Faculty Bibliography

OBJECTIVE:Poly(ADP-ribosyl) polymerases (PARPs) are nuclear enzymes with roles in DNA damage recognition and repair. PARP1 inhibition enhances the effects of DNA-damaging agents like doxorubicin. We sought to determine the recommended phase two dose (RP2D) of veliparib with pegylated liposomal doxorubicin (PLD) in breast and recurrent gynecologic cancer patients. METHODS:on day 1 of a 28-day cycle. Dose escalation proceeded in two strata: A (prior PLD exposure) and B (no prior PLD exposure). Patients underwent limited pharmacokinetic (PK) sampling; an expansion PK cohort was added. RESULTS:44 patients with recurrent ovarian or triple negative breast cancer were enrolled. Median age 56 years; 23 patients BRCA mutation carriers; median prior regimens four. Patients received a median of four cycles of veliparib/PLD. Grade 3/4 toxicities were observed in 10% of patients. Antitumor activity was observed in both sporadic and BRCA-deficient cancers. Two BRCA mutation carriers had complete responses. Two BRCA patients developed oral squamous cell cancers after completing this regimen. PLD exposure was observed to be higher when veliparib doses were > 200 mg BID. CONCLUSIONS:PLD on day 1 of a 28-day cycle. Anti-tumor activity was seen in both strata. However, given development of long-term squamous cell cancers and the PK interaction observed, efforts should focus on other targeted combinations to improve efficacy.

Evidence suggests that stromal cells play critical roles in tumor growth. Uncovering new mechanisms that control stromal cell behavior and their accumulation within tumors may lead to development of more effective treatments. We provide evidence that the HU177 cryptic collagen epitope is selectively generated within human ovarian carcinomas and this collagen epitope plays a role in SKOV-3 ovarian tumor growth in vivo. The ability of the HU177 epitope to regulate SKOV-3 tumor growth depends in part on its ability to modulate stromal cell behavior because targeting this epitope inhibited angiogenesis and, surprisingly, the accumulation of alpha-smooth muscle actin-expressing stromal cells. Integrin alpha10beta1 can serve as a receptor for the HU177 epitope in alpha-smooth muscle actin-expressing stromal cells and subsequently regulates Erk-dependent migration. These findings are consistent with a mechanism by which the generation of the HU177 collagen epitope provides a previously unrecognized alpha10beta1 ligand that selectively governs angiogenesis and the accumulation of stromal cells, which in turn secrete protumorigenic factors that contribute to ovarian tumor growth. Our findings provide a new mechanistic understanding into the roles by which the HU177 epitope regulates ovarian tumor growth and provide new insight into the clinical results from a phase 1 human clinical study of the monoclonal antibody D93/TRC093 in patients with advanced malignant tumors.

OBJECTIVE: /st> To determine the feasibility and toxicity profile of topotecan administered as a seven-day continuous intrathecal infusion for patients with leptomeningeal metastasis secondary to recurrent or progressive central nervous system cancer. Study design Two patients with central nervous system leptomeningeal metastasis were treated with a seven-day continuous infusion of topotecan (0.2 mg/day) administered via continuous intrathecal/intraventricular infusion at a rate of 0.6 mL/h, totaling 1.4 mg/course. CSF and plasma concentrations of topotecan closed lactone (the active metabolite) were quantified at various points during topotecan infusion. Patients were monitored for neurologic and systemic toxicities according to NCI common toxicity criteria. RESULTS: /st> Both patients tolerated the seven-day continuous topotecan without any significant adverse events. One patient received a second course 21 days after treatment initiation. CSF concentration of topotecan closed lactone ranged from 3.73 to 312 ng/mL (median = 131 ng/mL) and plasma topotecan closed lactone ranged from 0.44 to 1.78 ng/mL (median = 0.92 ng/mL). The median CSF topotecan concentration was greater than the median serum topotecan concentration by a 44-fold magnitude when samples were obtained at the same time point. None of the patients experienced any grade 3 or higher hematological toxicities or signs of arachnoiditis. CONCLUSION: /st> A seven-day continuous intrathecal infusion of topotecan is well tolerated and has the potential of maximizing central nervous system drug exposure.

Nanometer-scale architectures assembled on cell surface receptors from smaller macromolecular constituents generated a large amplification of fluorescence. A targeted dendrimer was synthesized from a cystamine-core G4 PAMAM dendrimer, and contained an anti-BrE3 monoclonal antibody as the targeting group, several fluorophores and an average of 12 aldehyde moieties as complementary bio-orthogonal reactive sites for the covalent assembly. A cargo dendrimer, derived from a PAMAM G4 dendrimer, contained several fluorophores as the cargo for delivery and five hydrazine moieties as complimentary bio-orthogonal reactive sites. The system is designed to be flexible and allow for facile incorporation of a variety of targeting ligands.

BACKGROUND: Continuous-infusion topotecan with erlotinib has the potential to reverse topotecan resistance due to drug efflux mechanisms. We assessed the activity of such a regimen in ovarian cancer patients previously failing bolus topotecan. Assay for shed collagen epitopes recognized by antibody HU177 during treatment explored its ability to reflect tumor invasion. METHODS: Topotecan 0.4 mg/m(2) per day was administered by continuous infusion for 9-10 days every 3 weeks. Erlotinib, 150 mg orally, was administered on days 1-10 of each cycle. Cycles were repeated until progression or toxicity. Serum for shed HU177 collagen epitopes was collected weekly. This was a two-stage design to detect a CA-125 response rate of at least 20% in 30 patients after completing two treatment cycles. The trial would be terminated early if there were less than two CA-125 responses in 16 patients. Four or more CA-125 responses in 30 patients would justify further study of this regimen in prior topotecan treatment failures. RESULTS: Six patients were enrolled, with four receiving three or more cycles and one achieving a partial response by cancer antigen 125 (CA-125) criteria. Shed epitope levels became undetectable on at least one measurement in all patients who received three or more cycles (Fig. 1A) and reappeared concomitantly with rises in CA-125 and clinical progression (Fig. 1B). After logistical delays, the trial was closed by the sponsor's decision to stop developing erlotinib in ovarian cancer. FIGURE 1: Monitoring of combination treatment. A, B, C, D, and F refer to patients. (A):: Topotecan and erlotinib. (B):: CA-125 in units/mL. CONCLUSION: Continuous-infusion topotecan with erlotinib was found safe in six pretreated ovarian cancer patients; one met CA-125 criteria for partial response. Serial shed epitope levels to reflect invasiveness deserve further study.

PURPOSE: The potential synergy of modulating platinum-induced DNA damage by combining the proteasome inhibitor bortezomib with oxaliplatin was studied in patients with solid tumors, with special attention to avoidance of cumulative neurotoxicity (NT). PATIENTS AND METHODS: In a 3 + 3 dose escalation design, patients received bortezomib at 1.0-1.5 mg/m(2) on days 1 and 4 and oxaliplatin at 60-85 mg/m(2) on day 1 of a 14-day cycle. NT assessments were performed at the start of every two cycles. Oxaliplatin pharmacokinetics (PK) were determined pre- and post-bortezomib. RESULTS: Thirty patients were enrolled with 25 (11 men, 14 women) fully evaluable for NT assessments at cycle 2. The median age was 56 years (range 35-74 years); median number of cycles received 2 (range 1-10). At dose levels 2-5 (B 1.3 mg/m(2)), patients manifested NT grades 3 and 4 at a median 3.4 cycles (range 2-9 cycles): 3 had ataxia (one also with sensory neuropathy or neurogenic hypotension, respectively) and 3 had just sensory neuropathy. A 6th dose-level reducing bortezomib to 1.0 mg/m(2) with oxaliplatin 85 mg/m(2)) was explored and no NT or dose limiting toxicities were noted among 7 evaluable patients (5 receiving two or more cycles). Four patients experienced a partial response-one with platinum-resistant ovarian cancer, another with gastroesophageal cancer, another with ampulla of Vater carcinoma, and a patient with cholangiocarcinoma. PK studies at dose levels 1 and 2 showed greater mean ultrafiltrable platinum when oxaliplatin was dosed after bortezomib. CONCLUSIONS: Bortezomib 1.0 mg/m(2) x 2 every 14 days combines safely with oxaliplatin. At higher doses, cumulative NT (i.e., cerebellar signs and sensory neuropathy) occurs at an accelerated pace perhaps from a PK interaction.

Background: PARP1 inhibition enhances the effects of DNA-damaging agents such as doxorubicin. We sought to investigate PK of PLD in a phase I study of veliparib (ABT-888, V) and PLD in patients (pts) with recurrent ovarian, fallopian tube, and primary peritoneal, and triple negative breast cancers. No prior PK interactions have been described in V clinical trials. Methods: Complete blood samples on day (D) 1 (pre-PLD and 1 hr post PLD), D 8, D 22 on cycle 1 and 2 of treatment in pts receiving PLD 40 mg/m2, day (D) 1 and V D1-14 at varying dose levels of 50,100,150, 200, 300, 350 mg twice daily, were collected in 25 of 31 pts enrolled to a previously reported dose finding phase I study of V and PLD (SGO2012). Plasma PLD levels were measured by HPLC methodology detailed by Gabizon et al Cancer Chemo Pharmacol 2008, 61:695. PK parameter estimates were obtained using non-linear modeling programs available in Winnonlin Ver 5.3. Affect of V dose on PK parameters was estimated with linear regression analysis. Due to a higher degree of GI toxicity with V dosages > 200 mg, we utilized a cut-off of 200 mg for V. PK parameters in the group with dosages greater than or equal to 200 mg (high V, n=18) and those less than 200 mg (low V, n=7) were compared, utilizing an unpaired, 2 sided t-test. Results: PLD clearance (CL) was reduced, half-life (hL) was increased, and AUC/mg was increased with higher dosages of V when compared to historical published data. We noted a positive correlation of the auc/mg dose, p=0.001 and a negative correlation with the CL, p=0.001 and increasing V dose. When analyzed as low and high V groups, the mean +/- SEM hL (hrs) was significantly lower in the low V when compared to the high V group, 83.2 +/- 11.7 vs 108.6 +/- 6.0(p =0.042) , and the mean PLD clearance (ml/h)was greater in the low V versus the high V group 35.9 +/- 5.6 vs 14.2 + 1.0, P<.0001. Similarly the AUC/mg dose (mg x h/L) was significantly lower in the low vs high V groups, 35.0 +/-5.2 vs 74.9 +/- 4.4, P<0.0001. Con!

PURPOSE: The combination of oblimersen, a bcl-2 antisense oligonucleotide, and dacarbazine lead to superior progression-free survival in advanced melanoma patients. Albumin-bound paclitaxel (nab-paclitaxel) has single-agent activity in melanoma. METHODS: In a phase I trial, chemotherapy-naive patients with metastatic melanoma and normal LDH levels were enrolled on 3 cohorts. The treatment regimen consisted of 56-day cycles of oblimersen (7 mg/kg/day continuous IV infusion on day 1-7 and 22-28 in cohort 1 and 2; 900 mg fixed dose, twice weekly in weeks 1-2, 4-5 for cohort 3), temozolomide (75 mg/m(2), days 1-42), and nab-paclitaxel (175 mg/m(2) in cohort 1 and 3, 260 mg/m(2) in cohort 2 on day 7 and 28). Apoptosis markers were tested in pre- and post-treatment specimens of a subset of patients. RESULTS: Six grade 3 events (neutropenia, renal insufficiency, hyponatremia, elevated creatinine, allergic reaction, and neuropathy) and 2 grade 4 events (neutropenia and thrombocytopenia) were seen in 32 patients. The objective response rate was 40.6 % (2 complete responses and 11 partial responses) and 11 patients had stable disease, for a disease control rate of 75 %. CONCLUSIONS: The combination of oblimersen, temozolomide, and nab-paclitaxel was well tolerated and demonstrated encouraging activity in patients with advanced melanoma.

PURPOSE: Skin metastases of breast cancer remain a therapeutic challenge. Toll-like receptor 7 agonist imiquimod is an immune response modifier and can induce immune-mediated rejection of primary skin malignancies when topically applied. Here we tested the hypothesis that topical imiquimod stimulates local antitumor immunity and induces the regression of breast cancer skin metastases. EXPERIMENTAL DESIGN: A prospective clinical trial was designed to evaluate the local tumor response rate of breast cancer skin metastases treated with topical imiquimod, applied 5 d/wk for 8 weeks. Safety and immunologic correlates were secondary objectives. RESULTS: Ten patients were enrolled and completed the study. Imiquimod treatment was well tolerated, with only grade 1 to 2 transient local and systemic side effects consistent with imiquimod's immunomodulatory effects. Two patients achieved a partial response [20%; 95% confidence interval (CI), 3%-56%]. Responders showed histologic tumor regression with evidence of an immune-mediated response, showed by changes in the tumor lymphocytic infiltrate and locally produced cytokines. CONCLUSION: Topical imiquimod is a beneficial treatment modality for breast cancer metastatic to skin/chest wall and is well tolerated. Importantly, imiquimod can promote a proimmunogenic tumor microenvironment in breast cancer. Preclinical data generated by our group suggest superior results with a combination of imiquimod and ionizing radiation and we are currently testing in patients whether the combination can further improve antitumor immune and clinical responses.

Accumulating evidence indicates that the malignant behavior of tumors depends not only on tumor cells themselves, but also on the stromal cells that comprise the malignant tumor mass. Experimental findings suggest that stromal cells such as cancer-associated fibroblast (CAF) may play important roles in promoting tumor growth and metastasis as well as regulating the efficacy of certain chemotherapeutic drugs. However, developing novel clinical strategies that selectively and simultaneously impacts tumor and stromal cells remains challenging. Alterations in the integrity and molecular composition of the extracellular matrix (ECM) are hallmarks of tumor progression. Our previous studies have shown that structural remodeling of the ECM can result in localized triggering of what we have termed "biomechanical ECM switches" or changes in the three-dimensional structure of pre-existing ECM molecules. A humanized antibody (TRC093/D93) specifically directed to the HU177 cryptic collagen epitope that is selectively exposed following triggering of a biomechanical ECM switch has been developed, and a human phase-I clinical trial was recently completed with encouraging results. Here we provide evidence that the HU177 biomechanical ECM switch is triggered within human ovarian tumors resulting in the exposure of the HU177 cryptic collagen epitope. The relative exposure of the HU177 cryptic site was significantly (P<0.05) enhanced in biopsies of malignant ovarian tumors as compared to benign ovarian lesions. Selective targeting of the HU177 cryptic collagen epitope by Mab D93 significantly (P<0.05) inhibited SKOV-3 tumor growth by approximately 70% as compared to controls. Tumors from these mice exhibited reduced angiogenesis, elevated levels of apoptosis and a significant reduction in infiltration of alpha smooth muscle cell actin (alphaSMC-actin) positive stromal fibroblasts. Importantly, while Mab D93 inhibited SKOV-3 tumor cell adhesion to denatured collagen type-I and enhanced the expression of the cyclin !