Ben Leach

Jeffrey A. Sosman, MD, is professor of Medicine and director, Melanoma and Tumor Immunotherapy Program at the Vanderbilt-Ingram Cancer Center in Nashville, Tennessee. At the 11th International Congress on Targeted Therapies in Cancer, held August 16-17, 2013 in Washington, DC,

The International Journal of Targeted Therapies in Cancer

spoke with Dr. Sosman about his conference presentation on advances in immunotherapy in melanoma.

IJTT: In your presentation, you mentioned that in clinical trials involving immunotherapies, you would see relatively few responses, but the responses that you did see when the patients received these treatments were durable, and that appears to be the case for all immunotherapies currently being studied. Can you explain this phenomenon and what research has been done to determine what is happening in these instances?

The patient is a 53 year old woman with metastatic melanoma to multiple subcutaneous sites and retroperitoneum.

Source: Dr. Steven Rosenberg, the Web site of the National Cancer Institute (http://www.cancer.gov).

There is no question that when we used drugs like interferon and interleukin 2 in the past we saw relatively few responses. However, among patients who responded, some responded for long periods of time. I think what&rsquo;s believed to happen is that memory forms in their immune response against their cancers. In fact, I have never seen a renal cancer patient who has relapsed after three or four years free of disease following interleukin-2. Now there are few of these post-IL-2 long-term responders, but it seems as if they have less chance of relapsing than someone who has been free of disease after nephrectomy. So there is some critical immunologic change. I think that the best example is the vitiligo seen in patients with melanoma who receive effective therapy. When we see responses, we frequently see some parallel destruction of their melanocytes, and this can even surround the melanoma. The immune response persists and even gets more consistent and more diffuse with time, as a sign of the memory that treatment has induced.

What is the mechanism of action of ipilimumab? What has allowed it to be proven to be efficacious for patients with metastatic melanoma?

There is an easy answer to that and then a little more complex answer. The simpler way of thinking is that the brakes on the car are bound by ipilimumab releasing the car (an immune cell, a T cell), and by binding the brakes it prevents the immune cell from braking normally, thereby basically running away without the ability to turn off its motor or its immune response. That&rsquo;s the simpler way of thinking about the drug action. There are some recent data that suggest that ipilimumab might also play a role in eliminating one of the subsets of cells that I spoke about that are immunosuppressive, the so-called &ldquo;regulatory T cells.&rdquo; They express CTLA4 on their surface, and Alan Korman has recently shown that these regulatory T cells are eliminated as part of the treatment in animal models. But most of us think of this loss of brakes as unregulated activation of the T cells, which leads to both the response and some of the toxicity as well.

In your talk you mentioned that using singleagent small-molecule inhibitors is not really an option once a patient&rsquo;s disease progresses. Can you tell us a little bit more about what you mean by that, and how immunotherapies fit into this discussion?

I think small-molecule inhibitors clearly are not going to be very effective for longterm, durable responses. In general, as a single agent, whether it is vemurafenib in melanoma or erlotinib or crizotinib in subsets of non-small cell lung cancer, despite the fact that they are all very good agents that can improve survival, 100% of the patients receiving these agents will ultimately relapse. So there are two strategies addressing that. The first strategy is to add more inhibitors, and the second is to add a different modality, and most of usat least those of us working with melanoma and probably those working with lung cancer as well&mdash; are thinking about adding the immune treatment. Now, whether the immune treatments directed at CTLA4 and PD-1 can be used in combination really will be the key. There is a big trial to look at that, and we will see if it really is better. But when you start talking about combining all of these, you do have some cost issues.

As someone who specializes in melanoma, it must be interesting that so many new agents have recently been approved for the treatment of this disease. What do we know about sequencing these various therapies at this point in time?

Right now in the practical sense, that&rsquo;s a really important question, and hopefully there will be a trial in the cooperative groups that looks at that. One of the problems with developing such a trial has been deciding what agents will make up the sequence. That&rsquo;s been changing. Initially, it was ipilimumab or vemurafenib, but now it is anti- PD-1/PD-L1 and maybe ipilimumab versus a MEK inhibitor, trametinib with dabrafenib. I think most of us right now believe that some patients will probably have a better chance and a long-term durable response if they receive immune therapy first. Now even with that said, it is not so easy that we can predict speed of disease progression. I think we make a judgment when we see the patient. If they have indolent disease, we might have more time, and we might be able to treat them later with a targeted agent. However, that&rsquo;s not always the case, and frankly, some of those patients I think are harmed by doing that, but we make a guess to the best of our ability. I think we need a trial--some of the patients who get targeted therapy, specifically small molecules initially and then, when they relapse, some of them relapse with fairly aggressive disease, and at that point it can be very difficult to give immunotherapy. So there is a belief and many anecdotes that in certain groups of patients, it would be better to give immune therapy first. That&rsquo;s not a patient who has rapidly growing disease or high LDH or poor performance status who needs response. That is instead the patient who has a performance status of 0 and feels great and has small-volume disease.

What should community oncologists know about the adverse events associated with these newer therapies, and are there any other challenges?

I think one thing I have to really applaud is something that the pharmaceutical industry is trying to do. It is trying to really regiment the approach to toxicity, and they have refined the algorithms over and over, and I think they are getting to a point where as long as there is good communication between the patient and the physician&rsquo;s office, these patients can probably be managed very effectively. Now, if you don&rsquo;t have that communication or if you don&rsquo;t have an effective algorithm that really needs to be done, you can get into big trouble. I think especially with ipilimumab where we have seen that toxicity, there are now really nice diagrams and approaches that have been very clearly outlined, and I think that&rsquo;s a real advantage. I tend to probably err on the overly aggressive side in terms of reacting to side effects. I think with PD-1 and similar molecules, I am actually a little concerned that sometimes we&rsquo;re overreacting to cough or to a little shortness of breath. We are still trying to refine the approach to anti-PD-1 pneumonitis.

So it is really still a learning curve for the therapy?

I think there is still some, but little by little, the pharmaceutical industry, and obviously with input from physicians in making those decisions, I think we have done a good job, and it is their role to really get that out into the community and make doctors comfortable. Community oncologists have not been overly comfortable with ipilimumab.

In your presentation, you mentioned a study presented at this year&rsquo;s ASCO meeting that looked at the combination of ipilimumab with GM-CSF. Could you give us your thoughts on the toxicity profile seen in that combination and what it really means for patients? You noted that there was a survival benefit but not a progression-free survival (PFS) benefit.

First, I don&rsquo;t know for any certainty why there was a survival benefit. It is feasible that it was just enhancing the immune response. The PFS being similar doesn&rsquo;t really bother me very much because we have seen that in immunotherapy- based trials. What makes me suspicious that it is more a toxicity issue is that those overall survival curves separate fast if you look, and it is not like vemurafenib versus DTIC. In that trial you are saving very sick patients who are still alive as opposed to being dead from their disease. I think if you look overall at that studyand I am not sure that this is a concern&mdash;it may not be so clear what GM-CSF is doing in that case. Now, if it is enhancing survival, truthfully who cares? The question is, it may enhance survival of a regimen that we may never use. That&rsquo;s why the 3-mg dosage versus 10-mg dosage study which has been completed, is really important to see what those results are.

Many clinical trials require patients to have a performance status of ECOG 0 or 1, yet many times, patients in the community might be presenting with ECOG 3 or 4. What should community oncologists do in those cases?

I think those are difficult patients, and I think the point [an audience member at the conference] made about treating them with these investigative drugs is totally valid. I think the immune- targeted drugs in those really sick patients, at least from what I have seen so far, are really unlikely to turn patients around if that performance status is due to the tumor. On the other hand, I do believe that the targeted small molecules that you will hear about can really have an impact on those patients rapidly and really have this so-called &ldquo;Lazarus effect.&rdquo; Now, the problem is that it is all too short, but it does provide some really quality time. On the other hand, Paul Chapman has tried to do a ECOG 2 and 3 study with vemurafenib, and it was very difficult to try to find patients who were poor performance status for a trial, yet not so poor that it is just totally unreasonable. I think we would like to do those studies because those are kind of real-world important studies, but they are not easy to do. So we will see if they can be done.

You have investigated the use of imatinib in melanoma for patients with KIT mutations. Could you tell us a little about how this wellestablished targeted therapy might be used in melanoma? How common are KIT mutations in patients with melanoma?

I think the latter question is really important, and I think that from our data and other people&rsquo;s data, the frequency of KIT mutations is quite low. It is really in the neighborhood of 1% to 3% of the whole population. It also appears that almost all of them come from three locations or three general body areas. One is the mucosal membranes, second is the hands and feet, the acral areas, and third to a much lesser degree is the head and neck area where there is chronic sun exposure. In some areas where primary melanomas come from the torso or extremities, you almost never see it. So it is a long, hard journey finding those patients. Also, some of those mutations are not responsive to KIT inhibitors, and we know that from imatinib, in GIST tumors or in other situations. But if you have one of a few mutations at a few codons, the studies now Stephen Hodi just published the study and Richard Carvajal had a previous study&mdash;if you look at that small group of patients, it looks like it is in the neighborhood of 50% and some are durable. Obviously they are all going to relapse with resistant disease, but I think if someone presents to me and has a KIT mutation, which we screen for, and it is one of those mutations, I certainly would consider using something, whether it is imatinib or a study looking at another KIT inhibitor.

Overall, where do we stand in terms of the use of immunotherapy in cancer? What is its future? Do you feel that it is firmly here to stay in cancer treatment? What are the most promising next areas of exploration?

I think I have become a believer again, for certainly the checkpoint inhibitors have had a huge impact. And I think they will get approved and probably in all three diseases, melanoma, renal cancer, and non-small cell lung cancer. I don&rsquo;t know if the 30% or 40% response rate will hold up, but I have confidence that the response rate will not be as low as 5% or 10%, and responses will be durable. So I think immune therapy agents are clearly here. How do we best utilize them? In the future, I think we will have to see. We certainly want to bring them to an earlier point, certainly from metastatic disease, and use them earlier upfront. Obviously the question is, can we use it in high-risk patients and prevent relapse? That&rsquo;s a really important question. I think the options now are going to be combining those two, and there is certainly a rationale to do so with ipilimumab and nivolumab. I think the other area that&rsquo;s of greatest interest is the combination of small molecules with these drugs, with immune therapy, because we know that a really high percentage of patients have tumor regression with targeted therapy. If we could somehow have the durability of the immune response to that, we can make a huge impact.

Could you talk about the trial comparing the 3-mg and 10-mg dosages of ipilimumab and the importance of doing that trial? What is it going to mean going forward with the use of ipilimumab?

Articles

At the 11th International Congress on Targeted Therapies in Cancer, The International Journal of Targeted Therapies in Cancer spoke with Dr. Sosman about his conference presentation on advances in immunotherapy in melanoma.

Immunotherapy in Melanoma

Although previously reported results of a randomized phase III study found that eribulin mesylate was not superior to capecitabine in the treatment of previously treated locally advanced or metastatic breast cancer, a subgroup analysis showed that certain patients may benefit more from eribulin, and quality of life (QoL) data suggest that patients may prefer eribulin because of a lower frequency of certain adverse events.

These additional analyses of the large, randomized trial were presented at the 2013 ASCO Annual Meeting.

In the eribulin 301 study, investigators randomized 1102 patients with locally advanced or metastatic breast cancer, <3 prior chemotherapy regimens (<2 for advanced disease), and prior treatment with anthracycline and taxane chemotherapy in a 1:1 ratio to either eribulin mesylate (1.4 mg/ m2 on days 1 and 8 of a 21-day cycle) or capecitabine (1250 mg/m2 orally twice daily on days 1-14 of a 21-day cycle).

At the 2012 CTRC-AACR San Antonio Breast Cancer Symposium in December 2012, the results of the study showed that the median overall survival (OS) was 15.9 months for eribulin compared with 14.5 months for capecitabine (hazard ratio [HR] = 0.879; 95% CI, 0.770- 1.003;

=.056) and the median progressionfree survival was 4.1 months and 4.2 months, respectively (HR = 1.079; 95% CI, 0.932-1.250;

Overall response rates were 11% for eribulin and 12% for capecitabine (

= .849). OS for HER2-negative patients was 15.9 months for eribulin and 13.5 months for capecitabine (HR = 0.838; 95% CI, 0.715- 0.983;

&ldquo;It&rsquo;s equivalent in efficacy to one of our best agents,&rdquo; said Joyce A. O&rsquo;Shaughnessy, MD, co-director of Breast Cancer Research at the Baylor Charles A. Sammons Cancer Center of Texas Oncology and PA/ US Oncology in Dallas, Texas, and a researcher who has had involvement in other trials involving eribulin in the treatment of breast cancer.

To determine whether certain patients may derive a more significant benefit with eribulin, a series of subgroup analyses were performed.

The researchers observed a nonsignificant trend favoring improved OS with eribulin compared with capecitabine in patients who received the treatment as either a first-, second-, or third-line regimen for advanced disease.

Additionally, a more pronounced benefit was observed in certain patients, including those with HER2-negative disease who received eribulin compared with capecitabine (15.9 months vs 13.5 months, respectively; HR = 0.84; 95% CI, 0.72-0.98;

= .03), ER-negative disease (14.4 months vs 10.5 months, respectively; HR = 0.78; 95% CI, 0.64-0.96;

= .02), and triple-negative disease (14.4 months vs 9.4 months, respectively; HR = 0.70; 95% CI, 0.55- 0.91;

In another analysis of the 301 study, investigators looked at QoL data to determine whether patients receiving eribulin preferred their experience compared with patients who received capecitabine.

The investigators found that over time, global health status (GHS) and overall QoL scores improved in both of the arms, but that improvement was significantly more pronounced in the eribulin arm compared with the capecitabine arm (estimated treatment effect = 6.52 GHS/overall QoL; 95% CI, 0.045-13.0;

Cognitive functioning significantly improved in patients receiving eribulin compared with those receiving capecitabine (estimated treatment effect = 15.33 QoL; 95% CI, 9.23-21.43;

<.01). Gastrointestinal side effects, including nausea, vomiting, and diarrhea, were less severe in the eribulin arm, whereas hair loss and systemic side effects were less common with capecitabine.

Kaufman PA, Awada A, Twelves C, et al. A phase III, open-label, randomized, multicenter study of eribulin mesylate versus capecitabine in patients with locally advanced or metastatic breast cancer previously treated with anthracyclines and taxanes. Presented at: 2012 CTRC-AACR San Antonio Breast Cancer Symposium; December 4-8, 2012; San Antonio, TX. Abstract S6-6.

Kaufman PA, Cortes J, Awada A, et al. A phase III, open-label, randomized study of eribulin mesylate versus capecitabine in patients with locally advanced or metastatic breast cancer previously treated with anthracyclines and taxanes: subgroup analyses.

Cortes J, Awada A, Kaufman PA, et al. Quality of life in patients with locally advanced or metastatic breast cancer previously treated with anthracyclines and taxanes who received eribulin mesylate or capecitabine: a phase III, open-label, randomized study.

A subgroup analysis showed that certain patients with breast cancer may benefit more from eribulin, and quality of life (QoL) data suggest that patients may prefer eribulin because of a lower frequency of certain adverse events.

Certain Patients May Benefit More From Eribulin

Amid a flurry of new drug approvals in recent years, the American Urological Association (AUA) has issued its first set of guidelines for the management of patients with metastatic castration-resistant prostate cancer (mCRPC). The guidelines, presented during the 2013 AUA Annual Meeting in San Diego, California, in May, incorporate an extensive role for abiraterone acetate at several stages of the disease.

Abiraterone is part of a class of newer agents called androgen synthesis inhibitors that are designed to target androgen receptors to block the signaling pathways associated with prostate cancer. In 2011, the FDA approved abiraterone in combination with prednisone for patients with mCRPC who had previously received docetaxel. In 2012, that approval was expanded to include patients with mCRPC who have not yet received chemotherapy.

&ldquo;These trials showed a definitive survival advantage,&rdquo; said Adam S. Kibel, MD, disease center leader of Urology at Dana-Farber Cancer Institute in Boston, Massachusetts. &ldquo;Patients didn&rsquo;t just show a quality of life improvement. Patients lived longer.&rdquo;

Kibel served as vice chair on a panel for the AUA that developed the guidelines for mCRPC patients. While abiraterone appears to help many patients with mCRPC, the guidelines make it clear that the drug may not be appropriate to use in all cases.

The AUA&rsquo;s board of directors approved the recommendations in April, before the FDA granted clearance for radium Ra 223 dichloride. As a result, the agent is not mentioned in the guidelines, although it may be in the future, according to a notice attached to the document.

The guidelines discuss treatment options for six different index patients that take into consideration symptoms, performance status, metastases, and whether docetaxel was administered (Table).

Two patient indices for which abiraterone plus prednisone should be offered are for patients who are asymptomatic or minimally symptomatic and have not received prior docetaxel chemotherapy (Index Patient 2), and patients who are symptomatic with good performance status who have received prior docetaxel chemotherapy (Index Patient 5).

For the first setting, Kibel said that while clinical trial data have definitely shown that survival is improved in these patients with abiraterone, it&rsquo;s not the only choice available.

&ldquo;The other two treatments, docetaxel and sipuleucel-T, also show similar survival advantages, but none of these have been tested head-to-head, so it&rsquo;s very hard to say which is better,&rdquo; Kibel said.

In the second setting, cabazitaxel or enzalutamide are also options, particularly if the patient already has received abiraterone.

There are two additional groups of patients with mCRPC for whom abiraterone plus prednisone may be appropriate: symptomatic patients with good performance status and no prior chemotherapy (Index Patient 3); and symptomatic patients with poor performance status and no prior docetaxel therapy (Index Patient 4). In both cases, Kibel said that randomized clinical trials have not specifically examined abiraterone in these groups of patients, but it is reasonable to believe that the agent is safe because of the other groups for which abiraterone is approved.

Table. Patient Categories for mCRPC Guidelines

Rising prostate-specific antigen (PSA) despite medical or surgical castration

No radiologic evidence of metastatic prostate cancer

Castrate levels of testosterone (<50 ng/mL)

Asymptomatic or minimally symptomatic mCRPC without prior docetaxel chemotherapy

Rising PSA in the setting of castrate levels of testosterone

Documented metastatic disease on radiographic imaging

No regular narcotics for pain relief for symptoms attributable to metastases

Symptomatic mCRPC with good performance status and no prior docetaxel chemotherapy

Documented symptomatic metastatic disease on radiographic imaging

Regular narcotics for pain relief for symptoms attributable to metastases

Symptomatic mCRPC with poor performance status and no prior docetaxel chemotherapy

Performance status Eastern Cooperative Oncology Group (ECOG) 3-4

Symptomatic mCRPC with good performance status and prior docetaxel chemotherapy

Castration-resistant disease based on serologic progression

Symptomatic mCRPC with poor performance status and prior docetaxel chemotherapy

&ldquo;I think we can extrapolate very reasonably that if it&rsquo;s been tested in asymptomatic patients pre-docetaxel, it&rsquo;s been tested in patients after docetaxel, and it&rsquo;s FDA-approved for patients who have a failure with metastatic disease pre-docetaxel, irrespective of symptomatology, it seems to me that it&rsquo;s very reasonable to include it in the guidelines,&rdquo; Kibel said.

However, patients with asymptomatic, nonmetastatic CRPC (Index Patient 1) should not receive treatment with abiraterone. Kibel explained that this patient population really has not been studied in a randomized trial with any therapy, let alone abiraterone. He said guidelines initially recommended observation-only for these patients, but he said the panel realized that some patients would still want some form of treatment.

While some first-generation antiandrogen agents such as bicalutamide and ketoconazole have been studied in some nonrandomized trials, and therefore may be prescribed, abiraterone has not, which is why it is not recommended in this particular setting.

Finally, patients who are symptomatic with poor performance status who have received prior docetaxel chemotherapy (Index Patient 6) are candidates for palliative care. However, the guidelines also say that clinicians may offer treatment with abiraterone plus prednisone, enzalutamide, ketoconazole plus steroid, or radionuclide therapy to selected patients.

&ldquo;We recognize that some patients really adamantly want some form of treatment, and it&rsquo;s very hard to say no to patients,&rdquo; Kibel said. &ldquo;We thought in selected patients who understood the risks and benefits and really weren&rsquo;t ready to start using palliative care that it would be reasonable to go ahead and try one of the treatments that had a low side- effect profile, which would include abiraterone but not chemotherapy.&rdquo;

Kibel said it is important to keep in mind that these are simply guidelines from the AUA, and that, because organizations use different methodologies for developing recommendations, there may not be uniform consensus across all groups about the proper way to manage patients. When it comes to abiraterone, though, Kibel said that its impact is very clear.

&ldquo;It&rsquo;s a treatment option that extends life expectancy, which is obviously a very positive thing,&rdquo; Kibel said. &ldquo;It&rsquo;s very well tolerated. And therefore, I think it&rsquo;s an excellent treatment alternative for patients that are in the appropriate category.&rdquo;

Cookson MS, Roth BJ, Dahm P, et al. Castration-resistant prostate cancer: AUA guideline. American Urological Association website. http://www.auanet.org/education/ guidelines/castration-resistant-prostatecancer. cfm. Published May 6, 2013. Accessed July 1, 2013.

Amid a flurry of new drug approvals in recent years, the American Urological Association (AUA) has issued its first set of guidelines for the management of patients with metastatic castration-resistant prostate cancer (mCRPC).

Impact of Abiraterone Evident in New AUA Guidelines for mCRPC

High doses of interleukin-2 (IL-2) could be used to treat patients with melanoma whose disease has metastasized to the brain, potentially expanding options in a population facing a grim prognosis, according to researchers from Saint Louis University in Missouri.

Investigators retrospectively reviewed outcomes for eight patients with stable melanoma brain metastases who were treated at the university between January 1999 and June 2011 with IL-2 intravenously at 720,000 IU/kg per dose on a regimen of 14 doses and two cycles per course for a maximum of two courses. Of the eight patients, one patient started treatment with lung lesions after resection of melanoma brain disease and had a partial response. The remaining seven patients had brain metastases when they began treatment with the immunotherapy.

The median overall survival (OS) for the cohort was 8.7 months (range, 2.1-19.0 months), and the median OS for those who began treatment with brain metastases was 6.7 months (range, 2.1- 18.2 months). All seven patients in this latter cohort experienced progressive disease. Patients received radiosurgery and whole brain radiation before and after IL-2 therapy.

By contrast, the median OS of patients with melanoma brain metastases is approximately four months with existing treatment options, which typically consist of radiation and surgery, Chu et al reported.

One patient did exhibit symptoms of neurotoxicity, but the investigators later learned of a history of alcohol abuse. When this patient was placed in a withdrawal program, the patient&rsquo;s symptoms improved. None of the patients who received IL-2 experienced treatment-related mortality, and for most patients, survival after diagnosis of brain metastases and IL-2 initiation was longer than expected when compared with their disease-specific graded prognostic assessment score, researchers said.

&ldquo;Our study shows that having this condition does not exclude a patient from getting this treatment and can in fact improve the length of their life,&rdquo; said principal investigator John Richart, MD, associate professor of Internal Medicine at Saint Louis University, in a statement.

Aldesleukin (Proleukin), a genetically engineered form of IL-2, has been approved by the FDA since 1998, with clinical data establishing a 600,000 IU/ kg dosage administered intravenously every eight hours for a maximum of 14 doses. &ldquo;Currently, it is the only FDAapproved regimen associated with continuous complete remission over long-term follow-up (>5 years),&rdquo; said Chu et al.

However, only 16% of patients respond to IL-2, and the therapy is associated with toxicities such as capillary leak syndrome, cardiac tachyarrythmias, and seizures, the researchers noted. As a result, it is typically reserved for &ldquo;a select group of patients with excellent performance status.&rdquo;

The researchers suggested that further studies in larger cohorts could better determine the role of IL-2 for this patient population, as well as the best use of new therapies that have been approved since the patients in this study were treated. They said prior studies have indicated a cumulative incidence of melanoma brain metastases at <10% of patients overall, but at least one-third to one-halfand possibly as many as 75%&mdash;of patients with advanced melanoma.

Chu MB, Fesler MJ, Armbrecht ES, et al. Highdose interleukin-2 (HD IL-2) therapy should be considered for the treatment of patients with melanoma brain metastases [published online ahead of print May 13, 2013].

Ben Leach | Authors

Articles

Immunotherapy in Melanoma

Certain Patients May Benefit More From Eribulin

Impact of Abiraterone Evident in New AUA Guidelines for mCRPC

High-Dose IL-2 May Benefit Brain Metastases in Melanoma