Skip to main content
Premium Trial:

Request an Annual Quote

Personalized Rx Highlights from American Association for Cancer Research Annual Meeting 2014


Below are the highlights from personalized medicine-focused studies presented this year at the American Association for Cancer Research annual meeting in San Diego.

Phase I Study of Novartis' BGJ398 in Cancer Patients with FGFR Genetic Alterations

In a Phase I study, Novartis' BGJ398 showed promising activity in cancer patients whose disease was being driven by FGFR genetic alterations. The first-in-human study aimed to gauge how well patients tolerated BGJ398. Researchers also enrolled patients with FGFR tumor alterations based on the hypothesis that this subgroup would be best responders to the treatment.

The study, funded by Novartis and conducted by Lecia Sequist of Massachusetts General Hospital and others, enrolled 107 patients with FGFR genetic alterations in their tumors. The majority of the participants had squamous cell lung cancer or breast cancer, but researchers also enrolled those with other types of cancer, such as cholangiocarcinoma and urothelial/bladder cancer.

In the dose-escalation phase of the study, 43 patients participated. Then, in a second expansion phase, there were three groups – 18 patients with FGFR1-amplified squamous cell lung cancer who received BGJ398 daily; 21 patients with other cancers who received the drug every day for four weeks; and 25 patients with other cancers who received the drug daily for three weeks and then had a week without treatment.

Patients who received 100 mg of the drug in the dose escalation phase experienced tumor shrinkage. Also four out of five patients with FGFR3-mutated urothelial cancer experienced tumor shrinkage, while two patients had tumor shrinkage of more than 30 percent on BGJ398, defined as a partial response. One patient with FGFR1-amplified squamous cell lung cancer had a partial response, and anti-cancer activity was also noted in those with squamous cell head and neck cancer, breast cancer, and cholangiocarcinoma.

The most common side effect seen in patients was heightened levels of phosphorus and researchers noted that in future studies of FGFR inhibitors, patients will need to take phosphate lowering drugs to manage this adverse event.

Dacomitinib Shows Promise in Molecularly Defined Head and Neck Cancer Patients

Researchers from Seoul presented data from a Phase II study that suggested that Pfizer's dacomitinib may benefit metastatic squamous cell head and neck cancer patients who have no defects in the PI3K pathway or excessive inflammation.

Earlier this year, Pfizer reported that two randomized Phase III studies of the irreversible pan-HER inhibitor dacomitinib – ARCHER 1009 and BR.26 – failed to meet their primary endpoints in advanced NSCLC patients. Following these failures, Pfizer noted that it is conducting molecular subgroups analysis in these studies, and also highlighted that in another trial, ARCHER 1050, researchers would enroll patients with EGFR-mutant NSCLC, randomizing them to dacomitinib or AstraZeneca's Iressa (gefitinib).

Meanwhile, the data reported at AACR by Byoung Chul Cho of the Yonsei Cancer Center in Seoul, provides insights into the activity of dacomitinib in another cancer group with very poor prognosis. The researchers hypothesized that because dacomitinib is an EGFR inhibitor, it will work well in squamous cell head and neck cancer patients because they have elevated levels of the protein.

In the study, researchers enrolled 48 patients with recurrent or metastatic squamous cell head and neck cancer and gave them the drug. Of these patients, 31 had stable disease. At median follow up of 8.4 months, the average time-to-progression was 3.4 months and the average overall survival was 6.6 months.

Then researchers performed genetic analysis and identified several markers associated with dacomitinib response. Patients with tumor mutations in PI3K pathway genes, PIK3CA and PTEN, experienced disease progression while on the drug twice as quickly as patients without these mutations. In patients with mutations, average progression-free survival was 2.9 months versus 4.9 months in the non-mutated group. In particular, two patients without these mutations had time-to-disease progression of 13.1 months and 18.9 months.

The researchers also reported differences in average progression-free survival between patients with tumors with high and low gene expression levels in IL6, IL8, PTGS2, and PLA2G2A, linked to inflammation. Cho and colleagues said their findings need to be confirmed in a larger Phase III trial, and noted they are conducting additional analysis to identify best responder populations for dacomitinib.

Researchers Describe Resistance to Xtandi in Castration Resistant Prostate Cancer Patients with AR-V7 in CTCs

A study conducted by Emmanuel Antonarakis of Johns Hopkins Sidney Kimmel Comprehensive Cancer Center and colleagues found that a molecule researchers detected in circulating tumor cells of metastatic, castration resistant prostate cancer patients may predispose them to having a poor response to the drug Xtandi (enzalutamide)

Xtandi is developed by Medivation for the treatment of metastatic castration-resistant prostate cancer, and has been shown to decrease prostate specific antigen levels by 89 percent in patients taking the drug for a month. The US Food and Drug Administration approved the drug for this indication in 2012.

In the study funded by the Prostate Cancer Foundation, Antonarakis and colleagues enrolled 31 patients with metastatic, castration-resistant prostate cancer who were prescribed Xtandi and collected blood samples from them before they started therapy, when they had maximum response, and at the time of disease progression. Researchers analyzed these samples for circulating tumor cells for the presence of AR-V7 mRNA.

As the researchers described it, hormones, such as testosterone, drive prostate cancer by binding to the androgen receptor. This receptor then attaches to other genetic material in prostate cancer cells and drives the disease. The AR-V7 is a truncated form of the androgen receptor in that it's missing a portion that testosterone can bind to, which is also the part that Xtandi binds to. However, even though testosterone can't bind to AR-V7, the protein can bind to other genetic material and fuel cancer growth, but this feature still makes the prostate cancer patient resistant to Xtandi treatment. AR-V7 mRNA, which Antonarakis and colleagues measured in circulating tumor cells, "is an intermediate between the AR gene and the AR-V7 protein," AACR said in a statement announcing the results.

Twelve of the 31 patients who had AR-V7 mRNA detectable in blood samples before starting Xtandi had worse response to the drug compared to those with no detectable AR-V7 mRNA in circulating tumor cells. PSA levels didn't drop in those with detectable AR-V7, but dropped by 50 percent or more in 10 of the 19 patients without detectable AR-V7. Additionally, those positive for the marker experienced disease progression earlier than those negative for the marker ─ 2.1 months time-to-radiographic progression versus 6.1 months, respectively.

This is the first study to link AR-V7 to Xtandi resistance in metastatic, castration resistant prostate cancer patients. Antonarakis said in a statement that this finding needs to be confirmed in other studies. Furthermore, his group is developing a lab test that can detect AR-V7 in circulating tumor cells and is performed in a CLIA certified lab. Antonarakis and colleagues are also conducting a small study to see if AR-V7 can similarly predict resistance to Zytiga (abirateron), another drug for metastatic, castration resistant prostate cancer, and hope to report results later this year.

NSCLC, Melanoma Patients with PD-L1-Positive Tumors Benefit from Merck Immunotherapy MK-3475 in Early Studies

Non-small cell lung cancer patients with high levels of the PD-L1 protein in their tumors experienced better outcomes following treatment with Merck's immune checkpoint inhibitor MK-3475 than those without this protein, Leena Gandhi of Harvard Medical School and colleagues reported at AACR.

Preliminary data from a Phase I study involving more than 100 previously treated NSCLC patients showed that six months after treatment 41 percent of those with high levels of PD-L1 had no disease progression compared to 17 percent of in the low PD-L1 group. While the study is ongoing, 72 percent of high PD-L1 patients are currently alive versus 53 percent of those with low protein levels, researchers said.

In an abstract presented at AACR, the researchers reported that patients in the PD-L1 high group had a progression-free survival rate at six months of 67 percent versus 11 percent in the low group. The overall survival rate at six months was 89 percent in the high PD-L1 group versus 33 percent in the low protein group. Median overall survival and progression-free survival levels have not been reached in the high PD-L1 cohort. Median progression-free survival and median overall survival in the low PD-L1 group was 2.1 months and 3.9 moths, respectively.

Previously treated NSCLC patients have poor outcomes, generally. Gandhi noted in a statement that the durability of responses to MK-3475 seen in PD-L1 high patients were promising.

Gandhi and her colleagues established the cut offs for high and low PD-L1 levels by analyzing pretreatment tumor protein levels in 146 patients and then gauging their response to the MK-3475. In this group, 37 percent of 41 patients with high PD-L1 levels responded to the drug, compared to 11 percent in the PD-L1 low group. Based on this, the researchers determined that tumor PD-L1 levels greater than 50 percent measured by immunohistochemistry differentiated between the responses seen in PD-L1-positive versus PD-L1-negative NSCLC patients.

Gandhi and her colleagues believe that patients with high levels of PD-L1 in their tumors respond particularly well to MK-3475 because the drug blocks interaction between proteins that keep their immune cells from being able to destroy cancer cells. PD-L1 proteins attach to another protein, called PD-1, on the surface of T cells, which in turn hinders the ability of the immune cells to destroy cancer cells.

Merck is conducting a randomized study to prospectively investigate the association of PD-L1 expression and NSCLC patients' response to MK-3475 compared to docetaxel.

Researchers also presented data at AACR looking at MK-3475 response in melanoma patients with positive and negative PD-L1 expression.

Based on data from around 100 melanoma patients, Adil Daud of the University of California, San Francisco, and colleagues, reported that 49 percent of 83 PD-L1 positive patients had responses and 13 percent out of 30 patients with PD-L1 negative tumors responded to MK-3475. The response rates increased a few percentage points in PD-L1 positive patients when the expression cut off separating positive and negative patients was increased from 1 percent or greater to 10 percent or greater.

The FDA granted Merck breakthrough designation for MK-3475 as a treatment for advanced melanoma. The company in January initiated a rolling biologics license application for the drug and said it plans to complete submitting the application with the agency in the first half of this year.