A Novel Antibody Drug Conjugate (ADC) Targeting Cell Surface Heat Shock Protein 70 (csHSP70) Is Active Against Pre-Clinical Models of Peripheral T-Cell Lymphoma (PTCL)

Background:

Neoplasms of T-cell or natural killer/T-cell origin account for 10-15% of all non-Hodgkin lymphomas (NHLs) in the United States, and 30% or more of NHLs in African and Asian countries, and tumors from post-thymic or peripheral T-cells are referred to collectively as PTCLs. Recent advances, including approval of brentuximab vedotin (BV), an anti-CD30 monoclonal antibody (mAb) drug conjugate (ADC) with monomethyl auristatin E (MMAE), deacetylase inhibitors (HDACis), and Anaplastic lymphoma kinase (ALK) inhibitors for ALK-positive anaplastic large cell lymphoma (ALCL) have improved outcomes. However, most PTCLs still have a poorer prognosis than comparable B-cell NHLs, and identification of novel targets and drugs retains importance in this area of unmet medical need.

Methods:

Pre-clinical studies were performed using PTCL and cutaneous T-cell lymphoma (CTCL) cell lines initially in vitro, and then using an in vivo xenograft model. Publically available databases were also leveraged, including the Broad Institute Cancer Cell Line Encyclopedia (CCLE), as well as our own RNA-sequencing (RNA-Seq) data from primary PTCL samples.

Results:

We examined the cell surface proteome of SUD-HL-1 (ALK+ ALCL), Mac-1 (ALK- ALCL), HH (CTCL), and HuT 78 (Mycosis fungoides with Sézary syndrome) cells by biotinylation and then mass spectrometry, and identified csHSP70 as being consistently expressed in all four lines. Analysis of the CCLE showed that HSP70 mRNA and HSP70 protein was expressed at the highest level in T-cell lymphoma cell lines, and our own RNA-Seq data confirmed HSP70 gene expression was higher in primary PTCL samples, and especially in ALCLs, compared with normal T-cells. To test its promise as a therapeutic target, we generated mAbs to human HSP70 and identified one clone, 239-87, which specifically bound csHSP70 on T-cell NHL cell lines but not on normal peripheral blood-derived mononuclear cells (PBMCs). Next, 239-87 was linked to MMAE to generate an ADC with a drug:antibody ratio of 4, and we confirmed that it was both internalized and then trafficked into acidic vacuoles in SUD-HL-1 cells. The 239-87-MMAE ADC induced a time- and concentration-dependent loss of viability in a panel of PTCL and CTCL cell lines associated with a G2/M arrest and induction of apoptosis, while normal PBMCs were unaffected. Comparisons of the activity of BV with 239-87-MMAE showed that the latter had similar efficacy against SU-DHL-1 and Hut 78 cells in vitro. When cells were propagated under conditions of hypoxia to mimic the tumor microenvironment there was an increase in csHSP70 expression, and the sensitivity of PTCL and CTCL cell lines to the 239-87-MMAE ADC was enhanced. Conversely, an inducible HSP70-targeted short hairpin RNA reduced total and csHSP70 protein expression, and reduced the efficacy of the ADC. Also of note, the HDACi vorinostat enhanced csHSP70 levels, and combinations of vorinostat with the 239-87-MMAE ADC enhanced loss of viability in these cells in a synergistic manner based on combination index analyses. Finally, we prepared an orthotopic in vivo PTCL model by subcutaneously injecting luciferase-labeled Mac-1 cells into C.B-17/IcrHsd-Prkdc ^scid mice. Disease progression occurred rapidly in all mice treated once weekly on days 10, 17, 24, and 31 with an IgG2A isotype mAb, as was the case for 7/8 mice treated with the 239-87-MMAE ADC at 1 mg/kg. In contrast, palpable tumor disappeared in 1/8 mice that received this ADC at 1 mg/kg, and 8/8 and 7/7 mice that received dosing at 5 and 10 mg/kg, respectively (Figure 1A). Tumor recurrence has not been seen at 105 days, including 74 days since the last ADC dose, and the one mouse at 1 mg/kg, and 3 each in the 5 and 10 mg/kg cohorts have had no disease by imaging, while the others have a small residual signal (Figure 1B) that has not progressed for two months.

Conclusions:

These pre-clinical in vitro and in vivo data support the possibility that csHSP70 could represent a novel therapeutic target for PTCL, and provide a rationale to translate ADCs based on our clone 239-87 mAb to the clinic for patients with advanced ALCL, and potentially other T-cell lymphomas as well.

Figure 1

View largeDownload slide

Figure 1

View largeDownload slide

Close modal