We saw a statistically significant and highly clinically significant improvement in overall survival with the addition of docetaxel for men starting hormone therapy for the first time.… But we also saw no evidence supporting the upfront use of zoledronic acid.
—Nicholas David James, MD, PhD
Our results show that there is no scientific rationale for advocating a strict time limit between primary excision and sentinel node biopsy [in melanoma patients].
—Charlotte Oude Ophuis, MD
We saw no improvement in overall survival when adding IMA901 to first-line sunitinib…. Immune responses need to be significantly improved before further development of IMA901.
—Brian Rini, MD
The 2015 European Cancer Congress (ECC), held recently in Vienna, represented the combined efforts of the European Cancer Organisation (ECCO), the European Society for Medical Oncology (ESMO), and other partner organisations, constituting the largest European platform for oncology education. At this year’s Congress, a global audience heard the first presentations of groundbreaking data, including dozens of “best” and “late-breaking” abstracts, and gained practical insights from some of the world’s foremost clinicians.
The ASCO Post has provided in-depth coverage of the meeting, both in print and online with The ASCO Post Evening News and The ASCO Post Newsreels. With the following brief summaries, we offer more noteworthy research coming out of the meeting.
STAMPEDE Update: No Benefit for Zoledronic Acid
An updated analysis of the STAMPEDE trial, which evaluated the early use of docetaxel and zoledronic acid in advanced prostate cancer, found no benefit of zoledronic acid in reducing deaths or skeletal-related events while confirming the significant overall survival benefit conveyed by docetaxel.1
“We saw a statistically significant and highly clinically significant improvement in overall survival with the addition of docetaxel for men starting hormone therapy for the first time. Docetaxel also significantly prolonged the time to first reported symptomatic skeletal-related event. But we also saw no evidence supporting the upfront use of zoledronic acid,” said Nicholas David James, MD, PhD, Director of the Cancer Research Unit at the University of Warwick and Queen Elizabeth Hospital Birmingham, in the United Kingdom.
For the comparison reported at ECC 2015, STAMPEDE accrued 2,962 advanced hormone therapy–naive prostate cancer patients, either newly diagnosed or relapsed. The current analysis focused on the comparisons between standard treatment with androgen-deprivation therapy with or without radiotherapy (n = 1,184), vs standard treatment plus zoledronic acid (n = 593), docetaxel (n = 592) or the combination (n = 593).
Dr. James presented updated data for survival and skeletal-related events, after a median follow-up of 43 months. Compared to standard of care only, the addition of docetaxel, zoledronic acid, or the combination yielded the following hazard ratios for survival: standard treatment plus docetaxel (hazard ratio [HR] = 0.78, P = .006), standard treatment plus zoledronic acid (HR = 0.94, P = .450), standard treatment plus the combination of docetaxel and zoledronic acid (HR = 0.82, P = .022).
Hazard ratios were similar for patients with newly diagnosed and previously treated disease. Patients with metastatic disease at presentation derived the most benefit from docetaxel. Only 20 deaths were observed in the nonmetastatic group receiving docetaxel.
In preventing skeletal-related events, Dr. James said, “Given what might have been expected, we saw a remarkably weak effect” for zoledronic acid—ie, a nonsignificant trend toward delaying events (HR = 0.89, P = .221). Among patients with bone metastases, the trend toward benefit was unexpectedly weak (HR = 0.94, P = .564).
In contrast, docetaxel was associated with a large treatment effect. The number of skeletal-related events was 328 with standard treatment vs 112 with the addition of docetaxel (HR = 0.60, P = .00000127), with similar protection seen among newly diagnosed men with bone metastases. The addition of zoledronic acid to this regimen did not enhance the benefit.
Enzalutamide in Triple-Negative Breast Cancer
In patients with advanced triple-negative breast cancer, treatment with a drug that inhibits the androgen receptor improved overall survival in patients with the relevant gene-expression profile, according to Javier Cortes, MD, PhD, Head of the Breast Cancer Program at Vall d’Hebron University Hospital, Barcelona.2
Triple-negative breast cancer comprises multiple subtypes and oncogenic drivers, including a subtype that may be driven by androgen receptor signaling. Enzalutamide (Xtandi), a potent androgen receptor–signaling inhibitor, significantly improves overall survival in metastatic castration-resistant prostate cancer and is currently being developed for patients with breast cancer who have an androgen-driven gene signature, Dr. Cortes explained.
MDV3100–11 is an open-label study evaluating enzalutamide in 118 patients with advanced triple-negative breast cancer patients, both untreated and previously treated. All patients expressed the androgen receptor, according to immunohistochemistry (> 0%). Previous studies have shown that more than three-fourths of triple-negative tumors express this receptor to some degree.
The study also evaluated patients according to the presence of an androgen-driven gene signature, as determined by the PREDICT AR assay. The assay was developed by prior gene-profiling analysis and validated in two independent data sets. Of 118 patients enrolled, 56 (47%) were PREDICT AR–positive.
“We found evidence of clinical activity in patients treated with enzalutamide in this phase II study,” Dr. Cortes reported. “We also found that a novel gene-expression profiling assay may identify triple-negative patients who appear to derive more benefit from enzalutamide.”
Median overall survival was 10.5 months longer in the PREDICT AR–positive vs PREDICT AR–negative patients (18.0 vs 7.5 months). Median overall survival for the whole population was 12.0 months. All other outcomes were also more improved in patients with a positive PREDICT AR status.
Median progression-free survival was 16.0 weeks in the PREDICT AR–positive group vs 8.0 weeks in the negative group. In the positive subset with no more than one prior treatment regimen (or none), median progression-free survival exceeded 32 weeks.
The clinical benefit rate at 16 weeks, which was the primary endpoint, was 39% in the PREDICT AR–positive group and 11% in the negative group. There were few responses, however: 9% and 3%, respectively. Median treatment duration for the positive group was 15 weeks.
In a multicovariate analysis of survival, both PREDICT AR status and line of therapy were significantly and independently associated with progression-free survival.
“Enzalutamide, a hormonal therapy that inhibits androgen-receptor signaling, may represent a novel therapeutic option in PREDICT AR–positive patients who would otherwise receive cytotoxic chemotherapy,” Dr. Cortes suggested.
Timing of Sentinel Node Biopsy in Melanoma
Restricting the time interval between primary surgical excision and sentinel node biopsy to less than 6 weeks does not appear to be necessary for patients with melanoma. Patients had similar survival outcomes at any cutoff interval from 1 to 8 weeks, according to a retrospective, multicenter cohort study presented during the Presidential Session.3
In both sentinel node–negative and –positive patients, melanoma-specific survival was not significantly different for any excision-to–sentinel node biopsy time interval cutoff from 1 to 8 weeks.
“Our results show that there is no scientific rationale for advocating a strict time limit between primary excision and sentinel node biopsy,” said Charlotte Oude Ophuis, MD, a PhD candidate at Erasmus MC Cancer Institute, Rotterdam, the Netherlands.
“Subanalyses in matched cohorts from the four centers with data on both sentinel node–positive and sentinel node–negative patients (n = 2,848) confirmed our finding that the time interval until sentinel node biopsy had no significant impact on melanoma-specific survival and disease-free survival,” she added.
Most international guidelines recommend reexcision plus sentinel node biopsy as soon as possible after primary excision, she explained, but before this study there were no data to support this recommendation.
The study analyzed data from 998 sentinel node biopsy–positive melanoma patients from nine different EORTC Melanoma Group Centers and 2,886 biopsy-negative melanoma patients from four different EORTC Melanoma Group Centers diagnosed between 1993 and 2012 with known primary excision dates. Compared with sentinel node–negative patients, those who were positive had tumors that were thicker (median Breslow thickness = 3.0 mm vs 1.7 mm) and more likely to be ulcerated (44% vs 25%).
The median time interval between primary excision and biopsy was similar in the two groups: 42 days in sentinel node–negative patients and 46 days in sentinel node–positive patients.
Among sentinel node–negative patients, older age and thinner tumors were associated with a shorter time interval until sentinel node biopsy. No significant associations were found between tumor or baseline characteristics and time interval in sentinel node–positive patients.
Advanced Melanoma: Sequencing of Checkpoint Inhibitors
In the treatment of advanced melanoma with checkpoint inhibitors, an unanswered question has been which agent to initiate first. The results of CheckMate 064, presented at ECC 2015 by F. Stephen Hodi, MD, Director of the Melanoma Center and Director of the Center for Immuno-Oncology at Dana-Farber Cancer Institute, Boston, suggested that starting off with nivolumab (Opdivo) is more efficacious but also more toxic.4
CheckMate 064 was a randomized, open-label phase II study evaluating the safety and efficacy of two immune checkpoint inhibitors given sequentially with a planned switch, in 138 patients who were treatment-naive or had received no more than one prior regimen that did not include antibodies against programmed cell death protein 1 (PD-1) or its ligand (PD-L1) or ipilimumab (Yervoy).
Patients received nivolumab at 3 mg/kg every 2 weeks for six cycles, followed by ipilimumab at 3 mg/kg every 3 weeks for four cycles (with a 2-week treatment-free interval in between) or the same regimens in the opposite order (with a 3-week treatment-free interval). All continued on nivolumab at 3 mg/kg every 2 weeks.
More patients had disease progression on the ipilimumab-followed-by-nivolumab schedule (56%) than with nivolumab followed by ipilimumab (27%). However, ipilimumab followed by nivolumab was associated fewer treatment discontinuations due to toxicity (14%) than nivolumab followed by ipilimumab (31%).
“Consistent improvement in efficacy outcomes was demonstrated with [nivolumab followed by ipilimumab] vs [ipilimumab followed by nivolumab],” said Dr. Hodi.
Confirmed responses were observed in 41.2% vs 20.0%; when unconfirmed responses were included, the response rates were 47.7% and 22.6%, respectively. Median changes in tumor burden from baseline at week 13 was −27% with nivolumab folowed by ipilimumab and +10% with ipilimumab followed by nivolumab and at week 25 were −50% and −17%, respectively.
During the induction periods, grade 3–4 adverse events were observed in 50% of patients receiving nivolumab followed by ipilimumab and in 43% of those given ipilimumab followed by nivolumab. There were no treatment-related events.
“The nature of the adverse events was similar to those previously observed with either agent, alone and in combination, and the frequency of adverse events was consistent with previous reports for the combination,” Dr. Hodi said.
In correspondence with The ASCO Post, he emphasized, “The trial was designed as a biomarker trial and had a forced switch from one agent to another. Patients did not switch at the time of progression, but switched whether they were responding or not. Efficacy data need more time to mature.”
PD-L2 Expression: Biomarker?
Efforts to identify a biomarker for response to anti–PD-1 therapy have focused on tumor expression of the PD-1 ligand, PD-L1. New insights suggest that PD-L2 expression may be a potential biomarker for clinical response to pembrolizumab (Keytruda), an anti–PD-1 antibody.5
“PD-L2 is one of two known binding partners of PD-1, and it has been shown to be involved with disease severity in murine models,” said Jennifer Yearley, DVM, PhD, the Anatomic Pathology Group Lead at Merck.
Using a proprietary immunohistochemistry assay to detect PD-L2 expression, the investigators sought to determine the potential relevance of PD-L2 expression in multiple cancers being treated with pembrolizumab in the clinic, including renal cell carcinoma, bladder cancer, non–small cell lung cancer, head and neck squamous cell carcinoma, triple-negative breast cancer, gastric cancer, and melanoma. In these cohorts, PD-L2 was expressed in varying degrees.
Each cohort was evaluated for overall expression of PD-L2 in stromal cells, tumor cells, and endothelial cells. Particularly high levels of PD-L2 expression were observed in gastric carcinoma and triple-negative breast cancer.
Both concordance and discordance were observed between PD-L1 and PD-L2 expression, but in general, the patterns of one closely mirrored the other. PD-L2 expression can occur in the absence of PD-L1 expression, Dr. Yearley said. In a pilot analysis of data from the KEYNOTE-12 study in patients with head and neck cancer, about two-thirds of all tumors scored PD-L2–positive, and response rates were twice as high for tumors positive for both PD-L1 and PD-L2 expression as for tumors positive only for PD-L1, she added.
In the KEYNOTE-12 cohort, PD-L2 expression was associated with higher response rates and longer progression-free survival in pembrolizumab-treated patients, she said.
“This is a tale of two ligands. The study is hypothesis-generating based on archived material. It needs to be validated with pembrolizumab and nivolumab. This will be very important to guide development of PD-1 blockers further,” said formal discussant Ignacio Melero, MD, PhD, Senior Investigator at the University of Navarra in Pamplona, Spain.
Vaccine Fails in Renal Cell Carcinoma
The IMA901 multipeptide vaccine failed to improve overall survival when combined with sunitinib vs sunitinib alone in patients with metastatic renal cell carcinoma in an open-label phase III trial.6
“We saw no improvement in overall survival when adding IMA901 to first-line sunitinib. Survival was comparable between the two arms of the study in favorable patients, and sunitinib alone achieved longer overall survival in intermediate-risk patients. Immune responses need to be significantly improved before further development of IMA901,” said lead author Brian Rini, MD, a hematologist/oncologist at the Cleveland Clinic, Ohio. The disappointing phase III results failed to replicate positive findings in a phase II trial, he noted.
The study enrolled 339 patients with favorable- or intermediate-risk metastatic renal cell carcinoma who were HLA-A–positive. Patients were randomly assigned 2:1 to receive first-line sunitinib plus IMA901 (n = 204) vs sunitinib alone (n = 135). Sunitinib was chosen as the comparator arm because it is standard first-line therapy.
Median overall survival was 33.1 months for the experimental arm vs not yet reached for sunitinib alone (P = .08). When survival was analyzed according to risk level, the curves overlapped for favorable-risk patients and favored the control arm in the intermediate-risk group. Progression-free survival was similar in both arms, about 15 months by central review.
Adverse events were similar across both arms. Dr. Rini hypothesized that the effect of sunitinib on monocytes may have had a negative effect on the vaccine response.
Although IMA901-specific CD8 T-cell responses were observed with the vaccine, there was no clear association between survival and observed immune response.
“This is another example of a huge difference between our goals with tumor vaccines and what actually happens. Many trials in tumor vaccines fail,” said Dr. Melero, who was formal discussant for the trial. “Many steps are needed to develop therapeutic vaccines.”
He added, “The data from phase II were not replicated. Those patients who mounted an immune response did not do better.”
Aspirin as Secondary Prevention
Aspirin improves survival in patients diagnosed with gastrointestinal tumors, according to results of a large study.7 This is the first study that analyzed survival from patients with tumors located at different sites in the gastrointestinal tract at the same time, rather than one type.
The study was based on 13,715 patients from the southern region of the Netherlands who had been diagnosed with a gastrointestinal cancer between 1998 and 2011. These cases were linked to drug-dispensing information from PHARMO, a database that contains prescription data from these patients.
About 30% of patients used aspirin before diagnosis, 8.3% used aspirin only after a diagnosis, and 61.1% took no aspirin either before or after diagnosis. Distribution of the most common tumor types was: colon (42.8%), rectum (25.4%), and esophagus (10.2%). Median follow-up was 48.6 months.
Survival was twice as high in patients who used aspirin post diagnosis. Overall survival at 5 years was 75% among aspirin users, compared to 42% among nonusers. An approximately 50% reduction in death was observed for all patients with a gastrointestinal malignancy. The effect persisted in all individual tumor types (esophageal cancer, colon cancer, rectal cancer, hepatobiliary cancer, stomach cancer). However, the effect was not present in pancreatic cancer.
The survival benefit of postdiagnosis aspirin was observed in an analysis that adjusted for confounding factors such as age, sex, stage of cancer, surgery, radiotherapy, chemotherapy, and comorbidities.
The researchers would like to analyze individual tumor specimens in patients with a beneficial response to aspirin to try to identify characteristics associated with benefit, said study coordinator Martine Frouws, MSc, a PhD candidate at Leiden University Medical Center in the Netherlands.
The authors hypothesized that the beneficial effect of aspirin is related to its antiplatelet effects. Aspirin’s inhibition of platelet function may allow the immune system to attack these circulating tumor cells.
An ongoing multicenter randomized, placebo-controlled trial in the Netherlands is evaluating daily aspirin (80 mg) in patients with colon cancer. The primary endpoint is overall survival.
“Given that aspirin is an inexpensive, off-patent drug with relatively few side effects, this will have a great impact on health-care systems as well as patients,” said Ms. Frouws. ■
Disclosure: Dr. James has served on the Advisory Board of and has received research funding and honoraria from Sanofi-Aventis. Dr. Cortes has been a consultant for Celgene and Roche and has received honoraria from Celgene, Roche, Novartis, and Eisai. Dr. Hodi has served as a consultant or advisor for Bristol-Myers Squibb, Novartis, and Merck and has received research funding from Bristol-Myers Squibb, Genentech, Novartis, and Merck. Dr. Yearley is an employee of Merck. Dr. Rini disclosed a relationship with Immatics and Pfizer. Dr. Oude Ophuis and Ms. Frouws reported no potential conflicts of interest.
1. James ND, Sydes MR, Mason MD, et al: Docetaxel +/- zoledronic acid for hormone-naive prostate cancer: First overall survival results from STAMPEDE and treatment effects within subgroups. 2015 European Cancer Congress. Abstract LBA19. Presented September 27, 2015.
2. Cortes J, Crown J, Awada A, et al: Overall survival (OS) from the phase 2 study of enzalutamide (ENZA), an androgen receptor (AR) signaling inhibitor, in AR+ advanced triple-negative breast cancer. 2015 European Cancer Congress. Abstract 1802. Presented September 26, 2015.
3. Oude Ophuis CO, Verhoef C, Rutkowski P, et al: The interval between primary melanoma excision and sentinel node biopsy does not affect survival, regardless of SNB status: An EORTC Melanoma Group study. 2015 European Cancer Congress. Abstract 2BA. Presented September 28, 2015.
4. Hodi FS, Gibney G, Sullivan R, et al: An open-label, randomized, phase 2 study of nivolumab given sequentially with ipilimumab in patients with advanced melanoma (CheckMate 064). 2015 European Cancer Congress. Abstract 23LBA. Presented September 27, 2015.
5. Yearley J, Gibson C, Yu N, et al: PD-L2 expression in human tumors: Relevance to anti-PD-1 therapy in cancer. 2015 European Cancer Congress. Abstract 18LBA. Presented September 27, 2015.
6. Rini B, Stenzl A, Zdrojowy R, et al: Results from an open-label, randomized, controlled phase 3 study investigating IMA901 multipeptide cancer vaccine in patients receiving sunitinib as first-line therapy for advanced/metastatic RCC. 2015 European Cancer Congress. Abstract 17LBA. Presented September 27, 2015.
7. Frouws M, Bastiaannet E, Van Herk-Sukel M, et al: Aspirin and gastrointestinal malignancies improved survival not only in colorectal cancer. 2015 European Cancer Congress. Abstract 2306. Presented September 28, 2015.