The concept of manipulating the immune system to treat cancer has experienced waxing and waning levels of enthusiasm over decades of clinical investigation.
F. Stephen Hodi, MD
Director, Melanoma Center
Director, Center for Immuno-Oncology
Dana-Farber Cancer Institute
The concept of manipulating the immune system to treat cancer has experienced waxing and waning levels of enthusiasm over decades of clinical investigation. With improved understanding of how the immune system functions, harnessing the power of T-cell checkpoint antagonists has dramatically changed this prospect and shifted a paradigm of cancer care.
The primary interaction in educating the immune system occurs between the professional antigen-presenting cell (APC) (eg, dendritic cell or B cell) and the T cell. The APC functions to process antigen proteins into peptides, which are presented on its cell surface within the groove of the major histocompatibility complex (MHC). The peptide is presented to the T cell via interaction with the T-cell receptor (TCR). For T cells to gain potency against the antigenic peptide, B7 on the APC side and CD28 on the T-cell side must interact, resulting in costimulation. The T cell proliferates and makes cytokines such as interleukin-2 in its immunologic capacity. If this proceeded unchecked, it would result in nonspecific reactivity or autoimmunity. Therefore, upon activation of the T cell, cytotoxic T-lymphocyte antigen-4 (CTLA-4) is expressed on the T-cell surface, which has greater binding affinity to B7 than CD28, resulting in T-cell inhibition.
In a randomized phase III trial involving previously treated patients with melanoma including treated brain metastases, ipilimumab, a fully human CTLA-4 blocking antibody administered at 3 mg/kg for 4 doses, demonstrated an improvement in survival when combined with a gp100 peptide vaccine or as a single agent when compared with the vaccine alone.1 Importantly, the significant benefit was witnessed in the tail of the Kaplan-Meier curve, demonstrating durable survival in a subset of patients. This led to the FDA approval of ipilimumab for the treatment of metastatic melanoma. Ipilimumab benefit was confirmed in a subsequent phase III trial in patients with treatment-naïve melanoma. Ipilimumab (10 mg/kg x 4 doses, plus maintenance every 3 months) combined with dacarbazine chemotherapy (850 mg/m2) demonstrated an improvement in overall survival compared with dacarbazine alone.2 Activity for ipilimumab has also been reported in a number of other cancers, including renal cell carcinoma (RCC), non—small cell lung carcinoma (NSCLC), and prostate cancer.
The development of ipilimumab revealed a unique set of inflammatory adverse events (AEs). Dermatitis is the most common AE, seen in approximately 40% of patients. Fortunately, this is typically low grade and commonly does not need medical intervention. Diarrhea/ colitis can be seen in approximately 30% of patients (high-grade, about 8%). The concern with patients experiencing high-grade or prolonged diarrhea is development of bowel perforation, as deaths have been associated with treatment. Other potential inflammatory AEs include endocrinopathies (hypophysitis and thyroiditis), hepatitis, pancreatitis, nephritis, uveitis, and neuritis. Algorithms have been developed to assist in the management of these side effects, with particular attention to diarrhea and liver function toxicities.
T cells that are activated to seek and destroy cells expressing a specific MHC-bound peptide express on their surface the programmed death-1 (PD-1) receptor. When other cells, such as tumor cells, express the ligand (PD-L1), the PD-1/PD-L1 interaction leads to apoptosis of activated lymphocytes.3,4 As a result, the tumor cells—via expression of PD-L1—can ward off an immune attack by placing a cloak around themselves or stiff-arming the immune system. The expression of PD-L1 (also referred to as B7H1) is significant in a wide variety of tumors, including melanoma and RCC.5 For some cancers, such as RCC, expression of PD-L1 has been associated with poor prognosis. Currently, there are no fewer than 7 companies developing antagonistic antibodies to PD-1 or PD-L1.
A multidose phase I trial of nivolumab, a fully human IgG4 PD-1 blocking antibody, treated patients with melanoma, RCC, NSCLC, colorectal cancer (CRC), and castrate-resistant prostate cancer every 2 weeks for up to 96 weeks.6.7 High-grade toxicities (>grade 3) occurred in approximately 20% of patients and included lymphopenia, fatigue, diarrhea, abdominal pain, and elevated lipase. Pneumonitis was reported in a small subset of patients, with 3 drug-related deaths (2 NSCLC and 1 CRC). The results reported for nivolumab in NSCLC, melanoma, and RCC included clinical response rates of 17%, 31%, and 29%, respectively; stable disease rates (at least 24 weeks) of 10%, 7%, and 27%, respectively; and median progression-free survival (PFS) of 2.3 months, 3.7 months, and 7.3 months, respectively. Even among patients who discontinued therapy for toxicities or who completed the planned 96-week course of therapy, 19 of 27 patients maintained responses while off drug treatment.
MK-3475, another IgG4 high-affinity PD-1 agonist antibody, was administered in a phase I study at every 2-week and every 3-week schedules with a favorable and expected safety profile.8 Cohorts of patients with melanoma included both patients previously treated with ipilimumab and those naïve to ipilimumab. The overall response rate was 38% and did not apparently depend on prior ipilimumab therapy.
MPDL3280A is an IgG1 antibody against PD-L1 engineered specifically to remove antibody-dependent cell-mediated cytotoxicity (ADCC) function. MPDL3280A has demonstrated responses in NSCLC, melanoma, and RCC of 22%, 29%, and 13%, respectively; 24-week stable disease rates of 12%, 5%, and 32%, respectively; and 24-week PFS rates of 46%, 43%, and 53%, respectively.
Since the field is moving very quickly, combination studies are being developed. A phase I study combin- ing ipilimumab and nivolumab concurrently demonstrated a >50% response rate, with all responses experiencing >80% reduction in tumor volume.9 This impressive clinical activity, however, was associated with approximately half of patients experiencing highgrade (>grade 3) toxicities, including elevated liver function tests and pancreatic enzymes in a significant proportion of patients.
Phase III trials testing nivolumab and MK-3475 for a variety of cancers are under way. In addition, a phase III trial comparing ipilimumab versus nivolumab, versus a nivolumab/ipilimumab combination, has recently been completed.
These important studies, along with the development of additional immune checkpoint agents including antagonists as well as agonists, provide a rich future of clinical investigation and opportunities for patients with cancer.