You are here
American Society of Clinical Oncology 2018
This year’s annual meeting of the American Society of Clinical Oncology (ASCO) hosted more than 40,000 medical professionals in Chicago from June 1 to 4. We review here key sessions across a wide range of disease states, with clinical findings in prostate cancer, non–small-cell lung cancer, breast cancer, esophageal cancer, pancreatic cancer, leukemias, metastatic colon cancer, and a session on molecular profiling guidance of therapy.
Overall Survival Differences Between African-American and Caucasian Men With Metastatic Castration-Resistant Prostate Cancer
- Susan Halabi, PhD, Professor of Biostatistics, Duke University, Durham, North Carolina
African-American men in the U.S. are diagnosed at a later stage of prostate cancer than Caucasian men and have higher prostate cancer incidence, higher mortality rates, and unequal access to care. Based on this, the study hypothesized that African-American men with advanced and lethal prostate cancer would have worse survival than Caucasian men. Analysis of the metastatic castration-resistant prostate cancer patients enrolled in the National Clinical Trials Network, however, revealed instead a 24% lower risk of death among African-American men than among Caucasian men, according to Dr. Halabi. Prior analyses have been limited by small sample size, she said at an ASCO press briefing.
The percentage of African-American men in the National Clinical Trials Network studies was 12%, compared with 4% in industry trials. All patients had been randomized to docetaxel/prednisone or regimens containing docetaxel/prednisone. The National Clinical Trials Network studies included 8,820 men enrolled in nine phase 3 clinical trials with overall survival (OS) as their primary endpoint. Among that total, 7,528 (85%) were Caucasian, 500 (6%) were African-American, 424 (5%) were Asian, and 368 (4%) were other or unknown.
Dr. Halabi’s analysis included Caucasian and African-American men only (n = 8,452), with a mean age of 69 years. Because poor prognostic factors, such as higher prostate-specific antigen levels and worse performance status, were more common among African-American men, the researchers conducted a multivariable analysis that compared outcomes in African-American men to Caucasian men with the same prognostic factors. Median OS was 21 months in both groups. Adjusting for additional prognostic factors, such as patient age and levels of alkaline phosphatase and hemoglobin, researchers found that African-American men were 19% less likely than Caucasian men to die from any cause. Median survival in the National Clinical Trials Network among Caucasian men was 20 months compared with 21 months among African-American men (hazard ratio, 0.76; P < 0.0001).
What explains the unexpected finding? The lower risk of death in African-American men may reflect differences in the biology of the disease, Dr. Halabi speculated, or perhaps African-American men better tolerate the docetaxel/prednisone combination. Dr. Halabi pointed out that all patients included in the study were eligible to be enrolled in clinical trials and had access to clinical trials. Further genomic analyses will explore whether biological variations explain differences in outcomes by race.
Pointing to the low percentage of African-Americans enrolled in clinical trials, she concluded, “New methodologies for enrolling higher numbers of African-American men should be actively pursued and vigorously implemented so minority groups are well represented on trials.” She also emphasized a further conclusion: “Results from clinical trials cannot be generalized to U.S. populations.”
IMpower131: Primary PFS and Safety Analysis of a Randomized Phase 3 Study of Atezolizumab + Carboplatin + Paclitaxel or Nab-Paclitaxel Versus Carboplatin + Nab-Paclitaxel as 1L Therapy in Advanced Squamous NSCLC
- Robert Jotte, MD, PhD, Rocky Mountain Cancer Centers, Denver, Colorado
IMpower131 data on risk of disease-worsening or death in patients with advanced squamous non–small-cell lung cancer (NSCLC) show greater benefit with front-line atezolizumab (Tecentriq, Genentech) plus chemotherapy than with chemotherapy alone.
Squamous classification applies to about 25% to 30% of NSCLC patients, Dr. Jotte said in a press conference. Squamous NSCLC is difficult to treat, and new treatment options over the last few decades have been limited. First-line standard of care with advanced squamous disease includes platinum-based chemotherapy regimens. Trials of first-line carboplatin plus nab-paclitaxel have reported median progression-free survival (PFS) of 5.6 months, median overall survival (OS) of 10.7 months, and objective response rates (ORRs) of 41%.
Investigators for the multicenter, open-label, phase 3 IMpower131 study randomized 1,021 chemotherapy-naïve patients with stage IV squamous NSCLC to first-line treatment with atezolizumab, carboplatin, and paclitaxel (arm A); atezolizumab, carboplatin, and nab-paclitaxel (arm B); or carboplatin and nab-paclitaxel as controls (arm C). Induction in arm A was four or six cycles of atezolizumab (1,200 mg) plus carboplatin (area under the curve, 6) and paclitaxel (200 mg/m2) given on day 1 of each 21-day cycle. Patients then received maintenance atezolizumab every three weeks. Arm B substituted nab-paclitaxel (100 mg/m2) for paclitaxel, but was otherwise the same as Arm A. Arm C patients received induction of carboplatin and nab-paclitaxel (100 mg/m2), with carboplatin on day 1 of each 21-day cycle, and nab-paclitaxel was administered on days 1, 8, and 15 of each 21-day cycle. The maintenance phase consisted of best supportive care. Investigator-assessed PFS and OS were the co-primary endpoints.
Analysis of Arm B compared with chemotherapy-only controls (Arm C) showed median PFS in the atezolizumab plus chemotherapy arm to be significantly longer (6.3 months vs. 5.6 months; hazard ratio [HR], 0.71; P = 0.0001) after a median follow-up of 17.1 months. One-year PFS was 24.7% and 12.0% in the atezolizumab plus chemotherapy and chemotherapy-alone arms, respectively. The PFS benefit was consistent across programmed death ligand-1 (PD-L1) subgroups—positive, high, low, and negative—but was greatest in the high PD-L1 group. In the atezolizumab plus chemotherapy arm, median PFS was longest in the high PD-L1 group (10.1 months) and shortest in the PD-L1 negative subgroup (5.7 months). Median PFS in the control group was approximately 5.6 months.
ORRs were higher and median duration of response was longer for atezolizumab plus chemotherapy versus controls (ORR, 49% versus 41%; duration of response, 7.2 versus 5.2 months, respectively). The first interim analysis revealed similar OS durations of 14.0 and 13.9 months for atezolizumab plus chemotherapy vs. chemotherapy alone (HR, 0.96; P = 0.6931).
Dr. Jotte noted that atezolizumab with chemotherapy has a manageable safety profile with no new safety signals. The next interim analysis is anticipated later in 2018.
“With this regimen we can double the chance that patients don’t have to hear the bad news that their disease has progressed in the first year of treatment,” observed ASCO expert commentator David Graham, MD.
TAILORx: Phase 3 Trial of Chemoendocrine Therapy Versus Endocrine Therapy Alone in Hormone Receptor-Positive, HER2-Negative, Node-Negative Breast Cancer and an Intermediate Prognosis 21-Gene Recurrence Score
- Joseph A. Sparano, MD, Montefiore Medical Center, Bronx, New York
A clinical trial designed to clear up uncertainty about the best treatment for women with hormone receptor-positive, HER2-negative, axillary node-negative early-stage breast cancer found that most do not need chemotherapy after surgery, according to Dr. Sparano. The investigation was focused on women with mid-range scores of 11–25 on a 21-tumor gene expression assay (Oncotype DX Breast Recurrence Score).
Dr. Sparano noted that the 21-gene assay, based on a tumor biopsy sample, is widely used to predict 10-year recurrence risk and to identify those women most likely to derive strong benefit from chemotherapy. About half of all breast cancers are hormone receptor-positive, HER2-negative, and axillary node-negative. Hormone therapy alone is generally given to women with low scores of 0–10, while women with high scores of 26–100 receive hormone therapy and chemotherapy. The substantial side effects of chemotherapy—potentially nausea, vomiting, alopecia, fatigue, infection, early menopause, infertility, neuropathy, cardiac damage, and leukemia—lend urgency to identifying women who may forego this treatment.
Hormone receptor-positive, HER2-negative, axillary node-negative breast cancer is the most common type of breast cancer. In TAILORx (Trial Assigning Individualized Options for Treatment), investigators enrolled 10,273 women with this cancer type, finding 6,711 of them to have mid-range recurrence scores of 11–25. They were then randomized to receive either hormone therapy alone (n = 3,399) or hormone therapy and chemotherapy (n = 3,312). In a noninferiority comparison, the primary endpoint was disease-free survival, defined as absence of recurrence of cancer in the breast, regional lymph nodes, and/or distant organs; a second primary cancer in the contralateral breast or another organ; or death from any cause.
Median age was 55 years; 33% were at least 50 years of age. Tumor sizes of 1–2 cm were present in 63% of women. Intermediate grade scores were found in 57% of women. Analysis showed endocrine therapy to be noninferior to endocrine therapy plus chemotherapy (hazard ratio, 1.08; 95% confidence interval, 0.94–1.24; P = 0.26). Nine-year invasive disease-free survival rates were 83.3% and 84.3% for endocrine therapy and endocrine therapy plus chemotherapy, respectively. Distant recurrence rates were 5% for both groups and overall survival at nine years was similar irrespective of chemotherapy administration. An exploratory analysis showed some benefit of chemotherapy in women 50 years of age or younger with scores of 16–25 (2% fewer distant recurrences with scores of 16–20 and 7% fewer with scores of 21–25).
Guiding therapy with TAILORx recurrence score findings would spare 69% of women from receiving chemotherapy and select chemotherapy for 31%, Dr. Sparano said. Among those spared chemotherapy, he listed: 1) all women older than 50 years of age with hormone-receptor positive, HER2-negative, node-negative breast cancer and a recurrence score of 0–25, and 2) all women 50 years of age or younger with hormone-receptor positive, HER2-negative, node-negative breast cancer and a recurrence score of 0–15.
He concluded, “Any woman with early-stage breast cancer [who is] 75 years or younger should have the test and discuss the results of TAILORx with her doctor to guide her decision regarding chemotherapy after surgery to prevent recurrence.”
Chemoprevention of Esophageal Cancer With Esomeprazole and Aspirin Therapy: Efficacy and Safety in the Phase 3 Randomized Factorial AspECT Trial
- Janusz Jankowski, MD, PhD, National Institute for Health and Care Excellence, United Kingdom
In patients with Barrett’s metaplasia, the best chemoprevention against esophageal cancer is high-dose esomeprazole, according to randomized, phase 3 AspECT trial results. Five-year survival for the more than 52,000 esophageal adenocarcinoma cases diagnosed each year is less than 10%, Dr. Jankowski said in an ASCO press briefing.
Esophageal cancer incidence is rising, most likely related to observed increases in gastroesophageal reflux/Barrett’s esophagus and associated inflammation. Both proton pump inhibitors (PPIs) and aspirin may have chemopreventive properties, Dr. Jankowski said, with PPIs reducing acid reflux and with aspirin reducing inflammation.
In some people who have chronic gastroesophageal reflux disease (GERD) or esophagitis, even in the absence of chronic heartburn symptoms, Barrett’s esophagus can develop. In Barrett’s, squamous cells in the lining of the esophagus turn into glandular tissue. While the risk is fairly low, Barrett’s esophagus increases the chances of developing adenocarcinoma of the esophagus, Dr. Jankowski said. However, while 80% to 90% of esophageal cancers are preceded by Barrett’s esophagus, the latter is rarely diagnosed first, because diagnosis requires endoscopy. Acid reduction with standard-dose PPIs or inflammation reduction with aspirin could interdict progression from Barrett’s esophagus to cancer, Dr. Jankowski stated.
In the AspECT trial, investigators tested four treatment strategies among 2,563 patients with Barrett’s esophagus. The treatment groups received either a high-dose PPI (esomeprazole 40 mg twice daily), high-dose esomeprazole with low-dose aspirin (300 mg a day, or 330 mg a day in Canada), low-dose esomeprazole (20 mg once daily) with low-dose aspirin, or low-dose esomeprazole. Combined death from any cause, diagnosis of esophageal cancer, or high-grade dysplasia was the primary endpoint.
The combined endpoint occurred significantly less often among patients receiving high-dose esomeprazole versus low-dose esomeprazole (P = 0.0459) after a median follow-up of 8.9 years. In addition, there was a 27% delay in the time to the composite endpoint with high-dose versus low-dose esomeprazole (P = 0.037). Evidence favoring esomeprazole use was strongest in the mortality benefit among the endpoint components. While even more benefit was shown for high-dose esomeprazole with low-dose aspirin, the trial was not adequately powered for conclusions in this subset. Also, the aspirin versus no aspirin comparison failed to show a benefit, except in a further analysis with censoring for prior nonsteroidal anti-inflammatory drug use, Dr. Jankowski said. Across treatment groups, total severe adverse events were similar.
Experts at the ASCO press briefing urged caution regarding interpretation and concern about the over-the-counter (OTC) availability of PPIs.
ASCO press briefing moderator Richard L. Schilsky, MD, warned, “These are OTC drugs. We don’t want people to rush out and start taking high doses of proton pump inhibitors with aspirin. Diagnosing Barrett’s requires an endoscopy.”
The Unicancer GI PRODIGE 24/CCTG PA.6 Trial: A Multicenter International Randomized Phase 3 Trial of Adjuvant mFOLFIRINOX Versus Gemcitabine in Patients With Resected Pancreatic Ductal Adenocarcinomas
- Thierry Conroy, MD, Institut de Cancérologie de Lorraine, Vandoeuvre-Les-Nancy, France
For pancreatic cancer patients who have good performance status after resection, a modified (m) FOLFIRINOX regimen should now be considered the standard of care, Dr. Conroy said in an ASCO press briefing.
There has been a clear need for a better regimen, given that 71% to 76% of patients will relapse within two years of receiving standard care. That has been six months of post-surgery gemcitabine and/or fluoropyrimidine since it was shown in 2008 that additional chemotherapy with these agents reduces recurrence risks. Dr. Conroy noted that FOLFIRINOX is more effective than gemcitabine as first-line treatment in metastatic pancreatic cancer in patients with good performance status and that mFOLFIRINOX maintains efficacy while avoiding the bolus fluorouracil associated with higher hematologic toxicities and diarrhea.
PRODIGE included 493 patients with resected pancreatic cancer and no tumor residue on mandatory postoperative computed tomography scans and good performance status. All had nonmetastatic pancreatic ductal adenocarcinoma. They were treated at 77 French and Canadian centers and were randomized 1:1 to either gemcitabine or mFOLFIRINOX (oxaliplatin, leucovorin, irinotecan, and 5-fluorouracil).
Dr. Conroy reported that the primary endpoint of median disease-free survival was 21.6 months with mFOLFIRINOX and 12.8 months with gemcitabine (hazard ratio [HR], 0.58;
P < 0.0001) after a median follow-up of 33.6 months. Three-year disease-free survival rates were 39.7% and 21.4% for mFOLFIRINOX and gemcitabine, respectively. Benefits for mFOLFIRINOX were observed in all subgroups. Similarly, the secondary endpoint of median overall survival (OS) was 54.4 months with mFOLFIRINOX and 35.0 months with gemcitabine (HR, 0.64; P = 0.003). Three-year OS rates were 63.4% for mFOLFIRINOX and 48.6% for gemcitabine.
While adverse events were more common in the mFOLFIRINOX group (diarrhea, fatigue, vomiting, sensory peripheral neuropathy, mucositis, and thrombocytopenia), febrile neutropenia rates were similar in both groups (2.9% for mFOLFIRINOX, 3.7% for gemcitabine). One death due to toxicity occurred in the gemcitabine arm. Also, grade 1–4 headache, fever, flu-like symptoms, lymphopenia, and transaminase increases were reported more frequently in the gemcitabine arm.
“In patients with good performance status, the mFOLFIRINOX regimen should now be considered a new standard of care after pancreatic cancer resection,” Dr. Conroy concluded. “While mFOLFIRINOX is more toxic than gemcitabine, it is a safe regimen with manageable toxicities.”
COG AALL0434: A Randomized Trial Testing Nelarabine in Newly Diagnosed T-Cell Malignancy
- Kimberly Dunsmore, MD, Professor of Medicine, Virginia Tech Carilion School of Medicine, Roanoke, Virginia
Nelarabine (Arranon, Novartis) added to standard chemotherapy produced the highest survival rates ever reported in children and young adults with T-cell acute lymphoblastic leukemia (T-ALL) or T-cell lymphoblastic lymphoma (T-LL). The randomized, phase 3 Children’s Oncology Group (COG) study (AALL0434) was federally funded, and was the largest ever conducted in this population, Dr. Dunsmore said.
T-ALL or T-LL affect one in six children and young adults diagnosed with acute lymphoblastic lymphoma. Twenty percent of children and adolescents enrolled in COG T-ALL trials did not survive beyond four years between 2000 and 2005, despite very intense and complex chemotherapy. “We felt we could and must do better,” Dr. Dunsmore said, noting that nelarabine, a T-cell-specific agent, has demonstrated success in relapsed patients. It was approved in 2005 for T-ALL or T-LL that had progressed after at least two chemotherapy regimens.
Investigators enrolled 1,895 patients 1 to 30 years of age newly diagnosed with either T-ALL (94%) or T-LL (6%). The trial had four arms, with all patients receiving the standard, complex, multidrug chemotherapy regimen known as COG augmented Berlin-Frankfurt-Munster chemotherapy. In addition, patients were randomly assigned to also receive either high-dose methotrexate in a hospital or escalating-dose methotrexate in an outpatient setting. Patients with moderate or high risk of cancer recurrence were also randomly assigned to receive or not to receive nelarabine and cranial radiation to prevent or treat brain metastases.
At four years, overall, the disease-free survival rate was 84.3% and survival was 90.2%. Among patients with T-ALL with increased risk of recurrence, 88.9% were leukemia-free at four years, compared with 83.3% of those not receiving nelarabine. Adding nelarabine did not benefit T-LL patients. However, more than 85% survived four years. Also, four-year survival in patients who did not achieve induction remission was 54%, more than double past survival rates, Dr. Dunsmore observed.
Contrary to results from previous, smaller trials, patients with T-ALL who received escalating doses of methotrexate fared better than those who received high-dose methotrexate. Four-year disease-free survival with the escalating dose was 89.8% versus 78% with the high dose.
Beyond better survival, patients receiving nelarabine had fewer central nervous system relapses. She noted that most physicians are moving to decrease the use of cranial radiation for T-cell leukemia because of possible post-cranial radiation side effects. She urged study of protocols of nelarabine without cranial irradiation that potentially could decrease long-term, late neurological side effects, which may include changes in cognitive abilities, learning disabilities, neuroendocrine changes, and development of secondary malignancies. Dr. Dunsmore stated, “Our trial shows that we could further increase survival rates by about 10%, which is very encouraging.”
Comparison of Systemic Therapy Use, Cost, and Survival in Patients With Metastatic Colorectal Cancer in Western Washington, U.S., and British Columbia, Canada
- Todd Yezefski, MD, Fred Hutchinson Cancer Research Center, Seattle, Washington
While systemic therapy for metastatic colorectal cancer is significantly more expensive in Washington state compared with nearby British Columbia, overall survival rates are similar, according to a comparison of data from separate registries: the British Columbia Cancer Registry and the Surveillance, Epidemiology, and End Results (SEER) Program linked to claims from two large commercial U.S. insurers.
“Few studies,” Dr. Yezefski said at an ASCO press briefing, “have compared health care use, cost, and outcomes between the U.S. and Canada for a specific disease type, such as colorectal cancer.” Dr. Yezefski noted that in the U.S., mostly private, employer-sponsored insurance plans are prevalent, while the elderly and poor are covered through publicly funded programs. Canada has a single-payer governmental system.
Investigators collected data to compare demographics, systemic therapy use, cost, and survival from 575 advanced colorectal cancer patients from western Washington state and 1,622 from British Columbia. In addition to being geographically close, British Columbia and western Washington both have a mostly white population, with a large Asian minority. They are also similar in income level and education. The median age was 60 and 66 years from the two locations, respectively, among whom 48% and 57% were male. Researchers could not access Medicare claims, skewing data collection toward younger patients. Systemic therapy was administered in 79% of patients in western Washington and in 68% of patients in British Columbia. The most frequently used first-line regimen was FOLFOX (oxaliplatin, 5-fluorouracil, and folinic acid) in western Washington (39%) and FOLFIRI (folinic acid, fluorouracil, irinotecan) plus bevacizumab in British Columbia.
The monthly mean cost of first-line therapy, in 2009 U.S. dollars, was $12,345 in western Washington and $6,195 in British Columbia (P < 0.01). The mean lifetime cost of therapy per month was $7,883 and $4,830, respectively (P < 0.01). Median overall survival with systemic therapy was 21.4 months in western Washington and 22.1 months in British Columbia.
Dr. Yezefski concluded, “Despite significantly higher cost for systemic therapy in western Washington compared to British Columbia, overall survival was similar. You got the same bang for more buck.” He observed that allowing Medicare to negotiate drug prices may lead to decreased costs. Medicare reimbursement rates, he pointed out, often guide those for private insurers.
Richard Schilsky, MD, the briefing moderator and ASCO’s Chief Medical Officer, said, “This study adds important context to the ongoing national conversation about rising treatment costs. As oncologists, we see the burden of high costs on patients and their families every day. In fact, as ASCO’s National Cancer Opinion Survey shows, many patients even forgo, delay, or skimp on treatments due to costs, potentially compromising their effectiveness.”
Precision Medicine: Clinical Outcomes Including Long-Term Survival According to the Pathway Targeted and Treatment Period–the IMPACT Study
- Apostolia Maria Tsimberidou, MD, PhD, Professor of Medicine, University of Texas MD Anderson Cancer Center, Houston, Texas
Patients whose treatment had been selected prospectively through molecular profiling had slower cancer growth and longer survival in a retrospective, multivariate analysis of advanced cancer patients. The finding was consistent, Dr. Tsimberidou said in an ASCO press briefing, across a diverse set of cancer types.
The IMPACT (Initiative for Molecular Profiling and Advanced Cancer Therapy) clinical trial was initiated in 2007 to evaluate the impact of personalized therapy selection based on molecular testing of tumors. Subjects were patients with difficult-to-treat cancers that had worsened despite standard therapy. Tumor molecular testing certified under the Clinical Laboratory Improvement Amendments (one to 50 genes analyzed) was performed in 3,743 consecutive patients, with one or more targetable molecular alterations identified in 1,307 (34.9%). Among these, matched targeted therapies were available among 711 (54.4%), while 596 (45.6%) remained unmatched. Patients’ median age was 57 years (range, 16–86); 39% were male. Cancer types were gastrointestinal in 24.2%, gynecological in 19.4%, breast in 13.5%, melanoma in 11.9%, and lung in 8.7%. The median number of prior therapies was four (range, 0–16).
Single-agent targeted therapy, as well as matched targeted therapy in combination with chemotherapy or other treatments, qualified as matched therapies. Among 697 patients with matched therapies, rates of objective response and stable disease for at least six months were 16.2% and 18.7%, respectively (34.9% combined). Among nonmatched patients (n = 571), the rates of objective response and stable disease for at least six months were 5.4% and 14.7%, respectively (20.1% combined). The difference between totals was significant (P < 0.001). Progression-free survival (PFS) was 4 months in the matched group and 2.8 months in the unmatched group (hazard ratio [HR], 0.67; P < 0.001). Median overall survival (OS) was longer in the matched therapy group versus the nonmatched therapy group (9.3 months versus 7.3 months; HR, 0.72; P < 0.001). In multivariate analysis, significant risk factors reducing OS included nonmatched therapy; P13K/Ak/tmTOR alterations; liver metastases; lactate dehydrogenase, albumin, or platelet count greater than the upper limit of normal; performance status greater than 1; and being 60 years of age or older.
“Matched therapy was an independent factor predicting longer survival in multivariate analysis,” Dr. Tsimberidou said, adding that the three-year OS rate was 15% in the matched group and 7% in the nonmatched group. Ten-year rates were 6% and 1%, respectively.
“It’s been 20 years since the first targeted therapy was introduced. These first ‘precision medicine’ therapies revolutionized cancer care and helped many patients live longer,” said ASCO expert Catherine Diefenbach, MD. “But we’ve just scratched the surface. Now with faster and more robust genetic tests, we can help even more patients by treating the cancer based on its genetic makeup rather than solely on its location in the body.”
The ongoing follow-on study, IMPACT 2, is a randomized, phase 2 study comparing PFS in patients who receive targeted therapy matched to the molecular characteristics of their tumor versus PFS in patients for whom treatment is not selected based on the molecular analysis of their tumor.