Improving dose optimization in oncology: The sotorasib example
Click Here to Manage Email Alerts
Until the last decade, efforts to develop a therapy that successfully targeted the KRAS mutation had been futile.
KRAS — a key player in promoting tumorigenesis via the mitogen-activated protein kinase (MAPK) pathway — is mutated in roughly one-quarter of all known cancers. Some malignancies, like pancreatic cancer, have an extremely high KRAS mutation rate (approximately 90%).
KRAS has been difficult to target primarily because these oncoproteins lack a deep binding pocket for small molecules to adhere to with high affinity.
In May, the FDA granted accelerated approval to the first KRAS-blocking drug, sotorasib (Lumakras, Amgen). The agent is indicated for adults with KRAS G12C-mutated locally advanced or metastatic non-small cell lung cancer. The FDA also approved two companion diagnostics — one each for plasma and tissue — recommending that if no mutation is detected in a plasma specimen, tumor tissue should be tested.
The FDA approved dose for most cancer drugs, including sotorasib, is based on the maximum tolerated dose observed in phase 1 trials. However, most early phase trials do not evaluate the minimum effective dose, a more logical and safer endpoint.
Although the oncology community applauded the first FDA-approved KRAS inhibitor, many remain cautious about dose selection and optimization.
The sotorasib example highlights the need for improved regulatory oversight on integrating dose optimization methods earlier in drug development.
Clinical data in lung cancer
KRAS is the most common oncogenic driver in lung adenocarcinoma. Approximately 13% of patients harbor KRAS G12C mutations.
First-line treatment typically consists of immune checkpoint inhibitors alone or in combination with chemotherapy. Outcomes for patients who progress after first-line therapy generally are poor.
A phase 1 trial by Hong and colleagues evaluated sotorasib for 129 patients with KRAS G12C-mutated advanced solid tumors. The cohort included 59 patients with NSCLC, 42 with colorectal cancer and 28 with other malignancies, such as melanoma and pancreatic, endometrial and appendiceal cancers.
The planned dose levels for the escalation cohorts were 180 mg, 360 mg, 720 mg and 960 mg.
Researchers observed no dose-limiting toxic effects. More than half (56.6%) of patients experienced any-grade treatment-related adverse events; 11.6% reported grade 3/grade 4 events, including elevated liver function tests, diarrhea, and anemia.
Efficacy appeared greatest in the NSCLC cohort, as one-third (32.2%) of patients achieved confirmed objective response and 88.1% achieved disease control. Median PFS in this cohort was 6.3 months (range, 0.0+ to 14.9).
By comparison, in the colorectal cancer cohort, 7.1% achieved confirmed response and 73.8% achieved disease control. Researchers also reported responses among patients with other cancers.
Skoulidis and colleagues conducted a phase 2 trial that included 126 patients with refractory KRAS G12C-positive NSCLC. All patients received 960 mg sotorasib once daily, the recommended phase 2 dose based on the dose-escalation study.
Results showed a 36.1% objective response rate and 80.6% disease control rate. Median duration of response was 11.1 months, with median PFS of 6.8 months and median OS of 12.5 months.
More than two-thirds (69.8%) of patients experienced treatment-related adverse events.
Twenty-five patients (19.8%) experienced a grade 3 event and one patient (0.8%) experienced a grade 4 event. These rates were slightly higher than those observed in the phase 1 trial, but adverse event types were similar and researchers reported no new safety signals.
As the researchers noted, the durable clinical benefit and safety data support sotorasib as an effective option for KRAS G12C-positive NSCLC.
Although these results are promising, drug resistance is a constant challenge with nearly any targeted therapy.
Mechanistic data suggest activated SOS competes with sotorasib for binding to KRAS G12C, thereby increasing resistance. Preclinical studies showed significantly greater activity with the combination of the EGFR inhibitor cetuximab (Erbitux, Eli Lilly) and KRAS G12C inhibition vs. KRAS inhibition alone.
In a study of another KRAS inhibitor, adagrasib (MRTX849, Mirati Therapeutics), Awad and colleagues determined nearly half of the 38 treated patients developed molecular lesions likely to cause resistance. These included novel RAS mutations and amplifications; MET amplification, activating mutations in BRAF, MAP2K1 or RET; oncogenic fusions in ALK, RET, RAF and FGFR3, and loss-of-function mutations in NF1 and PTEN.
KRAS inhibitors may have limited efficacy as monotherapy, and combination trials are a logical next step. Nonetheless, this novel class of targeted therapies provides a promising opportunity in a historically “undruggable” cancer type.
Call for improved dose optimization
The purpose of phase 1 studies is to evaluate drug safety and determine a recommended phase 2 dose. If the dose is too high, it can result in excessive toxicity; if a dose is too low, it can result in lack of efficacy.
However, the field of oncology historically has adopted the dogma of “more is better”; hence the traditional 3+3 dose-escalation design that involves a stepwise increase in total dose until approximately one-third of the population experiences dose-limiting toxicities.
Notably — unlike traditional chemotherapy, which is administered for a specific number of cycles in a predefined timeframe — targeted small molecules often are used continuously until disease progression,.
Short follow-up in phase I trials of targeted therapies may not reflect long-term toxicities that could appear after months or years of use (eg, atrial fibrillation among patients treated with ibrutinib [Imbruvica; Janssen, Pharmacyclics] and blood clots among those treated with ponatinib [Iclusig; Takeda Oncology]).
There is a need for better dose optimization of targeted therapies, including methods to determine the minimum effective dose that produces the maximum biological and clinical effect. Pharmacokinetic and pharmacodynamic data are critical to determine the lowest dose causing receptor saturation.
Depending on a drug’s metabolic pathway, pharmacogenomic studies also may help inform dosing in subgroups of patients who have impaired metabolism. This would allow for greater confidence of drug safety in phase 2 and phase 3 trials, during which grade 3 and grade 4 toxicities generally become more apparent.
The recommended phase 2 dose — and the FDA-approved dose — of 960 mg for sotorasib is based on the phase 1 trial, which utilized a Bayesian logistic regression model to guide dose escalation from 180 mg to 960 mg. Researchers aimed to elicit dose-limiting toxicity among 20% to 33% of patients.
This is despite its highly targeted nature, which may spare healthy normal tissue. Notably, there was no evidence of a dose-response relationship, raising the question whether such a high dose is justified.
In an article published earlier this year in The ASCO Post, Mark J. Ratain, MD, FASCO, and Allen S. Lichter, MD, FASCO, noted the phase 1 trial also mandated dosing under fasting conditions, which could impair absorption. They suggested a much lower dose of the drug administered with food may have a superior therapeutic index, and they urged the FDA to require the sponsor to optimize the dose.
Amgen agreed to conduct a randomized trial to compare 960 mg daily with 240 mg daily. Although none of the cohorts in the phase 1 trial received the 240-mg dose, Amgen intends to proceed with this dose based on preclinical, pharmacokinetic and clinical data. Results are expected in late 2022.
Shifting the paradigm
A HemOnc Today cover story published in 2019 — titled “Cancer drug doses: More is not always better” — explored approaches in hematology/oncology to achieve similar efficacy with lower therapeutic doses.
Several experts interviewed for the story discussed the pitfalls of a “more-is-better” approach.
Efficacy signals may be detected at lower doses, and it is critical that drug developers evaluate biological and clinical response at each dose level during phase 1 trials.
Ibrutinib, for example, is approved at a fixed dose of 420 mg daily for treatment of chronic lymphocytic leukemia.
However, a study by Advani and colleagues demonstrated 2.5 mg/kg resulted in 95% occupancy or blockade of Bruton’s tyrosine kinase (BTK). The extent of BTK occupancy at 4 hours and 24 hours did not increase with further dose escalation.
Chen and colleagues found that ibrutinib 420 mg daily resulted in higher trough plasma concentrations compared with 280 mg daily; however, they noted no difference in BTK percent inhibition.
Unsurprisingly, real-world data show nearly half of all patients discontinue treatment early or require dose reductions due to adverse events. These observations — which were not detected in early-phase trials — highlight the importance of earlier dose optimization.
An analysis by Chiuzan and colleagues determined 93% of 1,712 dose-finding trials of molecularly targeted agents and immunotherapies published between 2008 and 2014 utilized a rule-based design, such as 3+3, the rolling six or the accelerated titration designs.
The authors noted the percentage of dose-finding studies that used a model-based or novel design more than tripled between the previous literature review, conducted in 2007, and their study period. However, the actual proportion remained low.
Other approaches may be used to optimize drug exposure and dose. These include therapeutic drug monitoring, particularly for oral tyrosine kinase inhibitors with high interindividual pharmacokinetic variability and well-established therapeutic indices.
Another approach is to capitalize on drug-food effects to lower the dose and increase absorption, as seen with abiraterone for men with castration-resistant prostate cancer.
Weight-based vs. fixed dosing and extending time between doses, such as with immunotherapies, also can provide opportunities to limit dosing without compromising drug efficacy.
Patients and oncologists have come to expect some degree of adverse events, but this should not be the norm. Further, there is an inexorable rise in cancer drug spending, partly driven by over-dosing.
Sponsors should be required to provide evidence of dose optimization earlier, including identification of the minimum effective dose.
It is time for the oncology community to move on from the obsolete maximum-tolerated dose concept for molecularly targeted oncologic agents.
References:
Advani RH, et al. J Clin Oncol. 2013;doi:10.1200/JCO.2012.42.7906.
Amodio V, et al. Cancer Discov. 2020;doi:10.1158/2159-8290.CD-20-0187.
Awad MM, et al. N Engl J Med. 2021;doi:10.1056/NEJMoa2105281.
Chen LS, et al. Blood. 2018;doi:10.1182/blood-2018-06-860593.
Chiuzan C, et al. J Biopharm Stat. 2017;doi:10.1080/10543406.2017.1289952.
Groenland SL, et al. Am Soc Clin Oncol Educ Book. 2021;doi:10.1200/EDBK_319567.
Hong DS, et al. N Engl J Med. 2020;doi:10.1056/NEJMoa1917239.
Mato AR, et al. Haematologica. 2018;doi:10.3324/haematol.2017.182907.
Ratain MJ and Lichter AS. The ASCO Post. Empowering the FDA to require dose optimization of all new oncology drugs. Available at: ascopost.com/issues/january-25-2021/empowering-the-fda-to-require-dose-optimization-of-all-new-oncology-drugs. Accessed on Aug. 21, 2021.
Roda D, et al. Clin Cancer Res. 2016;doi:10.1158/1078-0432.CCR-15-1855.
Skoulidis F, et al. N Engl J Med. 2021;doi:10.1056/NEJMoa2103695.
Skoulidis F, et al. Abstract 9003. Presented at: ASCO Annual Meeting (virtual meeting); June 4-8, 2021.
For more information:
Jai N. Patel, PharmD, BCOP, CPP, is chair of cancer pharmacology and associate professor in the division of hematology/oncology at Levine Cancer Institute at Atrium Health. He also is a HemOnc Today Editorial Board Member. He can be reached at jai.patel@atriumhealth.org.
Collapse
Click Here to Manage Email Alerts