ASCO 2017-IMMUNOTHERAPY FLAT DOSE TOO COSTLY

The new immunotherapies are expensive drugs, but a new study suggests that they are costing more than they need to because of flat, rather than personalized, dosing.

Using a flat dose of pembrolizumab (Keytruda, Merck &Co) for first-line treatment of lung cancer instead of personalizing the dose to the patient's body weight results in an excess of 25% in drug dose, and hence a 25% increase in drug costs, say a group of oncologists. They calculate that this excess could mount up to nearly $1 billion annually in the United States.

This estimate was reported in a poster presentation (abstract 9013) at the recent American Society of Clinical Oncology (ASCO) 2017 Annual Meeting, and the study was simultaneously published online June 3 in the Journal of the National Cancer Institute.

Daniel A. Goldstein, MD, from the Davidoff Cancer Center, Rabin Medical Center, Petah Tikva, Israel, and colleagues reported that patients with non-small cell lung cancer (NSCLC) receiving first-line pembrolizumab at the flat dose of 200 mg every 3 weeks (Q3W) results in an excess of 25% in drug dose, and therefore a 25% increase in drug cost, compared with the originally approved personalized dosing of 2 mg/kg Q3W.

"Using personalized dosing may reduce the budget impact by approximately 25%, thus increasing access and improving health outcomes," the authors concluded.

Medical oncologist and a critic of drug pricing, Leonard B. Saltz, MD, professor and chair of pharmacy and therapeutics committee at the Memorial Sloan Kettering Cancer Center (MSKCC), New York City,  agreed.

"Pembrolizumab is only sold in the United States in 100-mg vials. An average lung cancer patient weighing 75 kg would require 150 mg of pembrolizumab based on weight. Because the 50-mg vials are only marketed outside of the US, 50 mg of drug, at over $48 per milligram, must be wasted for each 75-kg American patient," Dr Saltz told Medscape Medical News. "If we don't share

vials between patients, that 50 mg goes in the trash, but we pay for it anyway. The flat dosing of 200 mg just means that we are dumping the additional 50 mg into the patient for no additional benefit," he added.

Pharmacoeconomic Analysis 

Dr Goldstein and colleagues noted that pembrolizumab was initially approved at a dose of 2 mg/kg Q3W in patients with metastatic melanoma. The Keynote lung cancer studies (Keynote 001, 010) showed that response rates and toxicity were similar for pembrolizumab 2 mg/kg Q3W, 10 mg/kg Q3W, or 10 mg/kg every 2 weeks. Finally, an analysis of samples from patients in Keynote studies 001, 002, 006, 010, and 024 showed that the pharmacokinetics of pembrolizumab are consistent across weight-based and fixed-dose regimens and showed a flat exposure–response relationship.

When pembrolizumab was approved for the first-line treatment of patients with metastatic NSCLC overexpressing programmed cell death ligand 1, the approval was for the flat dose of 200 mg Q3W.

In a budget impact analysis from the US societal perspective, Dr Goldstein and colleagues compared costs associated with fixed dosing and personalized dosing.

For their analysis, the authors included 19,601 patients who were likely candidates for first-line pembrolizumab. On the basis of published data, the number of pembrolizumab cycles was 17.9; patient weight was 76 kg according to the average weight of a patient with lung cancer at Emory University in Atlanta, Georgia. Drug cost was based on Medicare pricing of $48.45/mg.

The authors' base-case model demonstrated that, in the United States, the total annual cost of pembrolizumab is $3,440,127,429 with fixed dosing and $2,614,496,846 with personalized dosing. Use of personalized dosing could therefore result in an annual saving of $0.825 billion.

Implications of Fixed Dosing, Large Vial Size 

Dr Goldstein indicated that potential barriers to implementing personalized dosing includes prescriber agreement, ineffective vial sharing, oversized vials, and the current policy of payment for wastage being a disincentive for reducing wastage.

"There is no incentive for physicians at community centers to use less drug since they get paid for the amount of drug they use," Dr Goldstein said. "Flat dosing means that infusion centers do not have to share vials between patients," he continued. "Flat dosing increases the revenue for a drug company, with no additional clinical benefit," Dr Saltz added.

"The most appropriate first step to avoid waste is to change or add personalized dosing to the FDA [US Food and Drug Administration] label. If the FDA changes the label, clinicians and payers will be more comfortable prescribing and reimbursing personalized dosing," Dr Goldstein and colleagues write in their discussion.

"This analysis is picking up on an issue we previously addressed — that drug companies are increasing the amount of drug they are selling by packaging it in large vial sizes so that a considerable amount is wasted, but paid for, with each dose," Dr Saltz said. "In case of flat dosing, the waste is infused in the patient for no benefit; in case of weight-based dosing, it is dumped into the sink," he added.

Dr Saltz was referring to the article published in March 2016 (BMJ. 2016;352:i788), which was widely publicized and reported by Medscape Medical News at the time. This analysis of 20 cancer drugs showed that as a result of large vial sizes, leftover cancer drugs in vials cost $3 billion annually.

In that BMJ paper, outspoken critic of drug pricing Peter Bach, MD, professor and director of the Center for Health Policy and Outcomes at MSKCC, along with his colleagues (Dr Saltz included), argued that under the buy-and-bill system, practices and hospitals buy single-dose vials of drugs and then bill insurers or patients when they are used. "The bill includes a percentage-based mark-up which can vary widely, but even low percentages can equate to large amounts of money given that many of the drugs cost thousands of dollars per vial," they wrote.

Dr Bach and colleagues urged drug companies to provide drugs in multiple vial sizes. They cite the example of bendamustine (Treanda, Teva Oncology), which is available in a broad array of single-dose vials (25 mg, 45 mg, 100 mg, and 180 mg) that can be combined to reach the precise dose of 100 mg/m². The availability of these vial sizes allows only 1% of the drug to be wasted.

Pembrolizumab was initially packaged in 50-mg and 100-mg vials in the United States. However, long before the approval of flat dosing, the 50-mg vial was discontinued and Merck marketed only the 100-mg vial.

Medscape Medical News reached out to Merck about pembrolizumab vial sizes because both vial sizes are indicated as available in the May 2017 KEYTRUDA Prescribing Information.

"In the US, when KEYTRUDA received initial approval (Sept 2014), we introduced a 50 mg lyophilized dose which required reconstitution and then quickly transitioned to the 100 mg liquid dose as planned, which was preferred by providers because it does not require reconstitution. KEYTRUDA currently is only available in the US in the ready-to-use, liquid formulation in the 100 mg vial size," Merck told Medscape Medical News.

However, both vials continue to be available in Europe, where the concept of a value framework exists. For example, the United Kingdom's National Institute for Health and Care Excellence provides a stamp of approval for drugs based on a cost-effective analysis, which needs to consider the amount of drug wasted, Dr Goldstein explained.

The FDA approves drugs on the basis of efficacy and safety. "If a drug is FDA-approved, Medicare has to provide the drug at the price set by the drug company. Medicare has no power to negotiate the price," Dr Goldstein said.

"In the US, there needs to be a step where someone assesses the value of a drug after it is approved and before Medicare agrees to pay for the drug," he explained.

"The FDA has no legal mandate, and appears to have no desire, to take on the issue of drug costs. As it stands now, no one can tell the drug companies how to package the drug," Dr Saltz told Medscape Medical News. "Once the FDA approves a drug for an indication, Medicare must pay for it at any price that the drug company chooses to set. If Medicare pays for the drug, others will," he added. "If drugs are priced higher, copays and Medicare payments increase," he said.

"Personalized dosing has the potential to decrease costs while maintaining efficacy," Dr Goldstein and colleagues write. "While personalized dosing should be encouraged, its rapid adoption and potential cost savings may be dependent on cooperation between the FDA, physicians, patients, payers, and manufacturers," they add.

Financial Toxicity, a Need for Transparency 

Dr Goldstein explained that although the 200-mg flat dose of pembrolizumab is not associated with additional toxicity, it is associated with financial toxicity for the patient, who has a higher out-of-pocket cost when he or she receives a higher dose than is warranted. Financial toxicity can manifest as patients cutting back on daily needs, selling their homes, not adhering to treatment, or declaring personal bankruptcy to pay for cancer care.

"Understanding cost matters," Dr Saltz told Medscape Medical News. He advocates the need for transparency with respect to any drug costs. It is important to know the price of the drugs one is prescribing, and the price should factor into decision-making. "We should be discussing the cost of drugs we prescribe as openlyDr Saltz argued that it is incorrect to assume that physicians should be concerned only with immediate out-of-pocket costs and that societal costs are not relevant. "Societal costs are ultimately distributed across the population. We all eventually bear the burden of increased drug costs in terms of increased insurance premiums," he said.

He commented  further that physicians openly talk with their patients about the intimate and personal details of their lives. They ask patients about bowel function, bladder function, sexual function, depression, alcohol and illicit drug use, and other details that are outside of normal social discourse. "There was a time decades ago when we as a profession did not talk to patients about the presence of cancer or about prognosis, but all this has changed," he said. "In this context, it is inconsistent that any conversation about costs is taboo," he said.

Dr Saltz also noted that it is important and routine for people to make informed decisions about any goods and services they might choose to purchase, and to consider alternatives and value. "Healthcare should be no different," he added.

The discussion of value in healthcare is too often confused by how the word is used. For a meaningful discussion of "value," the word should be used as a noun. The value of something can be quantitated in terms of high- and low-value care. In this context, the value of a drug can be seen as a ratio of its objective positives and negatives, with high price being one of the negatives, Dr Saltz explained.

as we discuss its risks, benefits, and other toxicities," he said.

"For any drug with fixed benefits and adverse effects, the higher price means lower value and vice versa," he said. "Drug cost is simply one component of value," he pointed out.

However, using "value" as a verb obfuscates the issue, he explained. When we say that patients "value" the benefits of a drug, we can no longer quantify it. "The subjective nature of this sort of use of the word 'value' resists a correlation with price," Dr Saltz said. "This focus of 'value' as a verb distracts us from a meaningful discussion of what is high-value care, and more importantly, what is not," he added.

"There needs to be a transparency about true risks, benefits, and alternatives of treatment strategies being considered," Dr Saltz said. "As much as it would be wrong to exclude a consideration of physical toxicities of a drug, it would be equally wrong to exclude consideration of price or financial toxicity," he said.

Dr Saltz contends that there will come a day when academic publications will include the drug price, and future cost issues will be incorporated into study design, as well as review and approval processes. "When a trial is designed that increases the length of treatment or increases drug dose to higher than the standard, investigators must think about what the costs of those changes will be and must take them into consideration," he said.

Dr Goldstein has disclosed no relevant financial relationships. Dr Saltz reports grants from Taiho Pharmaceuticals. 

American Society of Clinical Oncology (ASCO) 2017 Annual Meeting. Abstract 9013. Presented June 3, 2017.

ASCO 2017-PATIENT GENERATED DATA IMPROVED SURVIVAL

Over the past several years, there has been speculation as to whether digital tracking by patients via their mobile devices could change clinical outcomes. In fact, when I read a paper presented at the American Society of Clinical Oncology (ASCO) annual meeting in June 2016, I thought it might be too good to be true.^[1]

In that study, also covered in the Wall Street Journal,^[2] Fabrice Denis, MD, PhD, and colleagues at the Institut Inter-régional de Cancérologie Jean Bernard in Le Mans, France, randomly assigned 133 patients with advanced lung cancer to either receive standard care or use a smartphone app (MoovCare™; Sivan Innovation) tracking 12 symptoms. For the patient self-tracking strategy, an algorithm alerted the doctor.

In this small trial, there was a significant improvement in survival of a median of 7 months for the intervention. The 1-year survival increased from 49% in the standard-care group to 75% in the self-tracking arm. Furthermore, 50% fewer scans were performed, and it took only 15 minutes per week for oncologists to follow the 60 patients in the latter group. The trial was stopped prematurely because of these exceptional results.

I never expected that the validation for patient-generated data in digital medicine would be manifest in oncology with survival as the primary outcome.

My interpretation was that this was very interesting, but the glowing outcomes were unlikely to be replicated in a larger trial. And I never expected that the validation for patient-generated data in digital medicine would be manifest in oncology with survival as the primary outcome. But I was wrong on all counts.

In a large trial from Memorial Sloan Kettering that was presented at the recent ASCO 2017 meeting and simultaneously published in JAMA,^[3] 766 patients with diverse metastatic cancers underwent a 2:1 randomization of patient-tracking of 12 symptoms or usual care. As with the French trial, an email alert was triggered by significant worsening of symptoms, but it initially went to a nurse. The patient-reported outcomes group had a 5.2-month median survival improvement (31.2 vs 26.0 months). There were also improvements in the secondary outcomes of 1-year quality-adjusted survival and tolerance of a longer duration of chemotherapy.

Why is this trial so important? First, it is not a stand-alone result but the replication of one with concordant results, now with a fivefold increase in sample size and long-term follow-up (instead of premature discontinuation).

Second, the improvement in survival of > 5 months compares quite favorably with the median survival improvement in major oncology drug trials, such as those seen with the exciting results of immunotherapy. Many of these trials have shown a 3-month median survival improvement for new drugs, leading to commercial approval and costing at least $100,000 per treatment.

But unlike the drug trials, there was little promotion or coverage. The only major newspaper that reported on the trial was the Washington Post.^[4] This might have been anticipated; there is no sponsor. The ASCO meeting is characterized by the reporting of hundreds of drug trials, but it turned out that a digital medicine trial had striking results that largely outperformed most of the expensive and heavily promoted biopharma interventions. The digital tracking strategy costs very little: an app, an algorithm, and a strategy for clinician response. But that represents a de minimis amount considering the outcome of improved survival—or the new drug cost of $100,000 per patient.

Most important, this trial emphasizes the role of engaging patients in tracking their symptoms and of generating their own data. Patients have long been neglected for having a more active role in their care, when in fact, given the appropriate tools, they represent true "blockbuster" potential for improving their outcomes.^[5]

ASCO 2017-CDK 4/6 INHIBITORS IN BREAST CANCER

There have been a number of interesting breast cancer presentations at the American Society of Clinical Oncology (ASCO) annual meeting this year, including several reports of CDK 4/6 inhibitors in patients with hormone receptor–positive (HR+)/HER2-negative (HER2-) metastatic breast cancer (MBC) presented on Saturday, June 3.

To reiterate the landscape up until this point, there are currently two FDA-approved CDK 4/6 inhibitors in HR+/HER2- MBC: palbociclib and ribociclib. Prior to ASCO 2017, we knew that palbociclib received conditional approval in February 2015, based on a randomized, open-label phase 2 trial with front-line letrozole with or without palbociclib (PALOMA-1 trial). Subsequent randomized, placebo-controlled, phase 3 trials have demonstrated a significant benefit in progression-free survival (PFS) with front-line letrozole (PALOMA-2 trial) or with fulvestrant after prior progression on anti-estrogen therapy (PALOMA-3 trial). More recently, ribociclib was approved in March 2017 on the basis of the randomized phase 3 MONALEESA-2 trial of front-line letrozole with or without ribociclib.

So, what did we learn in Saturday's session at ASCO?

Updates on Abemaciclib

We saw the first presentation of a randomized, phase 3 abemaciclib trial.^[1]MONARCH 2 was a double-blind, placebo-controlled trial of fulvestrant with or without abemaciclib in women with HR+/HER2- MBC who had progressed while receiving endocrine therapy. (It was also a rapid communication published in Journal of Clinical Oncology .)Endocrine-resistant disease included those who relapsed on neoadjuvant or adjuvant therapy within 1 year or who progressed on first-line endocrine therapy. Median follow-up was 19.5 months. There was a significant improvement in PFS—the primary endpoint—from 9.3 months with fulvestrant alone to 16.4 months with the combination (hazard ratio, 0.55). The forest plot showed all subgroups favoring the combination.

Of note, the side-effect profile of abemaciclib is different from that of palbociclib and ribociclib. Neutropenia rates are relatively lower (total 46%, grade 3: 23.6%). Gastrointestinal side effects are higher, including diarrhea (total 86.4%, grade 3: 13.4%). Other notable aspects of this presentation: This is a different population from PALOMA-3's, including that no prior chemotherapy was allowed in MONARCH-2, potentially helping to explain the numerically longer PFS with the control arm in MONARCH-2 than previously observed in other studies, such as PALOMA-3. Also, there was a required dose reduction for all patients in the trial from 200 mg twice daily to 150 mg twice daily, which was better tolerated.

These data, plus the MONARCH-1 data presented last year at ASCO and recently published, are compelling. In addition, MONARCH-3, another phase 3 trial with front-line letrozole with or without abemaciclib, has been reported in press releases as positive for the combination. Stay tuned for these data in the future, as well as for potential approval for abemaciclib in MBC.

Is There an Overall Survival Advantage for CDK 4/6 Inhibitors?

Maybe. The updated results of PALOMA-1^[2] were presented today. With a median follow-up of 65 months, there was a numerical trend toward improvement with the combination (37.5 months vs 34.5 months)but it was not statistically significant. However, this study was not powered to look at overall survival. We will await future updates of studies like PALOMA-2 and MONALEESA-2 to answer this question. One other notable finding: the median time to chemotherapy initiation of 26.7 versus 17.7 months with the combination and single-agent letrozole, respectively (HR, 0.66).

Is There a Benefit of CDK 4/6 Inhibition in a Moderately Pretreated Population?

Yes. The TREnd trial was a two-stage, noncomparative, randomized, phase 2 trial.^[3]I Patients with one or two prior lines of endocrine therapy were randomly assigned to palbociclib or palbociclib plus the endocrine therapy that they recently received. Both arms demonstrated activity as demonstrated by a predefined clinical benefit rate of > 40%. In an exploratory analysis, the duration of clinical benefit (11.5 months vs 6 months) and PFS (10.8 months vs 6.5 months) were numerically higher in those who received endocrine therapy plus palbociclib as opposed to palbociclib alone.

As you can see, that's a lot of data. However, there are important questions to answer in the future, including:

• Can we identify which patients will benefit with endocrine therapy alone, given the associated cost and potential side effects of CDK 4/6 inhibitors?
• Is there a role for sequencing CDK 4/6 inhibitors in the metastatic setting?
• Will we identify markers of response/resistance to CDK 4/6 inhibitors, and markers that will demonstrate benefit with one agent over another?
• What will be the benefit with these drugs in the adjuvant setting?

Studies attempting to answer these questions are ongoing. But hearing the session at ASCO 2017, one cannot argue that this drug class remains an important step forward in the treatment of HR+/HER2- MBC.

ASCO 2017-IBRUTINIB+IMMUNOTHERAPY FOR CLL

Combining the kinase inhibitor ibrutinib (Imbruvica) with an investigational personalized cellular therapy known as CTL119 can lead to complete remission in patients with high-risk chronic lymphocytic leukemia (CLL), according to new research from the Perelman School of Medicine at the University of Pennsylvania and Penn's Abramson Cancer Center. The team presented results from its pilot study of this combination therapy at the 2017 ASCO Annual Meeting(Abstract 7509).

The research team is led by Carl June, MD, the Richard W. Vague Professor in Immunotherapy in the Department of Pathology and Laboratory Medicine and Director of Translational Research, along with David Porter, MD, the Jodi Fisher Horowitz Professor in Leukemia Care Excellence and Director of Blood and Marrow Transplantation, both at the Abramson Cancer Center. The data were presented by the study's first author, Saar Gill, MD, PhD, Assistant Professor of Hematology-Oncology.

“Combining ibrutinib with the CTL119 therapy achieved very powerful results for these patients, and with limited toxicity,” Dr. Gill said. “This newer, coupled approach gives us hope that personalized cell therapies could be an important option for high-risk CLL patients on these types of drugs.”

CTL119 manufacturing begins with a patient's own T cells, some of which are removed and then reprogrammed in Penn's Clinical Cell and Vaccine Production Facility with a gene transfer technique designed to teach the T cells to target and kill tumor cells. The engineered cells contain an antibody-like protein known as a chimeric antigen receptor (CAR), which is designed to bind to CD19 protein on the surface of cancerous B cells. The modified “hunter” cells are then infused back into the patient's body, where they multiply and are believed to attack the cancer cells.

Current Findings

The team presented data on the first 10 patients in the trial, each of whom had been taking ibrutinib for at least 6 months but had not achieved a complete remission. They were then infused with their own engineered “hunter” T cells. Eight of nine patients who are evaluable for response had no evidence of disease in their bone marrow at 3 months, and all remain in remission after a median follow-up of 6 months, with a range of 0.5 to 9 months. One patient was found to have a partial response in his marrow.

Patients in the trial had been on ibrutinib for a minimum of 6 months and had not achieved a complete response when they received an infusion of engineered cells split over 3 consecutive days. All patients had abnormalities of TP53 or ATM—two mutations associated with high-risk disease—and two patients had increasing BTK C481S clones, also a high-risk marker.

Earlier Studies

The new data builds off several preclinical studies supporting the use of ibrutinib with CAR therapy. In March 2016, Penn researchers published a study in Blood that showed long-term ibrutinib treatment reverses the dysfunction of T cells in CLL and that combining CAR therapy with ibrutinib enhanced engineered T-cell proliferation in mice.

CAR therapy alone has led to complete remissions and responses in some CLL patients, but not all patients respond—findings that led the Penn team to seek combination therapies that might enhance efficacy of the therapy. In 2015, Penn Medicine researchers reported in Science Translational Medicine an overall response rate of 57% in CLL patients treated with CAR therapy and a complete remission rate of 29%.

Ibrutinib is a well-tolerated, oral drug that improves symptoms and survival in high-risk CLL patients, but it is not curative and requires continuous treatment for life. It rarely induces complete remissions by itself. The first year on the drug may provide an optimal window for collecting T cells from patients and subsequently administering a potentially curative T-cell therapy, the authors said.

“These patients had a lower disease burden and were treated earlier in the course of their disease, which distinguishes it from other studies,” Dr. Gill said. “One of the challenges in treating CLL patients with personalized cellular therapy is not having healthy enough cells to manufacture. The results suggest that ibrutinib restored T-cell activity in the patients.”

Toxicities

All 10 patients who received the CTL119 cells experienced mild cytokine-release syndrome—a potentially lethal toxicity that can include varying degrees of flu-like symptoms and temporary neurologic symptoms, and in more severe cases, low blood pressure and breathing difficulties—within a few days after receiving their infusions. However, none required treatment with tocilizumab (Actemra), an immunosuppressant drug that blocks the effects of the inflammatory cytokine interleukin-6. All patients recovered from their cytokine-release syndrome. One patient in the study developed tumor lysis syndrome and recovered.

Longer follow-up will reveal the durability of these results, the authors said, and may support the evaluation of a first-line approach with ibrutinib and CAR therapy in an effort to remove the need for chronic therapy.

ASCO 2017-ANTI-HYALURONIC TREATMENT FOR PANCREATIC CANCER

By adding an experimental drug to a standard chemotherapy regimen, a subset of patients with metastatic pancreatic cancer had a significantly longer period before the cancer progressed as compared with those who received the standard treatment, according to a phase II clinical trial led by an investigator at Fred Hutchinson Cancer Research Center.

The randomized, controlled trial found that when the experimental therapy was given to participants whose tumors had high amounts of the drug's target molecule, they had 4 months more of progression-free survival than participants in the control group who only received standard chemotherapy.

First author Sunil Hingorani, MD, PhD, presented the study results at the 2017 ASCO Annual Meeting (Abstract 4008).

Dr. Hingorani said that the results reassure him that it was the right move to advance the drug, called pegylated recombinant human hyaluronidase (PEGPH20), into the worldwide phase III trial that opened last year.

About PEGPH20 and Hyaluronic Acid

Dr. Hingorani's earlier research led him to the drug because he believed it could address a challenge posed by many pancreatic cancers: the tumors have very high internal pressures that can collapse local blood vessels and prevent cancer-killing drugs from getting in. PEGPH20 reduces those pressures so chemotherapies circulating in the blood can penetrate tumors.

The experimental drug, which was created from the blueprint of a naturally occurring enzyme, breaks down a molecule called hyaluronic acid that is produced in bulk by many pancreatic cancers.

Hyaluronic acid, or HA, is naturally found in the human body; it readily binds water to create a gel fluid. But in pancreatic tumors, as the gel fluid builds up, it raises the tumor's internal pressure, squeezing local blood vessels shut. Patients whose tumors have high amounts of HA also tend to have a poor prognosis.

HALO 202 Background

Dr. Hingorani and his team first conducted studies in mice that showed how PEGPH20, in combination with chemotherapy, permanently reduced the amount of pressure-boosting HA inside the animals’ tumors. It caused the tumors to shrink and increased the mice's survival time.

In the phase II trial (HALO-109-202) patients with late-stage pancreatic cancer were randomly assigned to receive a standard-of-care, first-line combination chemotherapy either with or without PEGPH20. When the results of all 234 evaluable patients on HALO 202 were grouped together, the apparent benefit of PEGPH20 was small—a matter of just a couple extra weeks of progression-free survival.

“If this was all the potential that this strategy represented, I wouldn't pursue this [research further],” Dr. Hingorani said. “That's not enough for me.”

But a stark difference emerged when the results were divided up by how much of the drug's target, HA, patients' tumors contained: in the subset of 80 patients whose tumors had high levels of HA, adding PEGPH20 to chemotherapy resulted in an average of 9.2 months before disease progression; with chemotherapy alone, this timespan was 5.2 months.

Dr. Hingorani also reported that the unexpected, elevated risk of blood clots associated with PEGPH20—which resulted in a temporary halt of the trial in 2014—equalized between the patients receiving PEGPH20 and those in the control group, and dropped overall, after the study was restarted, due to the addition of a blood thinner to all patients' regimens.

“These are the real take-home messages to me, namely, the progression-free survival in target-rich [high-HA] patients and the ability to give the enzyme safely,” Dr. Hingorani said.

Treatment-related adverse events for trial participants included peripheral edema (63% of those receiving PEGPH20 vs 26% for the control group), muscle spasms (56% vs 3%), neutropenia (34% vs 19%), and myalgia (26% vs 7%).

Phase III Trial

Because the phase II trial results suggest that the benefit of the experimental drug is restricted to the patients with high levels of HA in their tumors, only patients with such tumors qualify for the new phase III trial. Additionally, the phase III trial is designed to offer a more stringent test of the benefits of the new drug than its predecessor—the trial's goal is to determine whether PEGPH20 actually increases participants' lifespans, not just their time to disease progression. The investigators' exploratory analysis of the phase II trial data suggested that the experimental drug boosted the lifespans of patients with high-HA tumors to an average of nearly a year after diagnosis, which, if shown definitively in the phase III trial, could be a new benchmark for this cancer, Dr. Hingorani said.

Dr. Hingorani launched the phase III trial before handing off its leadership to Margaret Tempero, MD, of the University of California, San Francisco, and Eric Van Cutsem, MD, PhD, at the University of Leuven in Belgium. 

ASCO 2017-LIQUID BIOPSIES FOR SCREENING

While the idea of using a drop of blood to detect an occult cancer is still elusive, use of so-called liquid biopsies for cancer screening is at least a little closer to reality with the development of a new high-intensity genomic sequencing approach outlined here at the American Society of Clinical Oncology (ASCO) 2017 Annual Meeting.

A new test (under development by Grail Inc) highlighted in the ASCO press program uses high-intensity sequencing on circulating tumor DNA (ctDNA). It offers "an unprecedented combination of breadth and depth," surveying 508 genes and yielding about 100 times more data than other currently used approaches, said investigator Pedram Razavi, MD, PhD, from Memorial Sloan Kettering Cancer Center in New York City.

Dr Razavi presented results from a concordance study in 124 patients with metastatic cancer, which showed that the new test picked up 89% of genetic changes that had been identified with regular tumor biopsy and 76% of "actionable" mutations that could be treated with targeted therapy.

"Our findings show that high-intensity circulating tumor DNA sequencing is possible and may provide invaluable information for clinical decision-making, potentially without any need for tumor tissue samples," Dr Razavi commented.

"The ultimate goal would be detection of cancer in early, treatable stages," he said during a press conference here.

Medscape Medical News approached Philip Mack, PhD, professor and director of molecular pharmacology at the University of California Davis Comprehensive Cancer Center, for comment. He was not involved in this study, but at last year's ASCO meeting, he presented a study about the Guardant360 liquid biopsy assay.

An advantage of this approach is the breadth of potential data obtainable from a simple blood draw," Dr Mack told Medscape Medical News.

"Simultaneous analysis of over 500 cancer-associated genes from ctDNA could provide nuanced information on companion mutations and tumor evolution, which could help fuel therapeutic development in cancer research."

However, Dr Mack added: "Whether this or other approaches can be used for early detection of cancer remains to be seen."

ASCO experts also acknowledged that this approach is still far from reality.

 While the data are "an important step forward," according to ASCO expert John Heymach, MD, PhD, "we are a long way from using liquid biopsies for detecting cancers," cautioned Sumanta Kumar Pal, MD, another ASCO expert at the press conference.

"This field is moving very rapidly and ctDNA assays have already entered the clinic, but there are still many uncertainties, so this is always a situation where one can run into trouble," warned Maximilian Diehn, MD, PhD, from Stanford Cancer Institute, during a liquid biopsy educational session at the meeting.

Although a few ctDNA assays are available for clinical use in the United States (cobas, Roche) and Europe (therascreen, Qiagen; cobas; OncoBEAM, Sysmex), with many potential applications, screening for cancer detection is still not one of them, he said."Screening is one of the most exciting and high-profile applications, but we are not there yet, and screening should not be done with ctDNA assays outside of clinical trials," he said.

Dr Diehn serves on a joint ASCO and College of American Pathologists group working on a joint position statement on ctDNA-based liquid biopsies.

"The goal of this to ensure that clinical use of ctDNA assays is supported by adequate evidence, including analytical validity, clinical validity, and utility," he said, adding that the final draft should be available by the end of this year.

"It will not be a formal practice guideline because the data are just not there yet — that will come later," he said.

This study was funded in part by GRAIL Inc. Dr. Razavi disclosed Research Funding (Institutional) from GRAIL. Dr Heymach disclosed stock and other ownership interests with Bio-Tree, Cardinal Spine; consulting or advisory roles with AbbVie, ARIAD, AstraZeneca, Boehringer Ingelheim, Bristol-Myers Squibb, Calithera Biosciences, Genentech, Medivation, Novartis, Oncomed, Synta; and research funding (institutional) from AstraZeneca. Dr Pal disclosed honoraria from Astellas Pharma, Medivation, Novartis; consulting or advisory roles with Astellas Pharma, Aveo, Bristol-Myers Squibb, Exelixis, Genentech, Myriad Pharmaceuticals, Novartis, Pfizer; and research funding from Medivation.

American Society of Clinical Oncology (ASCO) 2017 Annual Meeting. Abstract LBA11516. Presented June 3, 2017.

ASCO 2017-A NOVEL DRUG FOR EGFR+ NSCLC

The investigational agent dacomitinib (Pfizer) showed potential as a new treatment option for patients with advanced EGFR mutation–positive non–small cell lung cancer (NSCLC).

In a head-to-head trial, this second-generation EGFR inhibitor proved superior to one of the first of these targeted agents, gefitinib (Iressa, AstraZeneca), in progression-free survival (PFS) and duration of response.

"Dacomitinib should be considered as a new treatment option for first-line management of patients with advanced EGFR-mutated NSCLC," said lead author Tony Mok, MD, professor and chair of the Department of Clinical Oncology at the Chinese University of Hong Kong, in China, who presented the results of his study at the American Society of Clinical Oncology (ASCO) 2017 Annual Meeting.

However, toxicity was higher with dacomitinib, with 66% of patients requiring a dose reduction, compared to 8% for those receiving gefitinib.

Thus, patients should be aware of the potential side effects when making treatment decisions, Dr Mok noted.

ASCO expert John V. Heymach, MD, PhD, from the University of Texas MD Anderson Cancer Center, Houston, also highlighted the toxicity that was seen.

"In this study, we see more than a 5-month difference in progression-free survival," he said. "From the perspective of doctors who treat lung cancer, this is a substantial advance, and I think it really puts dacomitinib at the front of the pack in terms of efficacy."

But, that said, the efficacy does come at the cost of some toxicity. "About 10% of patients had grade 3 toxicity involving skin and diarrhea, and a substantial number had dose reductions, but I'd like to emphasize that these aren't life-threatening toxicities," said Dr Heymach. "These are toxicities that doctors who treat this for a living become accustomed to managing. This drug clearly requires close monitoring and careful surveillance by experienced care providers to manage toxicities."

Overall, dacomitinib appears to be a clear, potential option for patients, he added. "And as a physician, I think discussing the different options in the trade-off of toxicity vs efficacy would be entirely appropriate if the drug were approved in the United States," he said.

First Head-to-Head Study

The new results come from the ARCHER 1050 study, the first randomized phase 3 study to compare a second-generation and a first-generation EGFR inhibitor head to head for first-line treatment of patients with advanced EGFR-mutated NSCLC.

The study randomly assigned 452 patients who were newly diagnosed with stage IIIB or IV EGFR-positive NSCLC to receive dacomitinib (45 mg PO QD) or gefitinib (250 mg PO QD).

Nearly two thirds (64%) of patients in both arms were never smokers; 59% had an exon 19 deletion, and 41% an L858R mutation in exon 21.

The median PFS for dacomitinib was 14.7 months, as compared to 9.2 months with gefitinib, an increase of 5.5 months. Overall survival data are not yet mature.

Duration of response was 14.8 vs 8.3 months in favor of dacomitinib, and the overall response rate was similar for both study arms (74.9 vs 71.6, P < .3883).

Dose Reductions Needed

The most commonly reported grade 3 adverse events with dacomitinib were dermatitis acneiform (13.7%) and diarrhea (8.4%); for gefitinib, increased ALT levels (8.5%).

No new safety signals were identified for dacomitinib, Dr Mok explained, but dose modification was significantly higher among patients in the dacomitinib cohort.

A total of 87 (38.3%) patients who received dacomitinib required a dose reduction from 45 mg to 30 mg, and 63 (27.8%) required a further reduction to 15 mg. The median duration of dose reduction was more than double that required for gefitinib (11.3 months vs 5.2 months).

"We do have high toxicity due to the high rate of inhibition," said Dr Mok. "We do require dose reduction of dacomitinib, and this has been done successfully."

Dr Mok explained that dacomitinib is a more potent, second-generation EGFR inhibitor that is associated with increased side effects in the skin and gastrointestinal tract, such as those with another second-generation drug, afatinib (Gilotrif, Boehringer Ingelheim). "But in spite of this, the activity seen in this study should allow for consideration of this effective therapy in this patient population," he said.

He added that for the patients treated with dacomitinib, improvements in patient-reported measures of key disease-associated symptoms were similar to those reported in the gefitinib group.

Great Benefit but High Toxicity

Commenting on the study, Yanis Boumber, MD, PhD, assistant professor at Fox Chase Cancer Center, Philadelphia, Pennsylvania, noted that although the newer drug was more effective than gefitinib, more than half of patients who received the new drug had to reduce the dose because of moderate to severe diarrhea and skin rash.

"The results with dacomitinib are encouraging and could provide an additional effective option to this category of patients, but the significant side effects make it somewhat less appealing," he told Medscape Medical News. "Caution would be needed when deciding on the best treatment option in this patient population."

ASCO 2017-IMMUNOTHERAPY AND CHEMOTHERAPY FOR BREAST CANCER

At the 2017 ASCO Annual Meeting, results were presented from the phase II I-SPY 2 trial investigating pembrolizumab (Keytruda) in combination with standard therapy (paclitaxel followed by doxorubicin and cyclophosphamide) as a neoadjuvant treatment for patients with locally advanced triple-negative breast cancer or hormone receptor–positive/HER2-negative breast cancer (Abstract 506).  

Findings showed that the addition of pembrolizumab increased the estimated pathologic complete response rate nearly threefold in patients with triple-negative breast cancer (60% vs 20%) and in patients with hormone receptor–positive/HER2-negative breast cancer (34% vs 13%) compared to standard therapy. Overall, based on Bayesian predictive probability of success in a confirmatory phase III trial, pembrolizumab has graduated from the I-SPY 2 TRIAL for all signatures in which it was tested (triple-negative breast cancer, all HER2-negative, and hormone receptor–positive/HER2-negative).

“Pembrolizumab in combination with standard therapy tripled the rate of pathologic complete responses in HER2-negative patients in the I-SPY 2 Trial,” said Laura J. Esserman, MD, MBA, Professor of Surgery and Radiology and Director of the Carol Franc Buck Breast Care Center at UCSF Helen Diller Family Comprehensive Cancer Center, and the overall principal investigator for the I-SPY trials. “The regimen indicates a new and important treatment pathway and gives us well-grounded hope for new options for patients with these aggressive breast cancers—and that’s potentially very good news.” 

More Details About I-SPY 2

The I-SPY 2 trial (NCT01042379) is a standing phase II randomized, controlled, multicenter trial for women with newly diagnosed, locally advanced breast cancer (stage II/III), and is designed to screen promising new treatments and identify which therapies are most effective in specific patient subgroups based on molecular characteristics (biomarker signatures). The primary endpoint is to determine whether the combination of certain therapies increases the probability of pathologic complete response in the breast and the lymph nodes at the time of surgery. The data presented at ASCO from the I-SPY 2 trial were based on results observed in patients at high risk of relapse using upfront tumor profiling (including hormone receptor status, HER2 status, and the MammaPrint 70-gene signature test).

Patients were treated with weekly standard chemotherapy (paclitaxel) for 12 weeks, with or without pembrolizumab, followed by doxorubicin and cyclophosphamide every 3 weeks for four cycles. Sixty-nine patients were adaptively randomized to receive pembrolizumab in the trial from December 2015 until it graduated in November 2016. In total, 46 patients have undergone surgery; the other 23 have on-therapy magnetic resonance imaging assessments.

Major Findings

In patients with triple-negative breast cancer, an absolute increase in the estimated pathologic complete response rate of 40% was observed in the pembrolizumab arm (based on the estimated pathologic complete response rate of 60% with pembrolizumab plus standard therapy compared to 20% with standard therapy alone). In patients with HER2-negative breast cancer, an absolute increase in the estimated pathologic complete response rate of 30% was observed in the pembrolizumab arm (based on the estimated pathologic complete response rate of 46% with pembrolizumab plus standard therapy compared to 16% with standard therapy alone). In patients with hormone receptor–positive/HER2-negative breast cancer, an absolute increase in the estimated pathologic complete response rate of 21% was observed in the pembrolizumab arm (based on the estimated pathologic complete response rate of 34% with pembrolizumab plus standard therapy compared to 13% with standard therapy alone). The Bayesian model estimated pathologic complete response rates appropriately adjusted to characteristics of the I-SPY 2 population, including MammaPrint status.

The safety profile of pembrolizumab was consistent with that observed in previously reported studies across tumors. In the pembrolizumab arm, grade 3­ to 5 treatment-related adverse events include diarrhea (n = 5), febrile neutropenia (n = 5), fatigue (n = 4), anemia (n = 3), nausea (n = 3), neutropenia without fever (n = 1), peripheral motor neuropathy (n = 1), peripheral sensory neuropathy (n = 1), and vomiting (n = 1). Immune-mediated adverse events of grade 3 to 5 include adrenal insufficiency (n = 5), hepatitis (n = 2), colitis (n = 1), and hypothyroidism (n = 1). Five of six patients presented with adrenal insufficiency after completion of doxorubicin/cyclophosphamide (21–24 weeks after starting pembrolizumab), and one presented during pembrolizumab treatment (5 weeks after starting pembrolizumab).

“Not all breast cancers are the same—and there has continued to be a significant gap in the treatment options available for patients with certain subtypes, particularly triple-negative breast cancer,” said Rita Nanda, MD, a medical oncologist at The University of Chicago. “The results observed in this trial are not only encouraging, but demonstrate the potential for treatment combinations that can make a difference in patient outcomes.”

ASCO 2017-GUT MICROBIOME AND PROGRESSION OF MELANOMA

The blend of bacteria in the digestive tract of metastatic melanoma patients is associated with disease progression or delay in patients treated with immunotherapy, reported Wargo et al at the 2017 ASCO Annual Meeting (Abstract 3008).

Their study of fecal samples from 105 patients treated with immune checkpoint blockade indicates that certain characteristics of patients’ microbiomes correlate with slower disease progression, while other qualities are associated with rapid worsening of the disease.

“Greater diversity of bacteria in the gut microbiome is associated with both a higher response rate to treatment and longer progression-free survival,” said study leader Jennifer Wargo, MD, Associate Professor of Surgical Oncology at The University of Texas MD Anderson Cancer Center.

Dr. Wargo and colleagues also found that an abundance of specific bacteria is also associated with higher response rate and longer progression-free survival.

“The microbiome appears to shape a patient’s response to cancer immunotherapy, which opens potential pathways to use it to assess a patient’s fitness for immunotherapy and to manipulate it to improve treatment,” said Dr. Wargo, who is also Co-Leader of the Melanoma Moon Shot, part of MD Anderson’s Moon Shots Program.

In collaboration with the Parker Institute for Cancer Immunotherapy, Dr. Wargo’s team is developing the first immunotherapy-microbiome clinical trial, with a goal of launching it later this year. 

Researchers have found in recent years that single-celled organisms and viruses harbored in the human body outnumber the body’s own cells. This microbiome plays a role in many regular functions, including the initial priming of the immune system and then its normal operation.

Study Findings

Patients in the study were treated with immune checkpoint inhibitors that block the activation of programmed cell death protein 1 (PD-1), a protein on immune system T cells that halts immune response. The drugs’ effect is to free the immune system to attack tumors.

The team found that patients with more varied types of bacteria in their digestive tract had longer median progression-free survival, defined at the time point where half of studied patients have their disease progress.

As a median time to follow-up of 242 days, the patient group with high microbiome diversity had not reached median progression-free survival, while those with intermediate diversity had median progression-free survival of 232 days and those with lower diversity had median progression-free survival of 188 days.

Specific bacterial types also had an apparent effect. More than half of those with abundant Faecalibacterium prausnitzii had not reached median progression-free survival, while half of those with low abundance had their disease progress by 242 days.

An abundance of Bacteroidales was associated with more rapid disease progression, with patients at high abundance having median progression-free survival of 188 days, while those with lower levels of the bacterium had median progression-free survival of 393 days.

Research has shown that a persons’ microbiome can be altered by diet, exercise, antibiotic use, or, more recently, through transplantation of fecal material. Dr. Wargo cautions that there is much to understand about the relationship between the microbiome and cancer treatment and urges people not to attempt self-medication with probiotics and other methods. 

Illuminating Microbiome Mechanisms?

As they develop human clinical trials, Dr. Wargo and colleagues also are conducting lab and mouse model research to better understand the mechanisms that connect bacteria and the immune system. This will include a project funded by Stand Up to Cancer that involves fecal transplants from patients who responded to therapy and from nonresponders into germ-free mice providing favorable and unfavorable microbiomes to study in detail.

The team conducted 16S rRNA sequencing, an analysis of the presence of 16S ribosomal RNA used to identify bacteria, to determine microbiome composition from fecal samples.

Whole genome sequencing and immune monitoring were conducted on the tumors after treatment and in some tumors before treatment. The immune profiling showed that responders to anti–PD-1 treatment had significantly increased immune infiltrates in their tumors, including the presence of CD8+ killer T-cells, correlated to the abundance of a specific bacterium.

ASCO 2017-VITAMIN D SLOWS COLORECTAL CANCER PROGRESSION

In recent years, observational data have shown that higher plasma levels of vitamin D are associated with improved survival in colorectal cancer patients.

Now, for the first time, a randomized trial has shown that disease progression is slowed with high-dose supplements.

The results, from a phase 2 clinical trial known as SUNSHINE, indicate that a high dose of vitamin D supplementation significantly improved progression-free survival (PFS) by about 2 months compared to a low dose.

The trial was conducted in patients with previously untreated metastatic colorectal cancer. All participants received standard treatment with the mFOLFOX6 chemotherapy regimen (ie, folinic acid [leucovorin], fluorouracil, and oxaliplatin) plus bevacizumab.

This is the first-ever completed randomized trial of the use of vitamin D as a colorectal cancer therapy, said lead author Kimmie Ng, MD, of the Dana Farber Cancer Institute in Boston, Massachusetts, who presented the study here at the American Society of Clinical Oncology (ASCO) 2017 Annual Meeting.

"Patients seemed to do better on the high-dose vitamin D. I am really excited by the data," she told Medscape Medical News.

"A phase 3 trial is warranted," she added.

Another expert expressed similar enthusiasm about the trial. "The findings from this study are incredibly exciting," said Song Yao, PhD, a molecular epidemiologist at Roswell Park Cancer Institute in Buffalo, New York, who was asked for comment.

The findings from this study are incredibly exciting. Dr Song Yao

He pointed out that at the 2015 ASCO Gastrointestinal Cancers Symposium, the same team showed that in an observational study, patients with higher levels of vitamin D survived longer than those with lower levels. "This new study provides the much-needed evidence-based randomized trial design," he commented.

Another clinician already assesses vitamin D levels in colorectal cancer patients.

"I check vitamin D levels and replete vitamin D when necessary for my patients, but we need more data to know if this should be practice changing," said Allyson Ocean, MD, a gastrointestinal oncologist at Weill Cornell Medicine and New York–Presbyterian Hospital in New York City.

She also told Medscape Medical News that the results are "quite intriguing" and that a phase 3 trial is needed.

Dr Ng reported that in the high-dose group (n = 69), the median PFS, which was the primary endpoint, was 13.1 months, compared with 11.2 months for the low-dose group (n = 70). That translated into a 31% reduced relative risk for disease progression in the high-dose group (unadjusted hazard ratio, 0.69; P = .04).

Patients in the high-dose group received a loading dose of 8000 IU/day of vitamin D3 orally for 2 weeks followed by 4000 IU/day. Those in the low-dose group received a standard vitamin D3 dose of 400 IU/day.

Median follow-up was 16.9 months in the high-dose group and 17.9 in the low-dose group.

Each group received similar numbers of chemotherapy cycles, and both groups were highly compliant with vitamin D supplementation. The primary tumor locations (right, left, and transverse) were also similar for both groups.

The disease control rate in the high-dose group was 96% vs 84% in the low-dose group (P = .05).

The high dose did not increase toxicity. There was also significantly less serious (grade 3 and 4) diarrhea in the high-dose group (12% vs 1%; P = .02).

The results were even more impressive because there was an imbalance between the two study groups that favored the low-dose group: 60% of the low-dose group had the best possible performance status vs only 42% of the high-dose group.

In other words, the high-dose group fared better despite being less physically fit than the comparator group.

Notably, more patients in the high-dose vitamin D arm were able to undergo surgery after their chemotherapy (11 vs 6). However, the difference was not statistically significant (P = .19), Dr Ng acknowledged. "It's an intriguing finding," she said.

The trial and its results have not gone unnoticed. "There's a lot of interest from providers and patients," said Dr Ng.

Among the 139 patients who enrolled and ultimately participated in the trial, most were from New England; a minority were from Nashville, Tennessee (at Vanderbilt University).

Geography may have played a role in the results, suggested Andrea Cercek, MD, of Memorial Sloan Kettering Cancer Center in New York City, who acted as discussant of the study. In New England, she said, "there is a little less sunshine than other parts of America."

This raised a question about already existing levels of vitamin D in the study participants. (Vitamin D3 is made by human skin when exposed to sunlight.) "It's unknown if patients were deficient by US standards," she said.

It's unknown if patients were deficient by US standards. Dr Andrea Cercek

Dr Cercek also said that results from other studies of vitamin D supplementation in cancer patients are mixed. One study indicated no reduction in the risk for adenomas, and another had a negative finding ― reduced survival in patients with prostate cancer who received vitamin D supplements.

Reservations aside, she wanted to see more research: "I agree 100% with the investigators that a phase 3 study is warranted."

The study was supported by funding from the National Cancer Institute, Dana-Farber, Consano, Pharmavite, and Genentech. Multiple study authors, including Dr Ng, have financial ties to industry, including Genentech. Dr Yao has disclosed no relevant financial relationships.

American Society of Clinical Oncology (ASCO) 2017 Annual Meeting. Abstract 3506. Presented June 5.

Follow Medscape senior journalist Nick Mulcahy on Twitter: @MulcahyNick

ASCO 2017-GENETICS OF COLORECTAL CANCER IN YOUNG

Younger patients with colon cancer appear to have more than three times as many mutations in their tumors as older patients, which could lead to more effective treatment decisions, according to researchers at Georgetown Lombardi Comprehensive Cancer Center. In a new study, they found that tumor mutation load, as well as gene mutations that play an important role in DNA repair, were more predominant in the younger patients. Their findings were presented by Weinberg et al at the 2017 ASCO Annual Meeting (Abstract 3592).

The investigators undertook their analysis at a time when the rate of new colorectal cancers in patients aged 45 or younger is increasing.

According to Mohamed E. Salem, MD, Assistant Professor of Medicine at Georgetown Lombardi and senior investigator for the study, viewing high levels of mutations in tumors as a positive may seem counterintuitive, but it could be important to figuring out what therapies would work best. Importantly, he noted that immunotherapies work by taking the brakes off the immune system; the more mutations a cancer cell has, the more alien it appears to the immune-fighting cells hunting for cells that don't belong.

Study Details

In one part of their study, the researchers looked at a biologic mechanism known as mismatch repair, which occurs when a DNA strand is replicated and the wrong base is inserted into a strand being copied. If there are mutations in the genes that direct the mismatch repair, then more tumors can arise.

In the other part of the study, the investigators looked at tumor mutation load, which is a count of mutations in DNA. Tying the two factors together is the essence of their finding, as a greater number of mismatch repair gene mutations can contribute to a high tumor mutation load, which was seen in more young than old patients in the study.

The investigators zeroed in on the biology of distal tumors found in the part of the colon closest to the rectum; distal tumors also include those found in the rectum. Distal tumors are on the rise in younger patients and typically convey better survival odds than proximal tumors, which occur further up the colon.

The researchers looked at advanced distal tumors from 229 patients with colorectal cancer with a median age of 40 years, and compared them with distal tumors from 503 patients with a median age of 71. Most colorectal cancers appear after age 60.

Major Findings

Using advanced gene-sequencing techniques, the researchers determined the tumor mutation load for each tissue sample and catalogued which genes were most frequently mutated in those samples. Although the researchers found a wide array of mutated genes, many of which play important roles in various types of cancer, there was no statistically significant difference between younger and older patients in the rates of mutation in many cancer-causing genes.

A few genes, however, were more frequently mutated in younger patients: HER2, NF1, and the DNA mismatch repair genes MSH6, MSH2, and POLE. The investigators suggested that the DNA mismatch repair mutations may explain the higher tumor mutation load in younger patients. Significantly, high tumor mutation load was seen in 8.2% of young patients vs 2.6% of older patients—over a 3-fold difference.

“One of the leading theories for why rates of colorectal cancer are increasing in younger patients relates to lifestyle factors, including diet and exercise,” said the study's principal investigator, Benjamin Weinberg, MD, Chief Hematology/Oncology Fellow at Georgetown Lombardi. “There is also increasing evidence that bacteria and local inflammation of the colon can drive cancer growth. We can now add tumor mutation load to the list of factors and begin exploring whether there is a link between tumor mutation load and these lifestyle factors.”

ASCO 2017-PEMBROLIZUMAB IN UROTHELIAL CANCER

The PD-L1 inhibitor pembrolizumab continued to demonstrate an overall survival (OS) benefit compared with chemotherapy among patients with advanced urothelial cancer with no new safety signals, according to the third planned survival analysis of KEYNOTE-045 presented on June 5 (Abstract 4501).

KEYNOTE-045 was an open-label phase III trial in which 542 patients with advanced urothelial cancer that recurred or progressed after platinum-based chemotherapy were randomly assigned 1:1 to receive pembrolizumab (200 mg every 3 weeks) or the investigator’s choice of chemotherapy (paclitaxel, docetaxel, or vinflunine). It is now a mature trial, with 33 patients continuing to receive the study drug compared with no patients in the chemotherapy arm, Dean F. Bajorin, MD, FACP, FASCO, of Memorial Sloan Kettering Cancer Center, said.

Earlier data from the trial presented after a median follow-up of 14.1 months and published in The New England Journal of Medicine in March 2017 showed significantly longer median OS with pembrolizumab (10.3 months) compared with chemotherapy (7.4 months; hazard ratio [HR] for death 0.73, 95% CI [0.59, 0.91]; p = 0.002).1

That survival benefit was maintained at 18.5 months, Dr. Bajorin said, with an HR for death of 0.70 (95% CI [0.57, 0.86]; p = 0.0004). The benefit was maintained across all subgroups, regardless of age, ECOG performance status, prior therapy, liver metastases, histology, or investigator choice of chemotherapy.

The median survival in the pembrolizumab arm was the same as previously reported: 10.3 months (95% CI [8.0, 12.3]) compared with 7.4 months in the chemotherapy arm (95% CI [6.1, 8.1]), with significance remaining regardless of PD-L1 expression. At 18 months, 36.1% of patients receiving pembrolizumab were alive compared with 20.5% of patients receiving chemotherapy.

Progression-free survival (PFS) was not statistically significant. However, Dr. Bajorin said, the PFS at 18 months was 16.8% in the pembrolizumab arm compared with 3.5% for the chemotherapy arm, “inferring that the survival distribution and benefit will not change over further observation times.”

Patients receiving pembrolizumab demonstrated an objective response rate of 21.1% and a complete response rate of 7.8%, compared with 11.0% and 2.9%, respectively, in the chemotherapy arm.

The duration of response remains ongoing with pembrolizumab, Dr. Bajorin said, but was 4.4 months among patients receiving chemotherapy.

Fewer patients experienced treatment-related adverse events (AEs) with pembrolizumab compared with chemotherapy (any grade, 61.3% vs. 90.2%, respectively; grade > 3, 16.5% vs. 49.8%, respectively). The majority of AEs in the pembrolizumab cohort were pruritus, fatigue, and nausea, most of which were grades 1 and 2. Decreased neutrophil levels and neutropenia were rare among the patients who received pembrolizumab, but developed in approximately 15% of patients receiving chemotherapy, with a 7.5% febrile neutropenia rate.

Immune-related AEs were primarily grade 1 and 2, with hypothyroidism, pneumonitis, and hyperthyroidism as the most common. All immune-related AEs occurred in fewer than 10 patients.

“Data from this study support the use of pembrolizumab in patients following platinum chemotherapy as a new standard of care with level 1 evidence,” discussant Jonathan E. Rosenberg, MD, of Memorial Sloan Kettering Cancer Center, said. However, he added, “it is worth noting that there are late adverse events related to immune events that are seen even beyond 1 year of therapy, and, therefore, continued monitoring and vigilance with these patients is necessary.”

Dr. Rosenberg called immunotherapy a “groundbreaking, dramatic breakthrough” in the treatment of urothelial cancer. “It has transformed the treatment of previously treated patients and patients with cisplatin-ineligible metastatic bladder cancer.”

However, he said, “despite these advances, a minority of patients benefit, with an objective response rate in the second-line setting ranging from 14.8% to 21.1%, and the median survival less than a year. More work is yet to be done.”