1 Enhanced Delivery of SN-38 to Human Tumor Xenografts with An

Home , Milatuzumab

Author Manuscript Published OnlineFirst on June 23, 2015; DOI: 10.1158/1078-0432.CCR-15-0670 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

Enhanced Delivery of SN-38 to Human Tumor Xenografts with an Anti-Trop-2-SN-38 Antibody Conjugate (Sacituzumab Govitecan)

Robert M. Sharkey, William J. McBride, Thomas M. Cardillo, Serengulam V. Govindan, Yang Wang, Edmund A. Rossi, Chien-Hsing Chang, David M. Goldenberg

Immunomedics, Inc., Morris Plains, NJ 07950

Correspondence should be addressed to: David M. Goldenberg, Immunomedics, Inc., 300 The American Road, Morris Plains, NJ, 07950. E-mail: [email protected]; Telephone: 973-605-8200; Fax: 973-605-8282

Conflict of Interest Disclosure: All authors are employees of Immunomedics, Inc.

Keywords: Antibody-drug conjugate, Trop-2, SN-38

Running Title: Enhanced tumor targeting of SN-38 with sacituzumab govitecan

Word Count

Translational Relevance = 135

Abstract = 248

Introduction through Discussion = 4989

Tables = 3

Figures = 3 (no color)

Supplementary Information provided separately

Downloaded from clincancerres.aacrjournals.org on October 6, 2021. © 2015 American Association for Cancer Research. Author Manuscript Published OnlineFirst on June 23, 2015; DOI: 10.1158/1078-0432.CCR-15-0670 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

Translational Relevance

Sacituzumab govitecan (IMMU-132) is an antibody-drug conjugate composed of a humanized anti-Trop-

2 IgG coupled with a high ratio of SN-38 molecules. This conjugate was shown previously to improve therapeutic responses in human cancer xenografts compared to irinotecan, its prodrug. Recently, clinical trials with IMMU-132 have shown objective responses in several types of Trop-2-expressing epithelial cancers. In this report, we examine the advantage of delivering SN-38 with IMMU-132 compared to irinotecan in nude mice bearing 2 human tumor xenografts, as well as assess the concentrations of SN-38 and SN-38G in the serum, liver and intestine. A significant targeting advantage for IMMU-132 was observed, in terms of the delivery of SN-38 in a fully active form to the tumors, improved tumor/blood ratios, and reduced intestinal uptake that support the reduced rate and severity of diarrhea in patients.

Abstract

Purpose: This study examined the delivery of SN-38 to Trop-2-expressing tumors and assessed the constitutive products in the serum, liver, and small intestine in nude mice bearing human tumor xenografts (Capan-1 or NCI-N87) given a single injection of irinotecan (40 mg/kg; ~0.8 mg/mouse, containing ~460 µg SN-38 equivalents) or sacituzumab govitecan (IMMU-132), an antibody-drug conjugate composed of a humanized anti-Trop-2 IgG coupled site-specifically with an average of 7.6 molecules of SN-38.

Experimental Design: At select times, tissues were extracted and concentrations of the products measured by reversed-phase HPLC.

Results: In serum, >98% irinotecan cleared within 5 min; peak levels of SN-38 and SN-38G

(glucuronidated SN-38) were detected in equal amounts at this time, and no longer detected after 6-8 h.

IMMU-132 was detected in the serum over 3 days, and at each interval, ≥95% of total SN-38 was bound to the antibody. Intact IMMU-132 cleared with a half-life of 14 h, which closely reflected the in vitro rate of SN-38 release from the conjugate in mouse serum (i.e., 17.5 h), while the IgG portion of the conjugate cleared with a half-life of 67.1 h. In vitro and in vivo studies disclosed IgG-bound SN-38 was protected from glucuronidation. Area-under-the-curve (AUC) analysis indicated IMMU-132 delivers 20-fold to as much as 136-fold more SN-38 to tumors than irinotecan, with tumor/blood ratios favoring IMMU-132 by

20- to 40-fold. Intestinal concentrations of SN-38/SN-38G also were 9-fold lower with IMMU-132.

Conclusion: These studies confirm a superior SN-38 tumor delivery by IMMU-132 compared to irinotecan.

Introduction

Antibody-drug conjugates (ADC) represent a new therapy class based on the proposition of their being able to deliver cytotoxic agents more specifically to their intended target with less collateral damage. However, this technology has faced many challenges, but with 3 conjugates gaining FDA approval, each using drugs that are active at picomolar concentrations (so called ultra-toxic drugs), it appeared that the major hurdles facing this technology had been overcome, and that a core platform on which similar agents could be developed was available (1-5). Our group departed from the current popular approach, and instead developed an ADC with a more moderately-toxic agent, SN-38. SN-38, a topoisomerase I inhibitor, is a highly potent drug with activity in the low nanomolar range, administered in a prodrug form, irinotecan. Irinotecan’s pharmacology is well known, but complex, limiting its bioavailability. First, only a small portion of irinotecan is converted to SN-38, a process that occurs primarily in the liver, but also in the intestine and plasma, as well as in some tumors (6). Compounding a poor conversion rate is the fact that a sizable portion of SN-38 is readily converted to less active forms, most notably SN-38G, a glucuronidated product whose concentration is often 4- to 30-times higher than

SN-38 (7). Enterohepatic recirculation exposes SN-38G to bacterial enzymes in the intestine that reconvert SN-38G to SN-38, which is the most likely cause of late diarrhea, a common and serious toxicity associated with irinotecan therapy (8). SN-38’s lactone ring is another vulnerable conversion site, with irinotecan and SN-38 generally found in equilibrium between the less active carboxylate and the fully potent lactone ring forms (6).

Our antibody-SN-38 ADC platform emerged empirically by examining a number of different linkers. Selecting the best conjugate relied on efficacy studies in human tumor xenograft mouse models, with the linker designation CL2 providing the best responses (9, 10). Interestingly, in vitro serum stability studies of the initial three candidate conjugates revealed SN-38 was released at different rates, with one having a half-life of ~10 h, while another was highly stable; yet, the selected CL2-linked

conjugate had an intermediate stability in serum of 36 h. This derivative was later modified to simplify scale-up production for clinical use, with the new derivative designated CL2A (11). Numerous animal studies have shown that IMMU-132 improves therapeutic responses compared to irinotecan, suggesting

IMMU-132 delivers more SN-38 to the tumors (11-13) . The principal goal of this investigation was to quantify this difference, evaluating concentrations of the constitutive products in tumors and critical normal tissues (serum, liver, intestine), in order to gain a better perspective of the efficacy and toxicity properties of this conjugate.

Materials and Methods

Reagents. IMMU-132 was prepared from the hRS7 anti-Trop-2 IgG and the CL2A-SN-38 drug carrier, as described previously (11), with hydrophobic interactive chromatography (HIC) and liquid chromatography mass spectroscopy (LC-MS) showing a drug-antibody ratio of 7.6 (13). Conjugates prepared under these conditions have ~16 µg SN-38/mg IgG. The final lyophilized product was reconstituted in sterile saline immediately before use. Irinotecan was purchased from AREVA pharmaceuticals, Inc. (Elizabethtown, KY), and diluted in sterile saline to the desired concentration.

Standards for SN-38, SN-38G, and irinotecan were purchased from Toronto Research Chemicals

(Toronto, Ontario, Canada).

Study Design. Table 1 lists the 3 sets of studies performed for this analysis, 2 in nude mice bearing xenografts of the human pancreatic cancer cell line, Capan-1, and one with the human gastric carcinoma cell line, NCI-N87. Capan-1 and NCI-N87 were purchased from the American Type Culture Collection

(ATCC) (Manassas, VA), and were authenticated by short tandem repeat (STR) assay by the ATCC and routinely tested for mycoplasma. Surface Trop-2 expression on Capan-1 and NCI-N87 cells in culture are

157,000 and 247,000 molecules, respectively (12).

All animal studies were approved by Rutgers School of Biomedical and Health Sciences Institutional

Animal Care and Use Committee. Initiation of subcutaneous tumors in NCr female athymic nude (nu/nu) mice (4-8 weeks old; Taconic Farms; Germantown, NY) is described in the Supplementary Information.

Mice were administered intravenously with single doses containing a fixed amount of IMMU-132 (1.0 mg), whereas irinotecan was given at a dose of 40 mg/kg, using the average pre-study weight of the animals to determine a fixed dose for that group. By mass, 1.0 mg of irinotecan contains approximately

0.58 mg SN-38. Animals in the first and third study were examined at 5 different intervals, while animals in the 2nd study were examined only at 2-3 intervals. At the time of necropsy, mice were anesthetized and bled by cardiac puncture, with subsequent removal of tissues of interest that were weighed and frozen. Serum samples were analyzed by ELISA for detection of the IMMU-132 and hRS7 IgG (see

Supplementary Information) or by HPLC analysis (below).

HPLC analysis of SN-38, SN-38G, and irinotecan required extraction of the harvested tissues. A precipitating reagent reported by Hirose et al. that contained equal parts of methanol, ethylene glycol, and 1.0 M zinc sulfate was used for extraction/precipitation of proteins (14). Prior to extraction, serum volume was extended by diluting with an equal part of water; water was also added to the tissues (10 parts water to 1 part tissue), which were then homogenized on ice using a Powergen-35 homogenizer equipped with a 7 x 110 mm tip for 2 min at 30,000 rpm (Fisher Scientific, Pittsburgh, PA). An equal volume of the precipitating reagent was added to diluted serum or to a portion of the tissue homogenates prior to separation of solids. Each sample was spiked first with an internal standard, 10- hydroxycaptothecin (10-HCPT; 9 µL 3.33 mg/mL stock solution in DMSO; Toronto Research Chemicals).

The samples were vortexed for 15 sec and then centrifuged to collect the clarified supernatant, which was then then applied on a Phenomenex Chromolith high-resolution C-18 analytical column (EMD

Millipore Corp., Billerica, MA), using a linear gradient composed of 85:15 of buffer A/B to 100% buffer B

(buffer A: 50 mM KH2PO4 and 4 mM sodium 1-decanesulfonate in water, pH 3.5; buffer B: 600 mL buffer

A plus 400 mL acetonitrile), with a flow rate of 1 mL/min and fluorescence detection with excitation wavelength at 373 nm, and emission wavelength of 540 nm.

Portions of the homogenates and diluted serum taken from animals given IMMU-132 were subjected first to an acid hydrolysis step that ensured all SN-38 was released from the IMMU-132 conjugate. For this process, 0.15 mL of 6.0 M HCL was added to 0.15 mL of the homogenate or diluted serum spiked with 10-HCPT, then heated to 50oC overnight before neutralizing with 6.0 M NaOH. An equal volume of the precipitating reagent was added, vortexed, and centrifuged. Thus, the previous processing method measures the amount of free SN-38/SN-38G in the IMMU-132 and irinotecan samples, while the acid-hydrolysis processing isolates the total amount of SN-38/SN-38G specifically in the IMMU-132 sample (designated SN-38 [TOTAL]).

The HPLC analysis method is described in Supplementary Information. AUCs were calculated using GraphPad Prism version 6.0 for Windows (GraphPad Software, La Jolla, CA), using only the trapezoidal area where detectable levels of product were measured in ≥3 consecutive intervals.

Procedures that monitored SN-38 release rates in mouse serum and in vitro glucuronidation are provided in Supplementary Information.

RESULTS

Quantitative HPLC Methodological Assessments

Prior to initiating these studies, various aspects of the procedures used for the analysis of tissue samples were assessed. Details are reported in Supplementary Information (Fig. S1, Tables S1 and S2).

Overall, these studies provided sufficient evidence that the procedures developed for these assessments yielded reliable and reproducible results. The minimum sensitivity of detection in serum was 20 ng/mL and 110 ng/mL in tissue homogenates.

Of particular importance was the finding that the addition of the precipitating reagent to serum freshly spiked with IMMU-132 recovered only ~7% of the expected amount of SN-38 associated with

IMMU-132, indicating that the SN-38 bound to the IgG was not released effectively during the extraction process. Therefore, the acid-hydrolyzing step was introduced, which subsequently showed the expected amount of SN-38 associated with IMMU-132. Furthermore, triplicate samples of serum spiked with SN-

38G that were first processed by acid hydrolysis and then extracted found only 3.0 ± 0.2% of the SN-38G was converted to SN-38, indicating there was a negligible impact on the determination of SN-38G in the acid-hydrolyzed samples.

Concentrations of Products in Serum

Although animals given irinotecan in Study 1 examined serum taken over 24 h, none of the products was detected at this time, and therefore subsequent studies focused on intervals over 6-8 h.

Measurements of IMMU-132 and its SN-38 products were assayed over 72 h, with detection by both

ELISAs and SN-38 [TOTAL] over this entire period of time, but unbound SN-38 and SN-38G were within detectable levels only through 48 and 6 h, respectively (Fig. 1A-D).

In Study 1, irinotecan concentrations at 5 min averaged 1.7 ± 0.2% injected dose per mL (%

ID/mL), indicating that >98% of the irinotecan already had cleared from the blood. Both SN-38 and SN-

38G were detected at 5 min, illustrating rapid conversion and glucuronidation (Fig. 1A). Concentrations of SN-38 and SN-38G were similar over 6 h (AUC5 min-6 h of SN-38 and SN-38G were 2.52 and 2.84

µg/mL·h, respectively) (Supplementary Table S3). The combined AUCs of SN-38 and SN-38G divided by the AUC of irinotecan estimated the conversion rate to be ~25%, confirming the more efficient conversion in mice compared to humans (6, 15, 16).

In the IMMU-132-treated animals, at 1 h SN-38 [TOTAL] concentrations accounted for 58.7 ±

3.3% ID/mL of the estimated amount of SN-38 equivalents. At each interval examined, free SN-38 levels

were <5% of the total SN-38 in the serum (Fig. 1B). The concentration of free SN-38 in the 1-h sample was ~2-times lower than SN-38 levels at this same time in the irinotecan-treated animals. These clearance data are consistent with in vitro studies that indicated 50% of the SN-38 is released from the conjugate in mouse serum every 17.5 h (Supplementary Figure S2A). ELISA data also confirmed these findings, with IMMU-132 having the same rate of clearance as the SN-38 [TOTAL], while IgG levels were persistently higher (Fig. S2B). Pharmacokinetic parameters estimated from the ELISA data show IMMU-

132 clearing with a half-life of 14 h (Fig. S2C), reflecting the rate at which SN-38 is released from the conjugate (i.e., 17.5 h) plus the rate of IgG removal from the blood.

An important difference between irinotecan- and IMMU-132-treated animals was that there were only trace amounts of unbound SN-38G in the serum of animals given IMMU-132, with levels in the 1- and 6-h samples being 5- to 10-fold less than unbound SN-38; no free SN-38G was detected in the later samples. More importantly, at no time did the acid-hydrolyzed samples have detectable amounts of SN-38G, suggesting that all of the SN-38G generated was likely from the released SN-38, with the SN-

38 bound to the conjugate not susceptible to glucuronidation, despite the fact that the hydroxyl at the

10th position was available for this modification to occur.

The apparent protection of SN-38 from glucuronidation while bound to the conjugate was investigated further using an in vitro glucuronidation assay. HPLC analysis of the non-hydrolyzed SN-38 sample found 47% of the recovered products had been converted to SN-38G (recovery = 70%)

(Supplementary Table S4). For the IMMU-132 samples, there were 16.1 nM of SN-38 plus 7.1 nM SN-

38G, which based on initial concentration of 425.3 nM, the combined products represent a release of

5.5% of the initially bound SN-38. This indicates that the reaction conditions did not appreciably release

SN-38 from the conjugate. The acid-hydrolyzed IMMU-132 samples contained a total of 452.6 nM of SN-

38 and 4.5 nM of SN-38G, or just 1% of the total recovered product being in the SN-38G form (recovery

= 107%); however, since the amount of SN-38G in the acid-hydrolyzed sample was less than in non-

hydrolyzed samples, these results reveal there was no detectable SN-38G derived from the SN-38 bound to the conjugate, thereby confirming the in vivo data that SN-38 bound to IMMU-132 is protected from glucuronidation.

In Study 2, irinotecan and IMMU-132 were given to nude mice bearing larger-sized Capan-1, examining 2 of the same intervals for irinotecan (1 and 6 h) and 3 intervals (1, 6, and 24 h) for IMMU-

132 (Supplementary Fig. S3). Concentrations of all products in the serum generally were similar in these two studies, but unbound SN-38G was detected in the 24-h serum samples in 2 of the 3 animals in the

IMMU-132 group of Study 2 (21.1 and 56.5 ng/mL). Again, no SN-38G was detected in the acid- hydrolyzed serum samples at any time.

Although animals bearing NCI-N87 human gastric cancer xenografts and given irinotecan had differences in their clearance kinetics (i.e., in Study 3, the 3 products decreased ~90% from 5 min to 6 h;

Study 1, no more than 25% decrease over this same period), the AUCs for all 3 products were similar in both studies (Fig. 1A and C). Concentrations and clearance of the products from the IMMU-132-treated animals in Study 1 and 3 were similar. Overall, the serum clearance of the various constitutive products for irinotecan in all 3 of these studies was similar, as was the clearance of IMMU-132 and its products, but there were fundamental differences in the concentrations of these products in these 2 groups.

Concentrations of Products in Tumor

In Study 1, Capan-1 tumors were of similar size as used in previous IMMU-132 therapy studies

(11, 12), with the average tumor weights in the 15 animals given irinotecan being 0.113 ± 0.035 g, and in the IMMU-132-treated animals, 0.149 ± 0.064 g. Study 2 utilized Capan-1 tumors that were approximately twice the size as the first study to determine if this impacted tumor localization (6 irinotecan-treated animals, 0.365 ± 0.085 g; 9 IMMU-132-treated animals, 0.342 ± 0.125 g).

Concentrations of all products in the Capan-1 tumors from the irinotecan-treated animals were similar between the 2 studies (Fig. 2A). For example, irinotecan levels at 1 h in the first and second studies averaged 12.34 ± 5.23 µg/g (1.6 ± 0.7% ID/g) and 11.08 ± 0.14 µg/g (1.4 ± 0.02% ID/g), respectively, with SN-38 concentrations of 0.19 ± 0.05 µg/g and 0.21 ± 0.01 µg/g, respectively. Whereas in Study 1, where SN-38 was detectable in the tumors only through 2 h, in Study 2 several of the 6-h tumor samples had detectable SN-38 or SN-38G. AUC determinations for Study 1 alone and using data combined from both studies, which allowed SN-38 AUCs to be extrapolated over 6 h, found that only about 5% of the irinotecan in the tumor was converted to SN-38+SN-38G (Table 2), with about half being in the form of SN-38G, whereas in the serum, concentrations of irinotecan and SN-38 estimated the conversion rate to be ~25%. In the NCI-N87 tumors taken from the irinotecan-treated animals

(average weight = 0.31 ± 0.08 g, N = 15), peak levels of irinotecan remained at ~10 µg/g (~1.2% ID/g) for the first 2 h before decreasing to 1.69 ± 0.34 µg/g at 6 h (Fig. 2B). SN-38 concentrations were at their highest level at 5 min (0.35 ± 0.08 µg/g, representing 0.07 ± 0.02% ID/g of the total SN-38 equivalents in the administered irinotecan dose), and remained relatively constant over the 8-h sampling period at a level of about 0.25 µg/g. SN-38G was initially 40% lower than the SN-38 levels, and was undetectable at

6 and 8 h. While the absolute levels of SN-38 in the NCI-N87 tumors overall were not substantially different from Capan-1, the sustained detection of SN-38 in the NCI-N87 tumors over 8 h improved the

AUC by nearly 2-fold (Table 2).

In the IMMU-132-treated animals, the only product detected in the tumors at any interval was

SN-38 [TOTAL], indicating that all of the SN-38 residing in the tumor was bound to the conjugate (Fig. 2C and D). Concentrations in the Capan-1 tumors peaked at 6 h in both studies, averaging 1.88 ± 0.57 µg/g and 1.26 ± 0.36 µg/g (11.8 ± 3.6 and 7.9 ± 2.3% ID/g) in Study 1 and 2, respectively. Overall, a comparison of Capan-1 tumor uptake in the 2 studies again found generally similar values, with a tendency for the smaller tumors in Study 1 to have slightly higher concentrations on a per gram basis

(Figure 2D). Comparing the AUC for SN-38 content in the tumors from the first study for irinotecan- and

IMMU-132-treated animals, the data suggest as much as 136-fold higher concentrations of SN-38 could be delivered to the tumors using IMMU-132 (Table 2), but when the data from both studies are combined, the IMMU-132-treated animals had ~45-fold delivery advantage over irinotecan therapy. In the NCI-N87 xenografts (0.32 ± 0.16 g), peak levels were again found at 6 h, with an average of 1.28 ±

0.65 µg/g (6.4 ± 3.3% ID/g). Although the SN-38 [TOTAL] AUC for the NCI-N87 tumors was similar to that in Capan-1 tumors, the higher SN-38 AUC in the NCI-N87 tumors for the irinotecan-treated animals resulted in a lower, yet still an appreciably better targeting ratio of ~21:1. Since the clearance of SN-38 derived by irinotecan or IMMU-132 was similar in all 3 studies, the average AUCs of all the products in the serum is shown in Table 3. Using the combined AUCs for Capan-1 from Studies 1 and 2, we find the tumor/serum ratio for irinotecan-delivered SN-38 was 0.33, while in NCI-N87, the ratio was 0.65. In contrast, these same ratios for SN-38 delivered by IMMU-132 showed a 12.7- and 11.3-fold advantage for Capan-1 and NCI-N87, respectively.

Concentrations of Products in Liver and Intestinal Contents

In the 5-min liver homogenate sample, irinotecan concentrations were 91.72 ± 11.55 µg/g (11.8

± 1.5% of injected dose per gram), with 3.79 ± 0.51 and 0.62 ± 0.05 µg/g of SN-38 and SN-38G, respectively. All of these products cleared quickly over the first 6 h, with no products detected at 24 h

(Supplementary Fig. S4A and B).

SN-38 [TOTAL] was the primary form detected in the liver, but only detectable through 24 h, starting with 1.34 ± 0.13 µg/g (6.7 ± 0.7%) in the 1-h sample (Figure S4A). Free SN-38 in the liver homogenates of animals given IMMU-132 was only detected in the 1-h sample (140.7 ± 56.2 ng/g; 0.7 ±

0.3% of SN-38 equivalents injected). This compares to 1.57 ± 0.51 µg/g of SN-38 found in the 1-h liver sample of animals given irinotecan (0.4 ± 0.1% of administered SN-38 equivalents). SN-38G was not

detected in the acid-hydrolyzed or non-hydrolyzed liver homogenates from animals given IMMU-132. In

Study 2, concentrations of the various products were somewhat higher, but the pattern of uptake and clearance for each of the products in both the irinotecan- and IMMU-132-treated animals were similar

(Supplementary Fig. S4C).

Animals given irinotecan had detectable products in the small intestine even at 5 min after injection, peaking at 1 h (Figure 3). Combining data from Studies 1 and 2, irinotecan levels at 1 h were

90.6.30 ± 23.18 µg (~11% ID; AUC = 345.7 µg·h). The AUC for SN-38G in the small intestinal contents exceeded SN-38 AUC by nearly 18-fold (149.9 µg·h vs. 8.2 µg·h, respectively).

With IMMU-132, SN-38 concentrations were similar in the acid-hydrolyzed and non-hydrolyzed samples, indicating that the intact conjugate was not transported from the liver to the intestine. For example, in a 6-h fecal sample, unbound SN-38 levels were 0.14 ± 0.04 vs. 0.23 ± 0.07 µg for the non- hydrolyzed and hydrolyzed samples, respectively (2-tailed paired t-test; P = 0.053). The AUC for SN-38G was ~4-fold higher than SN-38 (14.3 µg·h vs. 3.3 µg·h). However, the combined AUC for SN-38 and SN-

38G in the IMMU-132-treated animals was still 9-fold lower than the combined AUC in the irinotecan- treated animals. This difference is even greater if the irinotecan AUC is added to the total amount of product in the intestine, since a portion of it could convert to SN-38, resulting in toxicity.

Irinotecan-treated animals also had high levels of irinotecan and SN-38 in the large intestine, with SN-38 levels being nearly 20-fold higher at 6 h than in animals given IMMU-132. SN-38G levels in animals given irinotecan were much lower than measured in the small intestine, likely reflecting the known conversion that occurs as a result of bacterial enzymes (8, 17-19) (Supplementary Table S5).

Discussion

SN-38 is a highly potent drug, but delivery by irinotecan is compromised by (a) rapid clearance from the blood, (b) poor conversion rate, and (c) a rapid conversion to the inactive SN-38G form. Thus, we developed a new type of antibody-drug conjugate that utilizes SN-38, expecting that this ADC would improve SN-38’s pharmacokinetics, but also provide a mechanism for selective tumor retention via the targeting antibody. However, several other advantages were found. First, SN-38 bound to the antibody is in an inactive state, and therefore the intact conjugate in the serum would not pose a toxicity concern.

Secondly, others had indicated that by coupling an agent to the 20th position, the lactone ring’s integrity is preserved (20). Conversion of the lactone ring to the less toxic carboxylate form accounts for as much as 30-40% of the total SN-38 after a 90-min irinotecan infusion, and this increases to as much as two- thirds at equilibrium (6). While we have not assessed lactone ring stability with IMMU-132, coupling to the 20th position should preserve lactone ring integrity. Finally, in vitro and in vivo studies found that

SN-38 bound to the conjugate is not susceptible to glucuronidation, which also significantly reduces SN-

38’s potency, with SN-38G concentrations in serum being 4- to as much as 30-times higher than SN-38 with irinotecan therapy (7). Thus, IMMU-132 retains SN-38 in its fully active form as it circulates in the serum. The IMMU-132 fraction that localizes in the tumor will have the opportunity to be internalized by the action of antibody binding to Trop-2, but because the antibody would hold the conjugate in the tumor for a sustained period, SN-38 also can be released locally, where its ability to be transported quickly across membranes would represent another path for internalization. SN-38 released from the conjugate elsewhere in the body, we anticipate, will behave in the same manner as SN-38 released from irinotecan.

Although the more recently approved ultra-toxic ADCs utilize linkers that bind the drug stably while in serum, since premature release of those agents would increase toxicity and reduce their therapeutic window, our in vivo experience revealed the optimal therapeutic activity for an SN-38 ADC

required a linker that neither held the drug stably nor released it too early (9, 10, 21). This conjugate system also permits the site-specific coupling of nearly 8 molecules of SN-38 per IgG without affecting the physiochemical nature, immunoreactivity, or the pharmacokinetics of the conjugate compared to unconjugated IgG (12). Conjugates using ultra-toxic agents are generally substituted with about half this amount of drug, mainly because higher substitution levels were found to alter pharmacokinetics and reduce the therapeutic index (22, 23).

In addition to these factors, IMMU-132’s delivery advantage can be highlighted in several other ways. First, a single dose of 40 mg/kg of irinotecan (~0.8 mg) administered in this study carried ~30-fold more SN-38 equivalents than IMMU-132 administered at a single 20 mg/kg dose (1.0 mg IMMU-132 containing 16 µg SN-38), yet by AUC comparisons, IMMU-132 delivered from 20-fold to as much as 136- fold more SN-38 into the tumor than irinotecan. Comparing the AUCs of SN-38 in the tumor to AUCs of free SN-38 in the serum also favored IMMU-132 by 20- to 40-fold. The single dose of irinotecan given to the mice, when converted to a human equivalent dose, is 3.25 mg/kg or 126 mg/m2 for a 70 kg/1.8 m2 patient, which is 2.8-fold lower than the recommend 350 mg/m2 for a single dose of irinotecan given once every 3 weeks. In contrast, the human equivalent dose of IMMU-132 is 4.1 mg/kg, yet IMMU-

132’s recommended dose level for Phase II clinical trials is 10 mg/kg weekly x 2 every 3 weeks, or ~5-fold higher equivalent to the dose given to mice. Thus, if the tumor AUC for irinotecan and IMMU-132 were adjusted to a human equivalent dose given weekly x 2 every 3 weeks, the IMMU-132 treatment would deliver as much as 40- to 220-times the amount of SN-38 as the irinotecan dose. However, this difference might be even higher in patients, since (a) humans are at least 5-fold less efficient than mice in converting irinotecan to SN-38, and therefore more SN-38 was available in the mice examined than would be in humans, (b) humans generate much more inactive SN-38G than SN-38 (mice had equal amounts of SN-38G and SN-38, whereas humans have at least 4-fold more SN-38G than SN-38) (7), which might be expected to alter the balance of SN-38 and SN-38G in the tumors between mice and

humans, and (c) although not measured in this study, a sizable portion of SN-38 derived from irinotecan is in the inactive carboxylate form, whereas SN-38 delivered to the tumor by IMMU-132 would be expected to be released in the active lactone form due to linker coupling to the 20th position on the lactone ring (20).

Both the Capan-1 human pancreatic and the NCI-N87 human gastric cancer cell lines express relatively high levels of Trop-2 (12), and in vivo studies have shown that both respond better to the

IMMU-132 treatment than irinotecan (11, 12). Peak concentrations of SN-38 in these tumors occurred between 1-6 h, and despite the 1.6-fold higher expression of Trop-2 in NCI-N87, tumor uptake of IMMU-

132 was similar to the Capan-1 tumors. Levels of SN-38 were found in the tumor of animals given

IMMU-132, and since the topoisomerase I activity of SN-38 is enhanced in S-phase cells, maintaining an

SN-38 presence for 3 days compared to 8 h would be another advantage. Since resistance to irinotecan therapy can occur for a variety of reasons (24-27), it is uncertain whether the ability to enhance SN-38 delivery with this ADC will impact resistance in a clinically meaningful manner. We are currently in the process of developing SN-38-resistant cell lines that should provide some insights.

Other agents designed to improve the bioavailability of SN-38 have been reported; two of these have been examined clinically. EZN-2208 was a PEG-conjugated SN-38, which couples a branched PEG moiety to the lactone ring of SN-38 and stabilized the lactone ring, but its linkage chemistry appeared to have ~12 min stability in human serum (28). While providing improved responses in animal models, clinically the agent failed to provide an indication of clinical benefit (29). Etirinotecan pegol is a PEG- conjugate composed of a 4-arm, branched PEG, with each arm harboring irinotecan, which has been reported to extend SN-38’s half-life to ~50 h (30-33). Preclinical studies in rats showed etirinotecan pegol serum concentrations decreasing from ~700 to 30 µg/mL over 12 h, with low levels (~30 ng/mL) of both irinotecan and SN-38 detected within the first few minutes; SN-38 concentrations then remained between 10 to 20 ng/mL over 14 days (31). Comparing 3 doses of etirinotecan pegol or irinotecan (days

0, 4, and 8) found that SN-38 concentrations in HT-29 tumor xenografts were 300-fold higher than with irinotecan (31), but SN-38 concentrations in the tumor were still ~17-fold lower than etirinotecan pegol, suggesting a ~5% conversion rate at the tumor site. SN-38G concentrations were not reported in this preclinical study, but clinical results have indicated substantially higher levels of SN-38G than SN-38.

Promising anti-tumor responses have been reported for etirinotecan pegol clinically, with dose-limiting toxicity being severe diarrhea that is not manifested until a median of 63 days from the start of therapy

(32, 33). Clinical studies with IMMU-132 found dose-limiting neutropenia occurring within 1 week of the start of treatment, but the incidence of severe diarrhea is greatly reduced. Diarrhea from irinotecan therapy can occur as a result of intestinal transport of irinotecan, SN-38 and SN-38G, with enterohepatic recirculation of SN-38G thought to be the primary cause of late diarrhea. With IMMU-132, free SN-38 concentrations in the intestine were as much as 20-times lower in mice. Since SN-38G levels in the serum are also very low with IMMU-132, we suspect that this reduces the pool of drug available for enterohepatic recirculation. These two factors likely combine to explain the lower incidence and severity of diarrhea observed clinically with IMMU-132 than other agents delivering SN-38 (34-37).

Concentrations of the intact conjugate in the serum, as monitored by an anti-SN-38 capture x anti-idiotype antibody probe ELISA or by total SN-38 using HPLC, showed IMMU-132 cleared at a rate that was slightly faster than SN-38 released from the conjugate when held in serum in vitro. This was expected, since the IgG clearance would contribute to the overall rate in vivo. ELISA monitoring of IgG in the serum found it clears more slowly, staying in the circulation for a longer period. Similar results are being found in patient samples (34-37), where (a) the IgG component clears more slowly than the conjugate, (b) >95% of the total SN-38 in the serum is bound to the IgG, (c) the amount of free SN-38G in the serum is only a fraction of the free SN-38, and (d) there is no evidence of glucuronidation of SN-38 bound to IgG. Analysis of serum samples taken from patients given 8 to 10 mg/kg of IMMU-132 on days

1 and 8 of 21-day cycles has found free SN-38 concentrations at levels of ~100 ng/mL 30 minutes after

the end of a 2-3 h infusion, while the total SN-38 levels average ~4000 ng/mL (34). These levels decrease steadily, being undetectable within 2-3 days; SN-38G levels in sera never exceeded SN-38.

Thus, we suspect the levels of free SN-38 in the serum of patients receiving IMMU-132 therapy, which are similar to the peak levels of SN-38 seen in patients given irinotecan therapy (18), contribute to the dose-limiting neutropenia in patients given IMMU-132, whereas the significantly lower levels of SN-38G with IMMU-132 reduces the incidence of severe diarrhea.

In conclusion, this ADC platform departs from the current practice using stably-linked ultra-toxic agents (subnanomolar), employing instead a moderately-toxic agent (low nanomolar) that is released in serum. IMMU-132 was found to deliver much higher levels of SN-38 to tumors than irinotecan and, importantly, all of the SN-38 delivered to the tumor by IMMU-132 is released in its most potent form. In relationship to gastrointestinal toxicity, the much lower amounts of SN-38G in the serum and significantly lower amounts of SN-38/SN38G in the intestine with IMMU-132 are expected to reduce the risk for severe diarrhea in patients, which is confirmed in clinical studies (34-38). Thus, IMMU-132 is an

ADC that appears to have unique properties that combine to make an effective therapeutic agent in the solid cancers tested.

Acknowledgements

We thank Jennifer Inghrim, Nalini Sathyanarayan, Jayson Jebson, Roberto Arrojo, and Preeti Trisal, Maria

Zalath for their technical assistance.

References

1. Bouchard H, Viskov C, Garcia-Echeverria C. Antibody-drug conjugates-A new wave of cancer drugs. Bioorganic & medicinal chemistry letters. 2014;24:5357-63.

2. Drachman JG, Senter PD. Antibody-drug conjugates: the chemistry behind empowering antibodies to fight cancer. Hematology / the Education Program of the American Society of Hematology

American Society of Hematology Education Program. 2013;2013:306-10.

3. Govindan SV, Goldenberg DM. Designing immunoconjugates for cancer therapy. Expert Opin

Biol Ther. 2012;12:873-90.

4. Leal M, Sapra P, Hurvitz SA, Senter P, Wahl A, Schutten M, et al. Antibody-drug conjugates: an emerging modality for the treatment of cancer. Annals of the New York Academy of Sciences.

2014;1321:41-54.

5. Panowksi S, Bhakta S, Raab H, Polakis P, Junutula JR. Site-specific antibody drug conjugates for cancer therapy. mAbs. 2014;6:34-45.

6. Mathijssen RH, van Alphen RJ, Verweij J, Loos WJ, Nooter K, Stoter G, et al. Clinical pharmacokinetics and metabolism of irinotecan (CPT-11). Clin Cancer Res. 2001;7:2182-94.

7. Xie R, Mathijssen RH, Sparreboom A, Verweij J, Karlsson MO. Clinical pharmacokinetics of irinotecan and its metabolites: a population analysis. J Clin Oncol. 2002;20:3293-301.

8. Stein A, Voigt W, Jordan K. Chemotherapy-induced diarrhea: pathophysiology, frequency and guideline-based management. Ther Adv Med Oncol. 2010;2:51-63.

9. Govindan SV, Cardillo TM, Moon SJ, Hansen HJ, Goldenberg DM. CEACAM5-targeted therapy of human colonic and pancreatic cancer xenografts with potent labetuzumab-SN-38 immunoconjugates.

Clin Cancer Res. 2009;15:6052-61.

10. Moon SJ, Govindan SV, Cardillo TM, D'Souza CA, Hansen HJ, Goldenberg DM. Antibody conjugates of 7-ethyl-10-hydroxycamptothecin (SN-38) for targeted cancer chemotherapy. J Med Chem.

2008;51:6916-26.

11. Cardillo TM, Govindan SV, Sharkey RM, Trisal P, Goldenberg DM. Humanized anti-Trop-2 IgG-SN-

38 conjugate for effective treatment of diverse epithelial cancers: preclinical studies in human cancer xenograft models and monkeys. Clin Cancer Res. 2011;17:3157-69.

12. Cardillo TM, Govindan SV, Sharkey RM, Trisal P, Arrojo R, Chang C-H, et al. Sacituzumab govitecan (IMMU-132, an anti-Trop-2-SN-38 antibody-drug conjugate: Characterization and efficacy in pancreatic and gastic cancers. Bioconjug Chem. 2015;(in press).

13. Goldenberg DM, Rossi EA, Govindan SV, Cardillo TM, McBride WJ, Zalath M, et al.

Characterization of an anti-Trop-2-SN-38 antibody-drug conjugate (IMMU-132) with potent activity against solid cancers. J Clin Oncol. 2014;32:suppl; abstr 3107.

14. Hirose K, Kozu C, Yamashita K, Maruo E, Kitamura M, Hasegawa J, et al. Correlation between plasma concentration ratios of SN-38 glucuronide and SN-38 and neutropenia induction in patients with colorectal cancer and wild-type UGT1A1 gene. Oncol Lett. 2012;3:694-8.

15. Morton CL, Wierdl M, Oliver L, Ma MK, Danks MK, Stewart CF, et al. Activation of CPT-11 in mice: identification and analysis of a highly effective plasma esterase. Cancer Res. 2000;60:4206-10.

16. Zamboni WC, Stewart CF, Cheshire PJ, Richmond LB, Hanna SK, Luo X, et al. Studies of the efficacy and pharmacology of irinotecan against human colon tumor xenograft models. Clin Cancer Res.

1998;4:743-53.

17. Gupta E, Mick R, Ramirez J, Wang X, Lestingi TM, Vokes EE, et al. Pharmacokinetic and pharmacodynamic evaluation of the topoisomerase inhibitor irinotecan in cancer patients. J Clin Oncol.

1997;15:1502-10.

18. Xie R, Mathijssen RH, Sparreboom A, Verweij J, Karlsson MO. Clinical pharmacokinetics of irinotecan and its metabolites in relation with diarrhea. Clin Pharmacol Ther. 2002;72:265-75.

19. Yamamoto M, Kurita A, Asahara T, Takakura A, Katono K, Iwasaki M, et al. Metabolism of irinotecan and its active metabolite SN-38 by intestinal microflora in rats. Oncol Rep. 2008;20:727-30.

20. Zhao H, Lee C, Sai P, Choe YH, Boro M, Pendri A, et al. 20-O-acylcamptothecin derivatives: evidence for lactone stabilization 1. J Org Chem. 2000;65:4601-6.

21. Govindan SV, Cardillo TM, Sharkey RM, Tat F, Gold DV, Goldenberg DM. Milatuzumab-SN-38 conjugates for the treatment of CD74+ cancers. Mol Cancer Ther. 2013;12:968-78.

22. Lewis Phillips GD, Li G, Dugger DL, Crocker LM, Parsons KL, Mai E, et al. Targeting HER2-positive breast cancer with trastuzumab-DM1, an antibody-cytotoxic drug conjugate. Cancer Res. 2008;68:9280-

90.

23. Erickson HK, Lewis Phillips GD, Leipold DD, Provenzano CA, Mai E, Johnson HA, et al. The effect of different linkers on target cell catabolism and pharmacokinetics/pharmacodynamics of trastuzumab maytansinoid conjugates. Mol Cancer Ther. 2012;11:1133-42.

24. Garcia-Carbonero R, Supko JG. Current perspectives on the clinical experience, pharmacology, and continued development of the camptothecins. Clin Cancer Res. 2002;8:641-61.

25. Rasheed ZA, Rubin EH. Mechanisms of resistance to topoisomerase I-targeting drugs. Oncogene.

2003;22:7296-304.

26. Gongora C, Vezzio-Vie N, Tuduri S, Denis V, Causse A, Auzanneau C, et al. New Topoisomerase I mutations are associated with resistance to camptothecin. Mol Cancer. 2011;10:64.

27. Ashour ME, Atteya R, El-Khamisy SF. Topoisomerase-mediated chromosomal break repair: an emerging player in many games. Nat Rev Cancer. 2015;15:137-51.

28. Zhao H, Rubio B, Sapra P, Wu D, Reddy P, Sai P, et al. Novel prodrugs of SN38 using multiarm poly(ethylene glycol) linkers. Bioconjug Chem. 2008;19:849-59.

29. Garrett CR, Bekaii-Saab TS, Ryan T, Fisher GA, Clive S, Kavan P, et al. Randomized phase 2 study of pegylated SN-38 (EZN-2208) or irinotecan plus cetuximab in patients with advanced colorectal cancer.

Cancer. 2013;119:4223-30.

30. Alemany C. Etirinotecan pegol: development of a novel conjugated topoisomerase I inhibitor.

Curr Oncol Rep. 2014;16:367.

31. Hoch U, Staschen CM, Johnson RK, Eldon MA. Nonclinical pharmacokinetics and activity of etirinotecan pegol (NKTR-102), a long-acting topoisomerase 1 inhibitor, in multiple cancer models.

Cancer Chemother Pharmacol. 2014;74:1125-37.

32. Awada A, Garcia AA, Chan S, Jerusalem GH, Coleman RE, Huizing MT, et al. Two schedules of etirinotecan pegol (NKTR-102) in patients with previously treated metastatic breast cancer: a randomised phase 2 study. Lancet Oncol. 2013;14:1216-25.

33. Vergote IB, Garcia A, Micha J, Pippitt C, Bendell J, Spitz D, et al. Randomized multicenter phase II trial comparing two schedules of etirinotecan pegol (NKTR-102) in women with recurrent platinum- resistant/refractory epithelial ovarian cancer. J Clin Oncol. 2013;31:4060-6.

34. Goldenberg DM, Vahdat LT, Starodub A, Bardia A, Chuang E, Moroose RL, et al. IMMU-132, a potential new antibody-drug conjugate (ADC) for the treatment of triple-negative breast cancer (TNBC):

Preclinical and initial clinical results. San Antonio Breast Cancer Symposium. 2014:P5-19-08.

35. Bardia A, Starodub A, Moroose RL, Mayer IA, Diamond JR, Chuang E, et al. IMMU-132, a new antibody-drug conjugate (ADC) against Trop-2, as a novel therapeutic for patients with relapsed/refractory, metastatic, triple-negative breast cancer (TNBC): Results from Phase I/II clinical trial

(NCT01631552). San Antonio Breast Cancer Symposium. 2014:P5-19-27

36. Starodub AN, Ocean AO, Shah MA, Guarino MJ, Picozzi VJ, Vahdat LT, et al. First-in-human trial of a novel anti-Trop-2 antibody-SN-38 conjugate, sacituzumab govitecan, for the treatment of diverse metastatic solid tumors. Clin Cancer Res. 2015;[Epub ahead of print] May 5. pii: clincanres.3321.2014.

37. Starodub AN, Ocean AO, Shah MA, Vahdat LT, Chuang E, Guarino MJ, et al. SN-38 antibody-drug conjugate (ADC) targeting Trop-2, IMMU-132, as a novel platform for the therapy of diverse metastatic solid cancers: Initial clinical results. Cancer Res. 2014;74:abstract CT206.

38. Goldenberg DM, Cardillo TM, Govindan SV, Rossi EA, Sharkey RM. Trop-2 is a new candidate target for solid cancer therapy with sacituzumab govitecan (IMMU-132), an antibody-drug conjugate

(ADC). (submitted for publication). 2015.

Table 1. Examination of SN-38 concentrations in tissues of nude mice bearing 2 human cancer xenografts. IMMU-132 Irinotecan Intervals (SN-38 (SN-38 Study Tumor (N = 3 animals/interval) Tissues examined equivalents)a equivalents)a Irinotecan Tumor, serum, liver, 5 min, 1, 2, 6, and 24 h 1.0 mg 773 µg 1b Capan-1 small intestine IMMU-132 (16 µg) (448 µg) contents 1, 6, 24, 48, 72 h Irinotecan Tumor, serum, liver, 1 and 6 h 1.0 mg 808 µg 2 Capan-1 small and large IMMU-132 (16 µg) (468 µg) intestine contents 1, 6, and 24 h

Irinotecan 5 min, 1, 2, 6, and 8 h 1.0 mg 840 µg 3 NCI-N87 Tumor, serum IMMU-132 (16 µg) (486 µg) 1, 6, 24, 48, 72 h a SN-38 equivalents for IMMU-132 based on spectrophotometric determinations of protein and SN-38 concentrations. SN-38 equivalents for irinotecan based on mass, with SN-38 representing approximately 58% of irinotecan’s mass. .

Table 2. Concentrations of products over time (AUC) in mice bearing Capan-1 or NCI-N87 tumors administered irinotecan or IMMU-132a.

Irinotecan-treated IMMU-132-treated SN-38 deliver ratio Tissue SN-38 SN-38G Irinotecan SN-38 [TOTAL] IMMU-132/Irinotecan Capan-1 Study 1 only 0.40 1.08 48.37 54.25 135.6 [Study 1+2] [1.09] [1.04] [46.73] [49.02] [45.0] NCI-N87 2.11 0.31 45.69 43.88 20.8 a AUC are expressed as (µg/g·h). In Study 1, SN-38 AUC derived from 5 min – 2 h; SN-38G and irinotecan from 5 min to 6 h. AUC from 5 min to 6 h for irinotecan-treated mice combining Studies 1 and 2. For IMMU-132-treated animals SN-38 [TOTAL] AUC are derived from 1 h to 72 h.

Table 3. Determination of tumor/serum AUC ratiosa. Capan-1 Tumor/serum AUC ratio Serum [Study 1+2] NCI-N87 Capan-1 NCI-N87 Irinotecan-treated SN-38 3.27 1.09 2.11 0.33 0.65 SN-38G 4.02 Irinotecan 27.48

IMMU-132-treated Free SN-38 3.87 49.02 43.88 12.7 11.3 Free SN-38G 1.00 SN-38 [TOTAL] 157.21 aAUCs, expressed as µg/g·h were averaged from all 3 studies, with values derived from 5 min to 8 h for irinotecan-treated animals or from 1 h to 72 h for IMMU-132-treated animals.

Figure Legends

Figure 1. Clearance of constitutive products in tumor-bearing nude mice given irinotecan or IMMU-

132. (A) Study 1, Capan-1-beaing mice; irinotecan. (B) Study 1; IMMU-132. (C) Study 3, NCI-N87-bearing mice; irinotecan. (D) Study 3; IMMU-132. Area-under-the-curve (AUC) estimates are provided in the table for products with 3 or more samples.

Figure 2. Concentrations of products extracted from Capan-1 and NCI-N87 tumors isolated from animals given irinotecan or IMMU-132. (A) Concentrations of SN-38, SN-38G and irinotecan in Capan-1 tumors from mice given irinotecan in Study 1 and Study 2. In Study 1, SN-38 was below the threshold of detection in the 6-h sample, whereas SN-38G and irinotecan was detected in all samples. Study 2 only examined samples taken at 1 and 6 h. (B) Concentrations of products in NCI-N87 tumors from animals given irinotecan (Study 3). (C) Uptake and clearance of SN-38 [TOTAL] in mice bearing Capan-1 (Studies 1 and 2 combined) and NCI-N87 (Study 3). No other products were detected in the tumors. (D)

Comparison of SN-38 [TOTAL] concentrations in Capan-1 xenografts from mice in Study 1 and Study 2.

Figure 3. Contents of the upper small intestine analyzed for the various products in mice treated with irinotecan or IMMU-132 (Studies 1 and 2 combined, N = 3 to 6). Irinotecan animals were examined from 5 min to 24 h, while IMMU-132-treated animals were examined 1 to 72 h, but no products were found at detectable levels after 24 h.

Enhanced Delivery of SN-38 to Human Tumor Xenografts with an Anti-Trop-2-SN-38 Antibody Conjugate (Sacituzumab Govitecan)

Robert M Sharkey, William J McBride, Thomas M. Cardillo, et al.

Clin Cancer Res Published OnlineFirst June 23, 2015.

Updated version Access the most recent version of this article at: doi:10.1158/1078-0432.CCR-15-0670

Supplementary Access the most recent supplemental material at: Material http://clincancerres.aacrjournals.org/content/suppl/2015/06/24/1078-0432.CCR-15-0670.DC1

Author Author manuscripts have been peer reviewed and accepted for publication but have not yet been Manuscript edited.

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/early/2015/06/23/1078-0432.CCR-15-0670. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.