	4	18,714.05	1,511.25	—*	—*	—*
	5	28,492.39	1,598.03	33,683.99	11,876.96	.53
CL, mL/h	1	14.92	2.50	17.16	5.71	.70
	2	12.31	3.38	9.35	2.02	.20
	3	18.23	20.51	10.62	2.46	.14
	4	12.00	2.03	—*	—*	—*
	5	18.52	2.61	16.39	5.05	.57
Cmax, μg/mL	1	4.59	0.89	10.54	7.11	.45
	2	24.60	10.19	31.34	4.74	.07
	3	47.29	10.19	83.58	0.16	.001
	4	84.14	11.00	—*	—*	—*
	5	131.3	20.05	170.20	52.08	.25

Abbreviations: AUC, area under the serum concentration-time curve; CL, total serum clearance; C_max, maximum serum concentration; ELISA, enzyme-linked immunosorbent assay; ¹¹¹In-CS-1008, indium-111 labeled to CS-1008; SD, standard deviation.

^*Outliers with a disproportionate effect on estimated parameters were excluded from data analysis; therefore, because of the lack of data, no statistics could be performed in cohort 4.

Table A5.

Serum Apoptotic Biomarker Levels

Time Points	Caspase 8 (μ/mL)	Caspase 3/7 (μ/mL)	M30 (μ/L)
	(Mean ± SD)	(Mean ± SD)	(Mean ± SD)
Baseline	.21 ± .20	.15 ± .26	490.86 ± 616.89
Day 1, 4 hrs postinfusion	.17 ± .19	.05 ± .04	442.03 ± 625.39
Day 2	.14 ± .13	.06 ± .07	429.63 ± 494.15
Day 4 or 5	.34 ± .66	.07 ± .06	401.22 ± 466.13
Day 36, preinfusion	.23 ± .31	.05 ± .06	714.65 ± 1264.56
Day 36, 4 hrs postinfusion	.21 ± .29	.04 ± .04	553.53 ± 734.82
Day 37	.28 ± .46	.06 ± .08	620.12 ± 979.83
Day 39 or 40	.23 ± .32	.10 ± .15	532.95 ± 616.73
Repeated measures ANOVA, P	.633	.042	.1

Abbreviations: ANOVA, analysis of variance; SD, standard deviation.

Table A6.

DR5 Expression, ¹¹¹In-CS-1008 Uptake, and Antitumor Response

Patient ID	111In-CS-1008 Uptake in Tumor*	DR5 IHC (H-score membrane staining)	DR5 IHC (H-score cytoplasmic staining)	Best Response (RECIST)
Patients with uptake in tumor and overall SD or PR
003	4	20	56	SD
005	4	14	21	SD
007	3	90	127	SD
008	3	63	93	SD
014	3	92	102	PR
018	4	15	110	SD
010	3	93	102	SD
Patients with uptake in tumor and overall PD
011	3	19	25	PD
012	4	70	80	PD
013	4	76	87	PD
016	2	34	68	PD
017	3	21	34	PD
Patients with no uptake in tumor
001	1	30	70	PD
002	1	110	135	SD
004	1	35	60	SD
006	1	90	127	PD
009	1	30	70	PD
015	1	106	124	PD
019	1	0	80	PD

NOTE. No significant associations were found between membrane staining (P > .9) or cytoplasmic staining (P = .5) and ¹¹¹In-CS-1008 tumor uptake. Tumor response was not associated with DR5 expression (P = .6 for membrane staining; P > .9 for cytoplasmic staining).

Abbreviations: DR5, death receptor 5; IHC, immunohistochemistry; ¹¹¹In-CS-1008, indium-111 labeled to CS-1008; PD, progressive disease; PR, partial response; SD, stable disease.

^*Visual grading of maximum intensity of ¹¹¹In-CS-1008 uptake in tumor: 1, no uptake; 2, mild uptake; 3, moderate uptake; 4, marked uptake.

AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

Disclosures provided by the authors are available with this article at www.jco.org.

AUTHOR CONTRIBUTIONS

Conception and design: Archie N. Tse, Mendel Jansen, Ralph Venhaus, Robert A. Beckman, Andrew M. Scott

Administrative support: Ralph Venhaus

Collection and assembly of data: Marika Ciprotti, Niall C. Tebbutt, Fook-Thean Lee, Sze-Ting Lee, Hui K. Gan, David C. McKee, Geoffrey Chong, Wendie Hopkins, Bridget Chappell, Fiona E. Scott, Martin W. Brechbiel, Rira Watanabe, Ralph Venhaus, Jonathan Greenberg, Andrew M. Scott

Data analysis and interpretation: Marika Ciprotti, Niall C. Tebbutt, Fook-Thean Lee, Sze-Ting Lee, Hui K. Gan, Graeme J. O'Keefe, Sylvia J. Gong, Fiona E. Scott, Archie N. Tse, Mendel Jansen, Manabu Matsumura, Masakatsu Kotsuma, Ralph Venhaus, Robert A. Beckman, Andrew M. Scott

Manuscript writing: All authors

Final approval of manuscript: All authors

AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

Phase I Imaging and Pharmacodynamic Trial of CS-1008 in Patients With Metastatic Colorectal Cancer

The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or jco.ascopubs.org/site/ifc.

Marika Ciprotti

Employment: Daiichi Sankyo

Niall C. Tebbutt

No relationship to disclose

Fook-Thean Lee

No relationship to disclose

Sze-Ting Lee

No relationship to disclose

Hui K. Gan

Honoraria: Novartis

Consulting or Advisory Role: Merck Serono

Speakers' Bureau: Merck Serono, Pfizer

Travel, Accommodations, Expenses: Bayer

David C. McKee

No relationship to disclose

Graeme J. O'Keefe

No relationship to disclose

Sylvia J. Gong

No relationship to disclose

Geoffrey Chong

Research Funding: Amgen (Inst), Novartis (Inst), Astellas Pharma (Inst), Incyte (Inst)

Wendie Hopkins

No relationship to disclose

Bridget Chappell

No relationship to disclose

Fiona E. Scott

Research Funding: Daiichi Sankyo (Inst)

Martin W. Brechbiel

No relationship to disclose

Archie N. Tse

Employment: Daiichi Sankyo

Stock or Other Ownership: Daiichi Sankyo

Patents, Royalties, Other Intellectual Property: Daiichi Sankyo

Mendel Jansen

Employment: Daiichi Sankyo

Stock or Other Ownership: Daiichi Sankyo

Manabu Matsumura

Employment: Daiichi Sankyo

Stock or Other Ownership: Daiichi Sankyo

Masakatsu Kotsuma

Employment: Daiichi Sankyo

Rira Watanabe

Employment: Daiichi Sankyo

Ralph Venhaus

Stock or Other Ownership: Agenus, Dynavax, Soligenix

Robert A. Beckman

Employment: Daiichi Sankyo

Stock or Other Ownership: Johnson and Johnson

Patents, Royalties, Other Intellectual Property: Onco-Mind

Other Relationship: Consultant for the Cancer Institute of New Jersey

Jonathan Greenberg

Employment: Daiichi Sankyo

Stock or Other Ownership: Daiichi Sankyo

Andrew M. Scott

Research Funding: Daiichi Sankyo, Abbvie, Avipep

Patents, Royalties, Other Intellectual Property: Patents held