You are here

Meeting Highlights

American Association for Cancer Research 2014

Walter Alexander

The American Association for Cancer Research (AACR) meeting is the premier forum for basic and translational oncology research. At this year’s 105th annual meeting, held from April 5 to 9 in San Diego, about 18,000 cancer researchers, other cancer-care professionals, and patient advocates from around the world convened for the presentation of more than 6,000 original papers. Unlike most major medical meetings, the AACR conference is devoted to early phase research. This report includes highlights on investigational agents (one for breast cancer and two for melanoma) selected by the AACR program committee for inclusion in the official press briefings based on the quality of the research and their potential clinical importance.

Final Results of a Randomized Phase 2 Study of Palbociclib (PD 0332991), a Cyclin-Dependent Kinase (CDK) 4/6 Inhibitor, in Combination With Letrozole vs. Letrozole Alone for First-Line Treatment of ER+, HER2− Advanced Breast Cancer (PALMOA-1/TRIO-18)

  • Richard Finn, MD, Associate Professor of Medicine, University of California at Los Angeles

While breast cancer is known to be very diverse at the molecular level, clinically it falls into only three groups, Dr. Finn said. The estrogen receptor–positive (ER+) group, which represents about 60% of breast cancers, has anti-estrogens as the backbone of therapy. The human epidermal growth factor receptor 2 (HER2) amplified group, about 25% of breast cancers, is treated with anti-HER2 agents such as trastuzumab (herceptin), and the 15% of patients who are ER, progesterone (PR), and HER2 negative represent an unmet need.

Palbociclib is a first-in-class oral, selective cyclin-dependent kinase (CDK) 4/6 inhibitor that inhibits cell proliferation and cellular DNA synthesis by preventing cell-cycle progression from G1 to S phase. Early research showed palbociclib to have in vitro activity in retinoblastoma-positive tumor cell lines and primary tumors.

Low nanomolar concentrations of palbociclib block retinoblastoma phosphorylation, inducing G1 arrest in sensitive cell lines. The retinoblastoma protein (Rb) is a tumor suppressor protein that is dysfunctional in some cancers. CDK 4/6 is dependent on another protein, cyclin B. Together, CDK 4/6 and cyclin B play a role in adding phosphates, hyperphosphorylating the Rb gene product. Rb is critical for regulating the transition from G1 to S phases of the cell cycle. Once Rb is hyperphosphorylated, it releases the brake from the G1/S transition. “So the idea is,” Dr. Finn said, “that by blocking this kinase, you block the hyperphosphorylaton of Rb, and therefore block cell-cycle progression.”

In preclinical research, palbociclib preferentially inhibited proliferation of luminal ER+ human breast cancer cell lines. When palbociclib was combined with anti-estrogens, the cell growth-blocking effect was synergistic. In the phase 2 PALOMA-1/TRIO-18 study, investigators divided the patient population into two parts, the first including 66 postmenopausal women with ER+, HER2-negative breast cancer not previously treated for advanced disease (locally recurrent or metastatic), and the second meeting the same criteria but also having CCND1 amplification and/or loss of p16 (n = 99). CCND1 encodes for the cyclin D1 protein involved in G1/S cell-cycle transition.

Women in both parts were randomized 1:1 to palbociclib (125 mg daily) plus letrozole (2.5 mg daily) or to letrozole (2.5 mg) alone in a three-weeks-on, one-week-off schedule. The second part, looking at genomic changes in cyclin B1 or loss of p16, was testing whether or not biomarkers could be further refined.

Median age was 63 years in the palbociclib plus letrozole group and 64 years in the letrozole-alone group. Previously, in the adjuvant setting, about a third of patients had had prior hormonal therapy and approximately 43% had had prior chemotherapy.

Median progression-free survival (PFS), the primary endpoint, was doubled in the 84-patient palbociclib plus letrozole group (20.2 months vs. 10.2 months, hazard ratio [HR], 0.488; 95% confidence interval [CI], 0.319–0.748; P = 0.0004) as compared to the 81 women taking letrozole alone.

Analysis of the part 1 and part 2 findings revealed similar dramatic improvements in the palbociclib plus letrozole arms, regardless of biomarkers. The control arms differed, however, with a longer PFS in the part 2 control arm (11.1 months vs. 5.7 months). “That might reflect how cyclin B1 or p16 selection affects responses to letrozole,” Dr. Finn speculated.

Secondary endpoint analysis revealed consistent benefits for the combination in objective response rate (43% vs. 33%) and in clinical benefit (complete response plus partial response and stable disease for at least 24 weeks, 81% vs. 58%).

Overall survival (OS) was not significantly longer for the combination (37.5 months vs. 33.3 months for letrozole; HR, 0.813; 95% CI, 0.492–1.345; P = 0.2105). However, Dr. Finn added, “Remember that survival events in ER-positive breast cancer take time to evolve. There’s only 30 events per arm at this point, so we are encouraged by the median overall survival increase.”

Also encouraging, Dr. Finn said, was the lack of new toxic signals, with neutropenia, leukopenia, and fatigue being most common in the combination arm. “This neutropenia is self-limiting; patients recover quickly, and most importantly, we do not see neutropenic fever.”

Dr. Finn noted that a randomized phase 3 study (PALOMA-2/TRIO-22) is ongoing in a similar patient population.

Antitumor Activity of the Anti-PD-1 Monoclonal Antibody MK-3475 in Melanoma: Correlation of Tumor PD-L1 Expression With Outcome

  • Adil I. Daud, MD, Co-Director of the University of California at San Francisco Melanoma Center

MK-3475, a humanized, monoclonal IgG4 antibody, is a very potent blocker of PD ligands 1 and 2, both of which play some role in triggering the PD-1 pathway. “It is increasingly recognized that the immune system plays a major role in controlling cancer—and one of the ways that cancer cells evade the immune system,” Dr. Daud explained, “is by using the PD-1/PD-L1 pathway.” Cancer cells, he continued, co-opt this pathway and, in some models, they block T-lymphocyte action by expressing PD ligand. “One way to defeat this is by using a PD-1 antibody such as MK-3475.”

This phase 1B KEYNOTE-001 study analysis of advanced melanoma patients treated with MK-3475 aimed to test the relationship of tumor PD-L1 expression to outcomes and to assess the effects of MK-3475 on peripheral T-cell phenotype as a pharmacodynamics marker. In the complex trial, the tested population included five groups subdivided according to prior ipilimumab treatment (naïve, treated, or refractory), dosage (10 mg or 2 mg), and schedule (every two weeks or every three weeks). The primary response assessment was conducted according to rigorous RECIST (Response Evaluation Criteria in Solid Tumors) 1.1 criteria per independent central review (with a first assessment at 12 weeks). Within 60 days prior to the initiation of treatment, patients had a mandatory biopsy by a clinical trial immunohistochemical assay. Eighty-three patients were identified as having measurable PD-L1–positive tumors and 30 as having measurable PD-L1–negative tumors. PD-L1 positivity was defined as staining of at least 1% of tumor cells.

Among the 125 patients (mean age 63 years, 61% male) with evaluable PD-L1 expression, 89 (71%) were found to be PD-L1– positive, typical of melanoma trials, Dr. Daud said. He also noted that 76% of patients were BRAF wild type, and 57% had advanced disease (M1c) with visceral metastases. All patients had advanced, unresectable melanoma, measurable disease (per investigator assessment), and were Eastern Cooperative Oncology Group (ECOG) performance status 0 or 1.

Reductions in tumor size using MK-3475 were reported in 67% of patients. Tumor growth was more common among PD-L1–negative patients.

In the overall unselected population, the overall response rate (ORR), based on the optimally defined cut-point of at least one tumor cell with the protein present per 100 tumor cells stained, was 40% (95% CI, 31–50). “That’s high for immunotherapy,” Dr. Daud commented. For PD-L1–positive patients the ORR was 49% (95% CI, 40–65), as compared with 13% for PD-L1–negative patients (95% CI, 4–31; P = 0.0007).

In the unselected population, the disease control rate (complete response plus partial response and stable disease) was 58%. It was 65% for the PD-L1–positive population and 37% for PD-L1–negative patients (PD-L1–positive vs. PD-L1–negative, P = 0.0070).

PFS after 12 months was 45% in the PD-L1–positive group, significantly higher than the 18% found in the PD-L1–negative group (HR, 0.52; 95% CI, 0.32–0.86; P = 0.0051).

Six-month OS estimates, however, were similar between groups, at 91% for PD-L1–positive patients and 79% for PD-L1–negative patients (HR, 0.83; 95% CI, 0.39–1.78; P = 0.3165). Median OS was not reached for either group. While Dr. Daud attributed the lack of significant difference to the high activity of the drug and to the immaturity of the data with few survival events, he commented, “It’s still interesting that there’s no separation in the survival curves.”

An analysis of changes from baseline in the percentage of CD8-positive (“killer”) T cells and in CD4-positive (“helper”) T cells showed significant median increases in both (14.6%; P < 0.001, and 15.7%; P < 0.001, respectively), but found no correlation between this T cell activation and response. Immune responses seemed to be improved similarly at all doses of MK-3475.

In patients whose tumors were PD-L1–negative, prior ipilimumab treatment did not affect outcomes.

“This is an extraordinarily active antibody in melanoma, with a durable objective response rate,” Dr. Daud said. He noted further that while fewer PD-L1–negative patients had responses, “their quality of responses was just as good.”

Studies with MK-3475, both as monotherapy and in combinations, are ongoing in patients with multiple solid tumors and hematological malignancies. They may provide more insight into the clinical utility of PD-L1 expression in patients treated with MK-3475, Dr. Daud said.

This study was funded by Merck. Dr. Daud has served on the advisory boards of Merck and GlaxoSmithKline.

A Phase I Study of IMCgp100: Durable Responses With a Novel First-in-Class Immunotherapy for Advanced Melanoma

  • Mark Middleton, MD, PhD, Professor of Experimental Cancer Medicine, University of Oxford, United Kingdom

Press briefing moderator Louis M. Weiner, MD, director of the Georgetown Lombardi Cancer Center, noted in introducing Dr. Middleton’s presentation: “Even with the best of our checkpoint inhibitor strategies to date, there still remains a reservoir of people whose cancers will not be responsive—because of the ways the cancers are defending themselves against the body’s immune system.”

IMCgp100, a first-in-class immunotherapy, is a bispecific biologic incorporating an engineered T-cell receptor with specificity for a peptide antigen derived from protein gp100, which is presented by HLA-A2. Expression of HLA-A2 antigen is important for tumor cell recognition by autologous T-lymphocytes. Expression of some HLA antigens may be selectively lost in several human tumors, including melanoma.

IMCgp100 binds tightly to tumor cells, activating adjacent killer T cells and stimulating a cascade of immune-activating molecules that recruit a further immune response. With the very tight binding of IMCgp100 to the cancer cell, T-cell receptor affinity is increased by 3,500,000. Cytotoxic T cells are recruited into proximity with the melanoma cells bearing HLA-A2 through binding to CD3 on the T cell. Then cytotoxic T cells are able to target the melanoma cells, leading to apoptosis.

HLA-A2, Dr. Middleton noted, is expressed in about 45% of melanoma patients. Existing immunotherapies target cell surface proteins, making only about 10% of potential targets within cancer recognizable to T cells. “By targeting HLA-presenting peptides, we bring many more targets (approximately 90%) into play. The results of this study offer proof of this concept.”

Thirty-one HLA-A2 patients (median age 61 years, 58% male) with stage IV or unresectable stage III melanoma with ECOG performance status of 1 or 0 and lymphocyte counts of at least 0.5 x 109/L were included in the study. Sixty percent had received systemic treatment (DTIC, 29%; ipilimumab, 6%; vemurafenib, 3%; or one or more experimental therapies, 22%) prior to the study.

IMCgp100 doses were escalated from 5 to 900 ng/kg until occurrence of dose-limiting toxicity. Rash was observed initially at 45 ng/kg. After two of four patients receiving IMCgp100 at 900 ng/kg developed grade 3 hypotension with edema, rash, and fever, the maximum tolerated dose (MTD) was defined at 600 ng/kg. Tumor flare was observed in patients with cutaneous or subcutaneous disease. “We generally see tumor flare as a good sign in immunotherapy. It’s a suggestion of an immune infiltrate, and it often precedes tumor shrinkage,” Dr. Middleton said.

Partial responses among the 16 patients receiving more than 135 ng/kg of IMCgp100 were reported in four, with three meeting RECIST criteria. “We see this as being extremely significant. It points to the potential of this drug for future therapy,” he said. Responses have lasted about a year in the first two patients, with further tumor shrinkage in one patient who remains asymptomatic.

“With IMCgp100, we have a first-in-class drug, a monoclonal T-cell receptor immunomodulator … that can be administered safely at biologically effective doses. It is associated with durable clinical responses in melanoma,” Dr. Middleton said.

For the ongoing study, a protocol amendment allowing daily dosing has been approved. “We feel that for pharmacokinetic reasons this may be more efficacious than once-weekly dosing,” Dr. Middleton said.

Dr. Weiner added, “This is an approach that essentially fills that gap in therapy … and I think that it, along with similar approaches, offers significant promise.”