Optimizing the Treatment Sequence for Metastatic Castration-Resistant Prostate Cancer

Liam Davenport


November 17, 2017

Much is made of the need to individualize cancer therapy, particularly for a disease like metastatic castration-resistant prostate cancer (CRPC), where an array of treatments are available when choosing an ideal therapeutic strategy.

Yet individualizing therapy to the patient sitting in front of you is often not straightforward, as it depends on how you approach the concepts of treatment, treatment responses, and even outcomes and on whether you take a treatment-centric or patient-centric perspective.

At a packed session at the European Society for Medical Oncology (ESMO) 2017 Congress, two prostate cancer experts trawled through the available evidence to assess whether there is an emergent optimal sequence of therapy in metastatic CRPC and examined the best approaches for providing individualized care.

Is There an Optimal Sequence?

Nicholas James, MD, PhD, a professor of clinical oncology at Queen Elizabeth Hospital Birmingham and University of Birmingham, United Kingdom, outlined the series of huge shifts in treatments for CRPC since the modern therapeutic paradigm began in 2010.

At that time, docetaxel (Taxotere®) was the only option for symptomatic and asymptomatic metastatic disease. Moreover, there was little available to treat patients without metastases.

The intervening years have seen not only the introduction of drugs such as abiraterone (Zytiga®), enzalutamide (Xtandi®) and cabazitaxel (Jevtana®) but also the shift of each drug into different time periods along the spectrum of disease progression, with these treatments available at earlier and earlier stages.

Radium-223 dichloride is a recent addition to the armamentarium, following the phase 3, randomized, double-blind, placebo-controlled ALSYMPCA trial demonstrating that it is associated with a significant 30% improvement in survival versus placebo.[1]

Despite a trial examining the treatment's effectiveness as a first- or second-line treatment in CRPC, it is currently reserved for symptomatic metastatic disease.

The problem with all of these drugs is that they were tested in silos.

"If you want to get the same results as we saw in the clinical trials, you have to do the same as the trial, but the problem with all of these drugs is that they were tested in silos; on their own, to the exclusion of all the other treatments," Dr James said. "So we don't know how they perform against each other, and we don't know how they perform in different sequences."

He also pointed out that the split between asymptomatic and symptomatic metastatic CRPC, with its implications for treatment choices, does not reflect what happens in the real world, as "most patients, if you manage them well, will only have a very short symptomatic space at the end."

The question then becomes whether, after all of the developments and additions to treatment options in recent years, there is indeed an optimal sequence in CRPC from all of the many possible combinations. Dr James believes that, in his opinion, the answer is no.

One solution to the question of how to sequence drugs, however, may come from examining the control arms of trials conducted in hormone-naive prostate cancer patients.

Dr James said that outcome data from the CHAARTED[2]; STAMPEDE, for both docetaxel[3] and abiraterone[4]; and LATITUDE trials[5] are instructive. Even though the treatments were used in different proportions of patients and in different ways, there was a similar effect on progression.

Indeed, in CHAARTED, and for both docetaxel and abiraterone in STAMPEDE, the median overall survival was 44 months, while survival was 34 months in LATITUDE and 32 months for patients in CHAARTED with high-volume disease.

Dr James also noted that data from the control arm of the STAMPEDE trial showed that the treatment a patient receives initially affects what they receive later, which may then have an impact on outcomes.

For example, an analysis of data on 566 patients from the trial shows that abiraterone plus standard of care is associated with significantly better failure-free survival and progression-free survival (PFS) than docetaxel in the same setting.

There is also weak evidence to suggest that abiraterone improves metastatic PFS over docetaxel but no evidence that it improves either symptomatic skeletal event rates, cause-specific survival, or overall survival. However, the toxicity profiles of the two drugs are "quite different."

Consequently, Dr James believes that "there isn't any good evidence" to support a particular sequence of therapies in CRPC, and there is no "one size fits all." Treatment therefore needs to be individualized to the patient.

"The one thing I would say is that, if you are going to use radium-223, you should use it first or second line and not fifth or sixth line...there are no data supporting its use in that setting," he said.

Dr James also noted that both abiraterone and docetaxel "improve survival, but they also alter the patterns of care," adding: "I think that's a very important point of discussion in your clinics because you'll have different access to these drugs."

If the available treatment options for CRPC offer generally comparable outcomes, and yet the choice of specific therapy is to be individualized, the question then becomes: On what basis?

After all, implicit in the notion of individualized therapy is the recognition that not all patients respond equally to all treatments. Moreover, once a patient has started a specific therapy, they may need to have their treatment changed.

Why Therapy Would Be Switched

There may be a number of reasons why a patient may need to switch therapy, said Howard I. Scher, MD, chief of the Genitourinary Oncology Service and D. Wayne Calloway Chair in Urologic Oncology at Memorial Sloan Kettering Cancer Center in New York, in the same session at ESMO.

One obvious example would be if the treatment didn't work and a patient experiences a rapid rise in prostate-specific antigen (PSA) level and/or new disease symptoms.

Or perhaps the drug worked but it caused intolerable adverse effects; or it worked for a while, but later there were emergent adverse events or even signs of disease progression.

However, the challenge with assessing treatment effects, and thus the point at which a patient should be switched to a different therapy, is that traditional measures of response are not necessarily helpful. Rises in PSA levels, for example, may not always be bad, just as falls in levels may not be good.

Assessing bone disease is also challenging, as the RECIST 1.1 criteria, which are tumor- and not patient centric, do not apply in this situation, and the interpretation of bone scans is not standardized.

Furthermore, it may not be possible to have a meaningful measure of progression, as, for example, lymph node involvement is common but may not affect the patient's prognosis. There are also few validated biomarkers.

Emphasizing that there is a distinction between following the rules of a drug development protocol and treating a patient, Dr Scher said, "The most important thing is to be a doctor."

Putting the Patient at the Center

This requires that the treatment objectives be defined based on a number of clinical and patient factors and that significant manifestations of the disease, such as symptoms or bone involvement, be acted upon. However, the intervention chosen should be one that provides the greatest clinical benefit.

Dr Scher pointed to work that he and his colleagues have done on the Prostate Cancer Clinical Trials Working Group (PCWG).[6] This frames treatment outcomes either as early response measures centered around the control, relief, or elimination of existing disease manifestations or as later responses measured in terms of delaying or preventing manifestations.

Using these precepts, it is possible to divide treatments into those that control pain or symptoms, those that delay the development of skeletal-related events, and those that delay death rather than those that achieve reductions in PSA levels, tumor shrinkage, favorable bone scans, or reductions in circulating tumor cells.

The PCWG principles urge clinicians to consider disease manifestations separately at baseline and after treatment. They reject global response criteria and aim to ensure that drugs are not stopped prematurely based on changes in disease manifestations that are, in fact, not clinically significant.

An important example is that apparent worsening on a bone scan should be followed up with a second scan to distinguish disease flare-ups from true progression.

Indeed, the use of the PCWG2 bone scan criteria in the COU-AA-302 trial,[7] which compared abiraterone plus prednisone with prednisone alone in 229 patients with metastatic CRPC, allowed 72% of patients to continue with treatment, as a subsequent bone scan did not confirm apparent disease progression seen on the first.

A Nuanced Approach to Treatment Changes

Since then, the PCWG3 was convened to identify subsets of CPRC patients in whom specific agents are appropriate or contraindicated and to develop intermediate response endpoints.[8]

Among other recommendations, the PCWG3 advocates the serial biologic profiling of patients using tumor biopsy samples, blood-based diagnostics, and/or imaging to offer insights into the resistance mechanisms and to identify predictive biomarkers.

Dr Scher emphasized that prior therapy can have an impact on the biology of a relapsing tumor, and so such biologic profiling "is important when a management decision is essential."

The decision to discontinue a therapy and switch to another therefore becomes more nuanced. In the COU-AA-302 trial,[7] a significant event leading to discontinuation was unequivocal clinical progression, defined by Dr Scher as:

  • Cancer pain requiring chronic opiate analgesia for at least 7 days

  • Immediate need to initiate cytotoxic chemotherapy

  • Radiation therapy or surgical intervention due to tumor progression

  • Deterioration in Eastern Cooperative Oncology Group performance status to grade 3 or higher

Conversely, it is not necessary to have disease progression to change therapy. A novel concept introduced in the PCWG3 addressed that notion: no longer clinically benefiting (NLCB).[8]

This allows providers and patients to decide to continue or discontinue a treatment based on its primary therapeutic objective, such as quality of life, patient-reported outcomes, or survival, which came from analyses of trials indicating that patients had been left on treatment even though it was known not to be effective.

Novel Predictive Markers on the Horizon?

Despite making better use of the established treatment outcomes and response measures to target, or discontinue, treatments depending on the individual patient, there is a need for more accurate and clinically meaningful predictive biomarkers.

In the final presentation of the session, Joaquin Mateo, MD, from the Institute of Cancer Research and the Royal Marsden Hospital, London, United Kingdom, said that the "introduction of abiraterone and enzalutamide have changed the landscape of androgen receptor aberrations."

This has opened up several avenues of research for identifying novel biomarkers that can predict treatment outcomes, centering around the PI3K/Akt/PTEN pathway.

However, Dr Mateo emphasized that any subsequent assays developed from these markers will need validation and clinical qualification, ideally by including them in clinical trials.

He concluded that even if such markers are not yet "ready for standard of care, genomics will impact patient care soon, as we have seen in other diseases."


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: