The still-high unmet medical need for highly effective and safe pharmacotherapies for cancer has resulted in 16 new anticancer drugs being approved by the U.S. Food and Drug Administration (FDA) in 2018 alone. Meanwhile, some of the new medicines relied on the results of clinical trials that selected inappropriate control groups. Again, this raises questions about the real effectiveness of such drugs.
Traditionally, the regulator requires randomized clinical trials, in which experimental treatment is compared to an existing treatment that has undergone controlled trials and has long been or has been successfully used in practice. If the candidate drug is superior to the control drug and has no obvious/unacceptable safety problems, the regulator gives a positive verdict in most cases.
It would seem obvious that the choice of drug for the control group is very important and should settle on the best option available at the time. However, in reality, it is not uncommon that the reference drug is already outdated or for some reason no longer used by practicing oncologists. Comparison with similar drugs, heroes of bygone days, gives grounds for suspicion of fictitiousness/truthfulness of the conducted clinical trials. Doctors and patients need to ask themselves: Is the new drug better than the existing ones, which are also cheaper?
Independent experts reviewed all of the randomized clinical trials that paved the way for new anticancer drugs (or expanding the list of indications for existing ones) approved by the FDA between January 2013 and July 2018. Current guidelines and scientific articles parsing cancer pharmacotherapy, published at least one year prior to enrollment in relevant clinical trials, were also reviewed.
As it turned out, 16 cancer drugs (17%) out of 96 FDA-approved were clinically tested against a control group that was not receiving the best of the previously approved and available therapies. And even a re-analysis involving a selection of treatment recommendations published two years before patients were randomized showed that 14 anticancer drugs (15%) turned to comparison against suboptimal control groups.
When parsing the nature of suboptimal comparative treatment choices, four clinical trials (25%) assigned placebos while effective comparison drugs were available, ten (63%) relied on a knowingly worse drug or did not allow combination therapy, and two (13%) reused previously prescribed drugs, although such exposure had no benefit to patients.
In other words, regulatory approval of a new drug does not always mean that it is better than those currently used by oncologists. For example, Imbruvica (ibrutinib), which received approval for priority therapy for chronic lymphocytic leukemia, was compared in the RESONATE-2 clinical trial (NCT01722487) to chlorambucil, which has repeatedly been inferior to alternative drugs and is not commonly used in practice. Opdivo (nivolumab), approved against previously untreated metastatic melanoma, passed the CheckMate-066 (NCT01721772) clinical trial comparing dacarbazine, although the best at the time was a combination of dacarbazine and Yervoy (ipilimumab).
The analysis also demonstrated that most (n=14/16) regulatory verdicts from randomized clinical trials with suboptimal control groups resulted in full approval without the need for additional postmarketing studies to confirm clinical efficacy. And that’s a problem. Unless an experimental drug is characterized by robust evidence of its own superiority over standard therapy, there is a great risk that physicians will offer patients a treatment that is no better or even worse than the existing one, with likely higher costs and, not unlikely, with some unusual adverse reactions.
Another problem is that serious adverse events such as disease progression or death can accumulate at an accelerated rate in the suboptimal control groups, meaning that the experimental drugs automatically look better.
Of course, choosing treatment options for comparison groups is a somewhat subjective task. Still, regulators should set a strict bar for clinical trial design requirements. The choice of therapies in the control groups should reflect actual clinical practice, taking into account the fact that standards of care are regularly updated. Cancer patients, who bear the incredible burden of physical and financial toxicity, will be immensely grateful for more effective and safer approaches.
By the way, Lartruvo (olaratumumab) has already suffered a well-deserved punishment when Eli Lilly completely recalled the drug due to its inability to treat soft tissue sarcoma.
List of drugs approved by the FDA based on randomized clinical trials in which a control group was given suboptimal therapy
|Comments on Suboptimal Control
|Adcetris (brentuximab vedotin)
|Primary cutaneous anaplastic large cell lymphoma (ALCL), or CD30+ expressing mycosis fungoides who received previous therapy
|Oral methotrexate or bexarotene
|ORR lasting 4 months
|Vorinostat and romidepsin, histone deacetylase (HDAC) inhibitors, not allowed in control despite FDA approval.
|Advanced neuroendocrine tumor of gastrointestinal or lung origin
|Placebo + BSC
|Standard of care is somatostatin analogue (SSA). Patients randomized to placebo and were not able to receive SSA (about 45% of patients in each arm SSA naive) with placebo, crossover not allowed except for emergent carcinoid symptoms.
|First-line chronic phase chronic myeloid leukemia (CML)
|MMR at 12 months
|Nilotinib and dasatinib had already demonstrated better MMR compared with imatinib.
|Advanced renal cell carcinoma (RCC) in following 1 line of therapy
|Selected the second-line drug that had not beaten an active competitor. Axitinib for second-line treatment (active against sorafenib) approved in 2012. Everolimus compared with placebo.
|Advanced gastric or gastroesophageal junction (GEJ) adenocarcinoma with disease progression on fluoropyrimidine or platinum-containing therapy
|Taxanes previously shown to be active but not allowed to be used in control arm.
|Gazyva/Gazyvaro (obinutuzumab) + bendamustine
|Follicular lymphoma who relapsed after/refractory to rituximab
|Single-agent benadamustine not optimal, 20% of enrolled patients deemed rituximab refractory after exposure to single agent rituximab without chemotherapy, 40% of enrolled patients deemed refractory while on maintenance rituximab, rituximab active when combined with chemotherapy in both populations.
|Gazyva/Gazyvaro (obinutuzumab) [+ chlorambucil]
|Previously untreated chronic lymphocytic leukemia (CLL)
|Chlorambucil OR chlorambucil + rituximab
|Almost no US patients. Rituximab optional choice in control arm but should not have been when comparing with another CD20 antibody.
|First-line metastatic non-small cell lung cancer (NSCLC) with EGFR exon 19 deletions or exon 21 L858R mutations
|LUX-Lung 3 (NCT00949650)
|Gefitinib approved in EU before start of enrollment on trial. Data showed superiority of gefinitib to chemotherapy in this population.
|Aragon Pharmaceuticals, Johnson & Johnson
|Nonmetastatic castration-resistant prostate cancer
|Antiandrogen bicalutamide had been used in this setting for several years prior to enrollment on this trial. Patients had to stop bicalutamide therapy before randomization.
|Merck & Co.
|Metastatic urothelial carcinoma who progress on platinum therapy
|Paclitaxel, docetaxel, or vinflunine
|Phase 2 data on nab-paclitaxel or pemetrexed published in 2006 but neither allowed in control arm.
|Lartruvo (olaratumab) + doxorubicin
|Soft tissue sarcoma for which anthracycline is appropriate
|Doublet chemo is SOC (docetaxel/ gemcitabine, doxorubicin/ifosfamide).
|AstraZeneca, Merck & Co.
|Germline BRCA-mutated HER2-negative metastatic breast cancer, previously treated with chemo
|Capecitabine, eribulin, or vinorelbine
|Platinum not allowed as investigator's choice. Efficacy of platinum chemotherapy well documented in BRCA-mutated cancers.
|Ninlaro (ixazomib) [+ lenalidomide + dexamethasone]
|Multiple myeloma after at least one prior therapy
|Placebo (+ lenalidomide + dexamethasone)
|Triplet regimen salvage therapy standard for fit patients in 2011 (bortezomib /lenalidomide /dexamethasone, CyBorD). Moreover, 31% of patients had not received bortezomib, parenteral proteasome inhibitor (ixazomib oral proteasome inhibitor).
|Chronic lymphocytic leukemia (CLL) with or without 17p deletion, who have received at least 1 line of therapy
|Bendamustine + rituximab
|Most patients (95%) previously treated with alkylating agent, 78% previously treated with anti-CD20 antibody; therefore many clinicians may not have chosen reexposure to control arm.
|Pfizer, Astellas Pharma, Medivation
|Nonmetastatic castration-resistant prostate cancer
|Antiandrogen bicalutamide had been used in this setting for several years prior to emolument on this trial. Patients had to stop bicalutamide therapy before randomization. Concurrent trial ran comparing enzalutamide to standard bicalutamide for the same indication.
|Metastatic ALK-positive non-small cell lung cancer (NSCLC)
|Crizotinib FDA approved in 2011 for ALK-positive NSCLC and was shown to be superior to platinum-based chemotherapy.