ASCO Roundup: Biosimilars, Precision Medicine, and New Inroads
By Neil Canavan
“Collective Wisdom,” the chosen theme for the annual meeting of the 2016 American Society of Clinical Oncology (ASCO) implies a certain amount of reflection. Indeed, this tone was echoed by any number of presentations that represent a clinical meditation – a consideration not only of the fervently new, but a drive to optimize gains already in hand.
Highlighted here: A biosimilar that will expand access to one of oncology’s most effective agents; inroads with liquid biopsies that have the potential to greatly facilitate patient assessments while on-treatment; a validation of methods to gauge the tumor-fighting ability of a patient’s immune system; and trial designs that explore new indications for already approved drugs.
As for the novel, two new agents will be touched on: one for small cell lung cancer, and the other for patients with advanced bladder cancer who are ineligible for treatment with the standard-of-care.
Herceptin: Take Two
Generic drugs based on small molecule formulations are nothing new, their design and manufacture are generally straightforward. This is not the case with drugs comprised of proteins, the so-called biologics, which by nature are much larger and more complex in design and often difficult to manufacture.
For reasons both technical and legal, (oh that pesky IP), very few generic biologics, or biosimilars have come to market. The first US approval of a biosimilar came in March 2015 with the nod going to filgrastim-sndz, a generic form of Amgen’s Neupogen; while a second drug, a generic version of the blockbuster rheumatoid arthritis medication, Remicade, was approved in April of this year. To date, there are no approved generic oncology biologics – but that situation is poised to change.
“This is one of the first trials with biosimilars in oncology to demonstrate similar efficacy, safety, and immunogenicity against the reference product,” said Hope Rugo, MD, University of California San Francisco, Helen Diller Family Comprehensive Cancer Center, adding that the eventual approval of the biosimilar in question, MYL-14010, a stand-in for Herceptin, would be a game changer.
“Biosimilars have the potential to significantly improve access to (these) expensive agents,” medications that are priced beyond the reach of patients in many developing countries.
The present phase III investigation, the Heritage Study, pitted MYL-14010 against Herceptin in a cohort of treatment-naïve patients with HER2+ metastatic breast cancer (N=500). Patients were randomized in a 1:1 fashion to either comparator agent, plus a taxane, and followed until their disease progressed.
Results showed that after 24 weeks of treatment the overall response rate was 69.9% for MYL-14010 vs. 64% for Herceptin – well within the pre-defined equivalence margin. A secondary endpoint of median progression-free survival has not yet been reached but event rates are, as yet, similar: 41 events for MYL-14010 vs. 48 events for Herceptin (Figure 1).
Results confirmed efficacy equivalence based on ratio of ORR & difference in ORR. Rates for adverse events for the two agents were also comparable.
“These results using a proposed biosimilar to Herceptin are incredibly exciting,” said Don Dizon, MD, clinical co-director of gynecologic oncology at Massachusetts General Hospital. “It has the potential to really broaden access to what has been a life-saving agent for women with HER2+ breast cancer.”
Going with the Flow: Liquid Biopsies
Tumors are dynamic, always growing, and ever changing, and the driver of change is oftentimes a therapeutic. Especially in an era of targeted agents, it is critical to know if, and when a tumor has lost the target. However, on-treatment biopsies are invasive by definition and impractical by default – patients can only be expected to tolerate so much disruption. Liquid biopsies can change that.
“It’s been known for decades that advanced cancers shed DNA into the blood stream,” said Philip Mack, PhD, University of California Davis Comprehensive Cancer Center. “This circulating tumor DNA (ctDNA) harbors all the same cancer associated mutations of potential use in (tissue-based) diagnostics.”
In validating this blood-based approach, Mack spearheaded the largest liquid biopsy study performed to date: Blood samples from over 15,000 cancer patients with multiple tumor types were analyzed.
Using a next-generation sequencing platform, Guardant360, Mack and colleagues assessed the accuracy and clinical utility of Guardant360 for the detection of 70 different cancer-related genes – even those expressed at very low frequencies in the blood. Accuracy was gauged by comparing liquid biopsy results with publically available population scale tumor sequencing projects (i.e. The Cancer Genome Atlas).
Results of this analysis showed that ctDNA mutation patterns were, “Highly consistent with distribution in tissue,” said Mack. “The key difference is that ctDNA analysis additionally detected resistance alterations, such as (EGFR) T790M, which were not present at the time of initial tissue biopsy.” And further, these alterations were reliably detected even when present at very low levels in the plasma.
Across all 70 genes of interest, ctDNA analysis correlated with tissue biopsy results from 94%-100% of the time (Figure 2).
ctDNA analysis additionally detected resistance alterations, such as T790M, not present at the time of initial tissue biopsy.
Of note, ctDNA represents a sampling from metastatic lesions as well as the primary tumor – something that tissue biopsies at the time of diagnosis don’t do. “Probably the biggest role for plasma analysis is down the road for patients that are progressing on therapies, to monitor that progression,” said Mack.
Awareness of such mutations has obvious implications for patient management: In this study, the identification of ctDNA-identified cancer mutations/biomarkers suggested the use of an FDA-approved targeted agent in 49% of patient cases.
Sumanta Kumar Pal, MD, City of Hope Medical Center was bullish on the study results.
“Tests such as this provide a useful alternative to tissue-based testing, acknowledging that fact that many of our patients in the clinic have tumors that are challenging to access (bone, brain, etc.). Further, “this study demonstrated not only feasibility of a test in detecting alterations but that the results are often potentially actionable, with half of patients having a result that can be linked to therapy.”
The advent of the so-called basket trial is in recognition of the fact that targeted cancer drugs are not targeted to a tissue type (lung, brain, etc.) but to a molecular abnormality that may be present in any number of tissues. For instance, while the HER2-targeting drug Herceptin is indicated for the treatment of HER2+ breast, and gastric cancers, HER2 may also be found in cancers of the lung, colon, bladder and several other tumor types—albeit to a far lesser extent, with rates as low as 2%—too low to justify a clinical trial to explore that potential new indication.
So how can the utility of targeted agents be expanded beyond current indications in these low-incidence, target-specific cancers? With a basket trial.
“The efficacy of approved therapies is often unknown in non-indicated tumor types harboring these molecular alterations,” said John Hainsworth, MD, Sarah Cannon Research Institute, and principal investigator for the MyPathway trial. “MyPathway was designed to evaluate agents targeting the HER2, BRAF, hedgehog or EGFR pathways in non-indicated tumors with those relevant molecular alterations.”
The trial treated relapsed/refractory cancer patients who had a given molecular abnormality – regardless of tumor type – with the drug indicated for that abnormality: For HER2, trastuzumab/pertuzumab; for BRAF, vemurafenib (indicated for melanoma); for EGFR, erlotinib (pancreas, NSCLC); for Hedgehog, vismodegib (basel skin carcinoma).
Tumor types included in MyPathway were a broad survey: colorectal, bladder, biliary, NSCLC, head and neck, and six others (N=129).
Patients identified with a given oncogenic driver were assigned the related agent.
Results for MyPathway showed that for the entire cohort one patient achieved a complete response, with 28 others achieving a partial response, and 40 patients achieving a status of stable disease. Best responses were seen for bladder, biliary, lung, and in particular, colorectal cancer. “That’s a very strong signal there,” said Hainsworth.
In all, activity was observed in patients with 12 different tumor types outside of current indications for these targeted agents.
Vivek Subbiah, MD, Department of Investigational Therapeutics at M.D. Anderson Cancer Center is very familiar with basket trial design, having been the PI on one of the first such studies ever performed (NEJM. 2015).
“Basket studies are seeking first tips, signal seeking first tips,” said Subbiah. His study that looked at treating the BRAF V600 mutation with vemurafenib in multiple tumor types produced some very interesting tips.
“Lung cancer was a surprise,” said Subbiah. “Lung cancer patients (with V600 mutation) had a very nice response to the drug,” whereas colorectal cancer patients did not respond to single agent vemurafenib, even though they had the target.
Basket trials are hypothesis generating to be sure, and may well suggest a pathway to a new, potentially lucrative indication.
Knowing the Score
Cancer immunotherapy is here to stay, and that means augmenting standard clinical practice.
“In the era of immunotherapy it is becoming essential to start classifying cancer patients based on immune parameters,” said Jerome Galon, Laboratory of Integrative Cancer Immunology, INSERM.
It’s been known for some time that the type, number, and location of immune cells is a major predictor of clinical outcome, but only recently, since the advent of immunotherapeutics has that information been seen as truly actionable—if not now critical—to deciding an appropriate treatment course.
“Today there is not a single immune characteristic that is taken into account in a cancer patient,” Galon observed. “We don’t know anything about the immune system of cancer patients because there is not a single standardized assay.”
Working toward that end, Galon and colleagues looked to validate such an assay, Immunoscore, in a multicenter, international population of stage I/II/III colon cancer patients (N=3,855).
To establish the various predictive cut points for the assay, information regarding the immune cell status from the first 1,336 patients, compared with known course of the patient’s disease was used as a training set. All patients had a median follow-up of 5.9 years.
An Immunoscore consists of cell density for CD3+ and CD8+ cells located both at the center of the tumor mass and at the margin. Cut points for low, intermediate, and high Immunoscores (IM) were determined and applied to data from the remainder of the patient cohort, and then compared with the time to disease recurrence (TTR) in these patients.
Results showed that TTR was shorter for patients with low IM vs. high IM (P<0.006). Immunoscores also correlated nicely with overall survival (OS), Galon said.
“The results of this international consortium may result in the implementation of the Immunoscore as a new component for the classification of cancer, designated TNM-Immune,” said Galon. “This will represent the first standardized immune-based assay for the classification of cancer.”
Galon further suggested that it is a logical progression of thought for patients with a high Immunoscore to be treated with immunomodulatory agents.
Two New Agents
Speaking of immunomodulatory agents, atezolizumab, an antibody drug that blocks the activity of the programmed cell death-ligand (PD-L1), a protein that suppresses the immune response, made news by providing significant benefit to patients with advanced bladder cancer.
Already approved in May of this year for second-line treatment of urothelial carcinoma, the most common type of bladder cancer, investigators tested the agent in a population of bladder cancer patients who are unable to tolerate the standard-of-care, front-line drug, cisplatin (N=119).
“We focused on a population of patients who have metastatic bladder cancer who are frankly often too sick to receive aggressive chemo, in particular with cisplatin which is the only treatment that has been shown to increase survival in this disease,” said lead author of the single arm, IMvigor210 study, Arjun Balar, MD, of the New York University Langone Medical Center.
“For cisplatin ineligible patients with (this disease) there are no other survival-prolonging therapies,” leaving the patient the only option of best supportive care.
Results from IMvigor may change that, given the striking results for atezolizumab in this patient cohort: The overall response rate was 24% (7% of patients achieved a complete response) and the median OS was 14.8 months, a figure that compares favorably to historical controls (Figure 3).
Unlike drugs that target the PD-1 receptor, whose activity may or may not be tethered to expression levels of PD-1 (it depends who you ask) the response to atezolizumab tracked closely with levels of PD-L1 expression—the higher the expression the greater the response. Of those patients who achieved a response, 75% of those responses are ongoing at a median follow-up of 14.4 months.
“I think these data represent a compelling argument for atezolizumab as a potential new standard of care in the first line setting,” said Balar, with studies specifically looking at this agent in the front line setting coming soon.
Small Cell Lung Cancer
“Small cell lung cancer (SCLC) is a terrible disease and things have not changed in 40 years in terms of clinically meaningful endpoints for patients with later stage disease,” said Charles Rudin, MD, PhD, of the Memorial Sloan Kettering Cancer Center.
Indeed, from the time of diagnosis the median survival rate for SCLC patients is less than 10 months, and for recurrent disease there is only one approved drug, topotecan, “So there’s a lot of room for improvement here.”
Improvement may come from a drug called rovalupituzumab tesirine, which targets delta-like protein 3 (DLL3), a molecule expressed in over 80% of SCLC tumors, but not at all in healthy tissue.
In a small, first-in-human study of 74 relapsed/refractory patients with advanced SCLC response rates were as high as 38%, depending on the level of DLL3 expression (as in the case with PD-L1, higher expression = better response).
Further, the OS rate at one year was 18% for the entire cohort, and 32% for those patients with high levels of DLL3 expression; this represents a doubling or better for OS as compared with historical controls.
Granted, this is an early investigation into the activity of rovalupituzumab tesirine, but, “these are impressive numbers,” said Rudin, “Yes, it’s a small trial—but this looks promising to us.”
So promising that a study using the drug in the first-line setting is already in the works.