Immune checkpoint blockade inhibitors are revolutionizing care for patients with metastatic melanoma and other malignancies. The first generation of FDA-approved checkpoint blockade inhibitors targeted the cytotoxic T-lymphocyte-associated protein-4 (CTLA-4), while the next generation of inhibitors inhibit either the programmed death 1 (PD-1) receptor or programmed death ligand 1 (PD-L1). These agents act by inhibiting negative regulators of the immune system, thereby providing a net effect of immune stimulation.
In September 2014, the FDA approved the first anti-PD-1 therapy, pembrolizumab (Keytruda), for the treatment of patients with metastatic melanoma who have previously progressed on ipilimumab, with the stipulation that if the patient’s tumor harbored a mutation in BRAF, the patient must receive a BRAF inhibitor as well. In December 2014, the FDA approved the second anti-PD-1 agent, nivolumab (Opdivo), for the same indication as pembrolizumab.
Both pembrolizumab and nivolumab have demonstrated a remarkable clinical efficacy-to-toxicity ratio. Specifically, these therapies are providing durable remissions to a substantial minority of patients (20% to 40%) while at the same time proving to be very safe, with very few grade 3/4 (<10%) adverse events. In addition to melanoma, these agents, as well as other anti-PD-1/PD-L1 agents, are showing promising results in multiple other tumor types including lung cancer, bladder cancer, kidney cancer, lymphoma, and many others.
Given the clinical success and lowtoxicity of anti-PD-1/PD-L1 agents, an exponential increase in their use in treatment regimens is inevitable. With the increased use of these agents as wellas the relatively substantial costs, now is a good time to begin thinking about how best to use these agents to optimize value both for the individual patient and for healthcare system at large. Moving forward, the following 3 actions are among those that could be taken to improve the value of these agents:
• Recognize both the patients who are most likely to benefit and those who are unlikely to benefit from the agent.
• Identify the optimal duration of therapy, which remains unknown. • Determine the best sequence or combination in which to administer these agents.
UPFRONT IDENTIFICATION OF PATIENTS LIKELY TO BENEFIT
Many in the field are exploring either clinical or translational biomarkers that can reliably predict who will benefit from anti-PD-1/PD-L1 therapies. Initially, Topalian et al demonstrated that the lack of tumor expression of PDL-1 had a very strong negative predictive value, with 0 of 17 patients with PD-L1-negative tumors not responding to nivolumab (anti-PD-1).1 In that same study, 9 of 25 (36%) patients with PD-L1-positive tumors did respond.1 Subsequent studies have confirmed that while PD-L1-positive patients are more likely to achieve clinical benefit with anti-PD-1/PD-L1 agents, patients whose tumors are PD-L1-negative can also respond, albeit at a lower rate than patients with PD-L1-positive tumors.2,3 Specifically, Daud et al demonstrated that while patients with PD-L1-positive tumors did have better outcomes with pembrolizumab, those with PD-L1-negative tumors had a response rate of 4% and disease control rate of 17%.3 Furthermore, Herbst et al demonstrated that PD-L1 expressio on the tumor-infiltrating lymphocytes, not the tumor, strongly correlated with improved outcomes in patients receiving Roche’s anti-PD-L1 antibody (MPDL3280A), while patients with PD-L1-negative or low PD-L1-expressing tumors also achieved a response rate of 13%.2 Outside of tumor expression of PD-L1, at least 1 study demonstrated that baseline tumor size was the strongest independent predictor of benefit from pembrolizumab, with patients with lower tumor burdens much more likely to achieve clinical benefit than those with larger tumors.4 In summary, further work is necessary to more reliably identify those who are more likely and less likely to benefit from anti-PD-1/PD-L1 agents. PD-L1 expression on either the tumor or tumor infiltrating lymphocytes is helpful to enrich for patients who are more or less likely to respond, but the positive and negative predictive value is far from perfect. Given the enormous potential for anti-PD-1/PD-L1 agents to provide clinical benefit, withholding the drug from patients who have a reasonable chance to benefit would not be ethical. However, when considering value to the healthcare system, perhaps withholding these agents from those who are extremely unlikely to benefit would be the ideal goal.
IDENTIFICATION OF OPTIMAL DURATION OF TREATMENT
A second key way to improve value of anti- PD-1/PD-L1 therapies is to determine the optimal duration of therapy for patients who are responding. For example, a hallmark of most successful immunotherapies, including the use of anti-PD-1/PD-L1 agents, is the ability to provide a durable remission and even a cure. While the therapy is still in its infancy, the median duration of response has not been reached in the majority of patients who achieved a clinical response to anti-PD-1/PD-L1 therapies, suggesting that many patients may respond indefinitely. Questions I am frequently asked by patients who are responding are, “How long do I need to stay on therapy?” or “What will happen if I stop therapy?” Unfortunately, we are currently lacking scientific answers to these questions. Future studies that address the added benefit of continuing therapy after patients respond should help answer this question.
On the flip side, and due to the well-described phenomenon of pseudoprogression (or late responders), it is often challenging to determine when to stop using these agents in patients who are not responding following initial scans. For example, a small minority of patients (5%-10%) treated with both pembrolizumab and nivolumab experienced disease progression at initial restaging (12 weeks) but then went on to respond at a later time point.5 With that being said, pseudoprogression is relatively rare, and therefore continuing therapy in all patients who are not initially responding decreases the overall value of these agents. Further work to identify patients who are unlikely to be late responders would help limit the use of these agents and thereby provide additional value. In summary, identifying the optimal duration of therapy in responders, as well as identifying markers that could reliably rule out pseudoprogression in patients who are not responding, could significantly improve the overall value of anti-PD-1/PD-L1 agents.
IDENTIFICATION OF THE PROPER SEQUENCE OF ANTI-PD-1/PD-L1 AGENTS IN THE COURSE OF THERAPY
At present, both pembrolizumab and nivolumab are FDA approved only in patients with metastatic melanoma who have previously progressed on ipilimumab, and if the patient’s tumor harbored a mutation in BRAF, the patient must also progress on a BRAF inhibitor. However, members of the National Comprehensive Cancer Network believe that anti-PD-1/PD-L1 agents should be considered in the frontline setting due to improved clinical efficacy and toxicity compared with ipilimumab.6 Furthermore, in a phase 3 study of previously untreated patients with metastatic melanoma, nivolumab demonstrated improved overall survival (OS) compared with those treated with dacarbazine, providing evidence that anti-PD-1 agents can and do provide OS benefit when given in the frontline setting.7 In addition, there are retrospective data, at least, that suggest that patients who are treatment-naïve have a smaller tumor burden and are perhaps more likely to benefit from these agents.4 Ongoing studies will further elucidate the benefit of anti-PD-1/PD-L1 agents in the frontline setting.
When discussing the value of anti-PD-1 agents, administration of anti-PD-1/PD-L1 molecules in the frontline could provide overall value to the healthcare system and to the individual. For example, anti-PD-1/PD-L1 agents given in the front line provide durable remissions, which could obviate the need for expensive and toxic future therapy. Fortunately, additional clinical studies testing these agents in the frontline setting are currently under way and will, it is hoped, provide more clues to where these agents are most effective.
Immunotherapies with anti-PD-1/PDL1 agents have revolutionized care for patients with metastatic melanoma, and preliminary data suggest that these agents likely will revolutionize the care of patients with other malignancies as well. Given the predicted widespread use of these immune-modulating agents, combined with their high costs, continued efforts are necessary to optimize their efficacy and thus improve their value. Specifically, improving our use of predictive biomarkers, further refining the optimal duration of therapy, and identifying the proper sequence of these agents will greatly enhance the value equation. Advanced clinical studies designed to address these questions are needed. EBO
Richard W. Joseph, MD, is assistant professor, Division of Medical Oncology, at the Mayo Clinic in Jacksonville, Florida.
1. Topalian SL, Hodi FS, Brahmer JR, et al. Safety, activity, and immune correlates of anti-PD-1 antibody in cancer. N Engl J Med. 2012;366(26):2443-2454.
2. Herbst RS, Soria JC, Kowanetz M, et al. Predictive correlates of response to the anti-PD-L1 antibody MPDL3280A in cancer patients. Nature. 2014;515(7528):563-567.
3. Daud A, Hamid O, Robert C, et al. Relationship between programmed death ligand 1 (PD-L1) expression and clinical outcome in patients with melanoma treated with pembrolizumab (MK-3475). Paper presented at: Society for Melanoma Research, 2014 International Congress; November 13-16, 2014; Zurich, Switzerland.
4. Joseph RW, Elassaiss-Schaap J, Wolchok JD, et al. Baseline tumor size as an independent prognostic factor for overall survival in patients with metastatic melanoma treated with the anti-PD-1 monoclonal antibody MK-3475. J Clin Oncol. 2014;32(suppl 5s). Abstract 3015.
5. Hodi FS, Ribas A, Daud A, et al. Patterns of response in patients with advanced melanoma treated with Pembrolizumab (MK-3475) and evaluation of immune-related response criteria (irRC). J Immuno- Therapy Cancer. 2014;2(3):P103. 6. NCCN clinical practical guidelines in oncology, melanoma. Version 2.2015. National Comprehensive Cancer Network website. http://www.nccn.org/professionals/physician_gls/pdf/melanoma.pdf. Accessed January 19, 2015. 7. Robert C, Long GV, Brady B, et al. Nivolumab in previously untreated melanoma without BRAF mutation [pulished online November 16, 2014].