More than 100 patients with thoracic malignancies have been treated on a series of clinical protocols examining toxicities and clinical responses following infusions of DNA demethylating agents (Decitabine; DAC), and HDAC inhibitors, such as romidepsin (DP) alone or in combination with other investigational agents. Collectively these trials demonstrated no objective clinical regressions, although prolonged stabilization of disease (4 - 12months) was observed in approximately 10% of patients. Nearly one quarter of all patients receiving DAC infusions exhibited increased expression of p16, MAGE-3, or NY-ESO-1 in tumor tissues. Serologic responses to NY-ESO-1 were observed in several patients receiving DAC for more than six months. Approximately 50% of patients receiving DP infusions exhibited increased intratumoral levels of H3Ac and p21. In addition, several patients exhibited enhanced expression of NY-ESO-1 and MAGE-A3 in tumor biopsies following DP infusions (120). Micro-array analysis of laser captured tumor cells from pre and post treatment biopsies from patients receiving DAC, DP or sequential DAC/DP infusions revealed a shift from a lung cancer gene signature to one observed in normal respiratory epithelia. These early trials provided proof of concept for the use of epigenetic regimens in combination with immunologic interventions for the treatment of thoracic malignancies. Because CT-X antigens appear to be preferentially expressed in pluripotent tumor cells, it is conceivable that autologous epigenetically modified tumor cells may be unique, personalized vaccines to induce immune responses to cancer stem cells. Difficulties regarding reliable establishment of primary cell lines limited our ability to formally evaluate this issue in the clinic, and these protocols were closed. In an attempt to circumvent these problems, we examined if vaccines produced from K562 erythroleukemia and H1299 lung cancer cells which exhibit high level CT-X gene expression without pharmacologic manipulation could induce broad immunity to CT-X antigens that potentially can be up-regulated in thoracic malignancies by gene induction regimens. In a phase II trial, 20 patients received six, monthly immunizations with live irradiated K562 cells constitutively expressing GM-CSF (K562-GM) in conjunction with metronomic oral cyclophosphamide and celecoxib (to inhibit immunosuppressive Tregs) as adjuvant therapy following complete resection of their malignancies. Metronomic chemotherapy was well tolerated, and no vaccine-related toxicities were observed. Whereas several patients exhibited increased serologic reactivity to purified CT-X antigens following their vaccinations, none met criteria for immunologic response. Several patients had uncharacteristically long disease free intervals while on therapy, and then developed rapid recurrence shortly after being removed from study. These findings raise the possibility that the vaccine induced transient antitumor immunity that was not detected by analysis of serologic responses to purified CT-X antigens; alternatively, the patients may have derived benefit from the metronomic chemotherapy. A manuscript pertaining to responses to carbohydrate antigens is under peer review, and we are completing additional analysis of immune responses to the vaccine and autologous tumor for another manuscript. In more recent efforts, we have evaluated the immunogenicity of a freeze thaw lysate of H1299. This NSCLC line was established at the NCI-NCI and exhibits broader and much higher levels of CT-X gene expression relative to K562-GM due to amplification of the X chromosome. In an ongoing trial, 20 thoracic oncology patients rendered NED by conventional therapy, yet at high risk for recurrence have been randomized to receive H1299 lysates with Iscomatix adjuvant with or without metronomic cyclophosphamide/celecoxib. Vaccines have been administered monthly for 6 months followed by treatment evaluation one month later. A variety of biologic and immunologic endpoints are being assessed in this First-in-Humans vaccine trial which has been on FDA Clinical Hold since May, 2015 due to problems in the Clinical Center PDS, which were not directly related to Iscomatrix, but nevertheless have prevented us from using the current lot of this vialed product. There were no vaccine related adverse events. Blinded analysis demonstrated sero-conversion or significant increases in existing titers to several CT-X and endogenous retroviral antigens in 8 of 14 patients (57%) evaluable for treatment response. A major concern regarding epigenetic therapy for solid cancers pertains to tumor heterogeneity and reliability of FNA specimens to reflect target gene modulation in cancer cells, as well as effects of drug treatment on the tumor microenvironment. This latter issue is particularly relevant given recent observations that DNA demethylating agents and HDAC inhibitors enhance responses to immune checkpoint inhibitors by inhibiting MDSC within the tumor microenvironment (127). Recently we speculated that patients undergoing bilateral pulmonary metastasectomy procedures might be ideal candidates for evaluation of novel epigenetic regimens provided their metastases are clonal. Surprisingly little information was available regarding this issue. As such, RNA and DNA were extracted from 65 trisected pulmonary metastases resected from 12 patients. qRT-PCR techniques were used to evaluate expression levels of a broad panel of CT-X and T-S genes as well as genes encoding epigenetic modifiers aberrantly expressed in human malignancies. Pyrosequencing techniques were used to quantitate DNA methylation levels in LINE1, NBL2, and D4Z4 repetitive sequences and promoter methylation status of differentially regulated genes. This analysis demonstrated that pulmonary metastases exhibit histology-related and patient-specific global DNA demethylation profiles. Significant inter-patient heterogeneity of gene expression was observed even among individuals with similar tumor histologies. Epigenetic signatures appeared consistent among metastases from the same patient, irrespective of time of resection (synchronous/metachronous) or anatomic location. These studies provided the rationale for a recently approved phase I dose escalation study of oral DAC/THU administered as adjuvant therapy for patients undergoing standard of care bilateral pulmonary metastasectomy procedures. Patients will undergo initial hemithoracic metastasectomy, and thereafter receive six weeks of oral DAC/THU therapy followed by contralateral metastasectomy if no contraindications. Comprehensive state of the art translational analyses of PD endpoints including multiplex quantitative IHC, multiparametric mass cytometry (CyTOF) and next-gen sequencing will be conducted in the Translational Imuno-Oncology Laboratory at Yale University. Additional studies including analysis of immune subsets, circulating tumor cells and methylation status of circulating tumor DNA will be performed at the NIH Clinical Center. Additionally, a phase II trial evaluating Nivolumab alone or in combination with oral DAC/THU as second line therapy for NSCLC will open in the near future. This protocol, which will be conducted in collaboration with Drs. Yogen Saunthararajah and Vamsidhar Velcheti at the Cleveland Clinic, and incorporate comprehensive PD analyses described above, are part of a UO-1 grant (D. Schrump, intramural PI; Yogen Saunthararajah, extramural PI) presently under review.
Showing the most recent 10 out of 12 publications
