BETHESDA, MD USA (UroToday.com) - Dr. David McDermott described efforts at Beth Israel Deaconess, and nationally, in targeting immune checkpoint inhibitors. The interest in immunotherapy in checkpoint blockade has been tremendous since last summer when the first PD1 antibodies were shown to have significant antineoplastic activity. PD1 inhibitors act by blocking the interaction between PD1 on T cells and PDL1 on tumor cells. This interaction promotes immune tolerance to tumor antigens.
The first cohort of RCC patients (n=34) had 29% objective responses, with several patients able to stop therapy, with sustained durability of response. Initial excitement is somewhat tempered by follow up studies that showed lesser response rates and progression-free survival. However, there were still sustained responses in these follow-up studies, which was encouraging. Dr. McDermott suggested that patient selection might be key to identifying patients who will respond to such treatments. He showed that PDL1 expression on tumors is associated with clinical benefit. However, there are still PDL1-negative patients who respond to PD1 antibodies, thus patients with PDL1-negative tumors should not be excluded from these treatments (or better biomarkers need to be identified).
Tumor heterogeneity may complicate biomarker development as there is discordance in PDL1 expression in different parts of large tumors, or between primary and metastatic sites. His institution has developed immune correlates of response to MPDL3280A (an anti-PDL1 antibody). Interestingly, patients from this trial with high disease burden, Fuhrman grade 4, or sarcomatoid histologies seemed to respond the best to checkpoint blockade. In retrospectively analyzing the initial nivolumab studies , he showed that patients with grade 3/4 disease also had the higher response rates compared to lower grades. Additionally, he referred to a global TCGA study showing that tumors with higher numbers of mutations due to carcinogenic exposure (melanoma, lung cancer, bladder cancer) also tend to have high responses. Indeed, smokers with lung cancer have 2-3 times higher likelihood of MPDL response. He suggested that a similar mechanism may be at play in RCC patients (although they have relatively less frequent mutations on a per genome basis). He suggested that inflamed tumors (tumors with infiltrating lymphocytes and active immune milieu) should be getting PD1 orPDL1 inhibitors. However, overcoming innate or acquired resistance will be important for non-responders. In staining tumors for CD8+ T, he showed that TIM-3 expression on these tumors may be responsible another mechanism of immune blockade. Higher numbers of T-effector or T-helper genes expressed in the tumor were associated with response but not other immune gene sets.
The last question he set out to answer was if the response duration can be extended by combining checkpoint inhibitors. In a pre-clinical model, PDL1 and VEGF blockade was effective, and a phase 1 trial combining MPDL3082A and bevacizumab echoed these findings. A randomized phase 3 trial is now underway to compare sunitinib, PDL1 blocckade, and PDL1 blockcade + bevacizumab. He noted that these combinations increase toxicity, which can be severe. He added that non-inflamed tumors might need vaccine-based therapies by highlighting preclinical mouse work. In addition to DC-based vaccines, neo-antigens can be created using tumor-specific mutant peptides or proteins for vaccination strategies. These agents have generated much excitement among the medical and surgical urological communities.
Presented by:
David F. McDermott, MD
Beth Israel Deaconess Medical Center, Boston, MA USA
Reported by:
Philip H. Abbosh, MD, PhD* from the 2014 Winter Meeting of the Society of Urologic Oncology (SUO) "Defining Excellence in Urologic Oncology" - December 3 - 5, 2014 - Bethesda, MD USA
*Fox Chase Cancer Center, Philadelphia, PA USA