How Do I Treat…?
Practical perspectives on cancer treatment by thought leaders, explaining how they would approach the treatment of a patient in their area of expertise.
Tuesday, October 29, 2013
HYMAN B. MUSS, MD, is Professor of Medicine and Director of the Geriatric Oncology Program at the University of North Carolina Lineberger Comprehensive Cancer Center.
Breast cancer, like most solid tumors, is a disease of aging, with an incidence that dramatically rises with increasing age. As the population ages, more and more older women with breast cancer will require oncology services that will pose challenges for surgical, medical, and radiation oncologists. What’s different about older women with breast cancer is that they frequently have major comorbidity, and breast cancer is not always their most important medical problem.
The first thing I do when seeing an older breast cancer patient is take a detailed social history so I get a feel of what the patient’s life is like, her social support system, and what’s important to her. After that I usually take a few moments and calculate her life expectancy -- exclusive of her breast cancer. I use www.eprognosis.org for this. It has several scales that give reasonable estimates of survival for community-dwelling elders. You should select a scale to use in your practice because some of the questions in the various scales relate to geriatric issues that most of us don’t incorporate (or ask about) in the usual history and physical examination (like how far can you walk?). Once you know your patient’s life expectancy, the next order of business is to define the goals of therapy. For older women with metastatic disease, the goals of treatment are ameliorating cancer-related symptoms while maintaining the highest quality of life. The best approach for almost all these older women with metastases is to use endocrine therapy for those with hormone-receptor positive tumors until it is clear that the tumor is refractory to endocrine treatment, and then consider sequential single-agent chemotherapy.
For those with earlier-stage disease, the goal of therapy -- as in younger patients -- is increasing the chances for cure. Unless older women have a very short estimated survival, definitive surgery with either lumpectomy or mastectomy should be considered. Older women with hormone-receptor positive tumors can be managed with primary endocrine therapy, but unless they have very short estimated survival, the majority will have tumor progression by five years, and initial surgery is a better option for most. In addition, older women with hormone-receptor positive tumors less than or equal to 2 cm treated with lumpectomy, and who are willing to take endocrine therapy, can be spared breast irradiation without any adverse influence on survival. Women who elect this path of treatment will have a higher rate of in-breast recurrence at 10 years -- about 10 percent compared with two to three percent for patients who receive breast irradiation – but if they relapse they can be treated again with surgery and then if needed, radiation.
For older women with hormone-receptor positive tumors, endocrine therapy is the mainstay of systemic treatment, and the decision as to whether to also suggest chemotherapy should be based on its potential added value to endocrine therapy. For most node-negative patients, and many with one to three positive nodes, genomic assays (Oncotype DX and others) can be helpful in estimating the potential benefits of chemotherapy. In addition, it always pays to remember that the bulk of relapses in women with hormone-receptor positive tumors treated with an adjuvant endocrine therapy occur after five years. This is why it’s so important to calculate the non-breast cancer related estimated survival as even older hormone-receptor positive patients with extensive nodal involvement will rarely benefit from chemotherapy if their estimated survivals are less than five to 10 years.
For those with early-stage hormone-receptor negative and HER-2 negative (“triple-negative”) tumors and estimated survivals exceeding five years, chemotherapy is the treatment of choice and should be considered for most patients. Those with small tumors less than 1 cm should be considered for chemotherapy on an individual basis. Online programs such as www.adjuvantonline.com and www.predict.nhs.uk/predict.shtml can help greatly in the selection of treatment. Unless there is a major benefit from more intensive anthracycline- and taxane-based therapy in these patients -- I like to see at least a 3% improvement in overall survival for the more aggressive, toxic, treatment -- I favor non-anthracycline regimens such as docetaxel and cyclophosphamide (TC). Cyclophosphamide, methotrexate, and fluorouracil can also be considered in this setting, although it is not as effective and takes longer to complete treatment than TC does. I use white blood cell growth factors in all older patients I treat with TC as in community settings the risk of neutropenic fever can exceed 20 percent.
What about older patients with HER2-positive breast cancer? For those with early-stage potentially curable cancers and reasonable life expectancy, I recommend chemotherapy and anti-HER2 directed treatment. It’s important to factor into your treatment decision that women with hormone-receptor positive, HER2-positive tumors (HR+/HER2+) have less aggressive clinical courses than those who present with hormone-receptor negative and HER2-positive phenotypes. I use www.predict.nhs.uk/predict.shtml to calculate the added value of chemotherapy and anti-HER2 based therapy in HER2-positive patients as HER2 status can be incorporated into the model. For those for whom I recommend chemotherapy and anti-HER2 directed treatment, I tend to use non-anthracycline combinations such at docetaxel, carboplatin, and trastuzumab. Trials of anti-HER2 based adjuvant therapy without chemotherapy are in progress. In addition, new agents such as ado-trastuzumab emtansine may prove to be ideal treatments in this setting. For those with metastatic HER2-positive breast cancer there are numerous choices, and such treatment decisions should be individualized and based on the tempo of the disease, the extent of the metastases, anticipated toxicity, and symptom burden.
A comprehensive geriatric assessment can play a major role in optimizing the management of older patients with cancer, regardless of type. Screening instruments and short versions of such assessments are gaining popularity and likely to be helpful in suggesting interventions that improve function and quality of life prior to and during treatment. In addition models that use a combination of clinical and geriatric assessment variables may prove to be extremely helpful in predicting chemotherapy toxicity. Consider also the frail elderly; here the overriding goals of care are maintenance of function and quality of life and any treatment that does not support these goals should be discouraged.
My hope is that this discussion will be helpful to you for treating the older woman with breast cancer. Expected survival, which is dependent on so much more than age, and the goals of treatment, are the prime considerations when approaching the older breast cancer patient. In addition, and most important, is asking the patient and family what their expectations of treatment are, and tailoring treatment options to their wishes. Providing the best options and making the right treatment decision for an older patient frequently is a major challenge but is always worth your time and effort.
Wednesday, August 28, 2013
Paul R. Helft, MD, is Associate Professor of Medicine in the Division of Hematology/Oncology at Indiana University School of Medicine and the Indiana University Melvin and Bren Simon Cancer Center, as well as Director of the Charles Warren Fairbanks Center for Medical Ethics at Indiana University Health.
I have spent my whole practice career in an academic environment, and have learners at various stages who work with me in both inpatient and outpatient settings. I tell all of our hematology/oncology fellows who begin a rotation with me -- only partly facetiously -- that oncology is easy, but talking to cancer patients is hard.
I truly believe this to be the case: that providing communication that accomplishes the many challenging tasks that are necessary in the context of the doctor-patient relationship with cancer patients is a herculean task. It is also a set of skills in which we provide trainees the least direct training. Any one of the many difficult types of conversations we engage in merits an extensive discussion: an informed consent conversation for standard or experimental chemotherapy; about risk of recurrence and the role of adjuvant therapy; about prognosis in the setting of a newly diagnosed or advancing life-ending cancer; about cessation of disease-directed therapy and the transition to end of life care. There are many others.
The best way I have found to teach the many skills needed for such complex and emotionally charged types of communication is to discuss with trainees the strategy we might take for a given conversation before going in to see patients together, to model the behaviors I have found to be most effective, and to discuss and analyze the process after the fact.
The following are observations I have arrived at following 15 years of study, practice, analysis, and research into complex communication between cancer clinicians and patients. These are in no way meant to serve either as a scholarly treatise or a comprehensive review, both of which are available in several forms in other literature. By the same token, many others have practiced, developed, written about, and studied ways of engaging in effective and compassionate conversations with patients, so I do not want to suggest that there is one right way.
1. Building trust is the single most important task
Nearly all of the most difficult tasks that face those of us who care for and communicate with cancer patients rely on trust. Think of the times when patients and their families have left your practice for good or when communication interactions have not gone well. Underlying these are frequently issues of trust. For patients to feel comfortable with our recommendations about prognosis and therapy, they must first trust us.
Decisions regarding cessation of disease-directed therapy in the setting of advancing illness only go well when there is substantial trust established. It is helpful to think about trust like a bank account: Sometimes, the account is full -- even without effort (think of the patients who arrive at their first visit after having been told what an amazing clinician you are) -- and sometimes it is empty and can only be slowly and painstakingly filled. What goes into the account can be “spent” later during times when a difficult decision relies on substantial trust, or when missteps happen (e.g., we forget something vitally important about the patient or his or her test results).
The behaviors that help establish and develop trust are those that demonstrate caring -- active listening, acknowledgment of emotion, committed presence, appearing unrushed, knowing and remembering details about the patient, and many others.
2. Ninety percent of communication is non-verbal
Communication experts teach us that only 10 percent of communication is in the words chosen. The rest is non-verbal (e.g., body language) and tone. Yet, we spend most of our effort on choosing the words. Mindfulness of one’s non-verbal communication takes some effort, but can be highly rewarding: where and how to sit with respect to the patient, what expression to wear on one’s face, appropriate eye contact, how and when to use touch, how to modulate one’s tone or use vocal mimicry as a sign of respect and caring. These are all issues that can have a major impact on the quality of communication.
All of us have experienced patients and family members who seem to remember little of what was said, especially under stress. But they always remember the sense that they were in the presence of a committed, empathic, and caring clinician. That impression does not come from the carefully chosen or scripted words we choose.
3. Spend more time listening
So many of the pressures we face in routine patient care push us to listen less: we try to make history-taking and review of systems and symptoms as efficient as possible; all of us experienced clinicians work from a kind of script (e.g., my “speech” about stage III colon cancer or about the risk of recurrence of pancreatic cancer).
These are “macros” that make communication more efficient for us. Taking more time to listen accomplishes two major, important things: first, most of the things that really make a difference in a patient’s care are subjective to them. We can only account for those things if we ask, and then listen carefully to the answer. Listening to patients’ descriptions of what they are most scared or most angry about, or what they feel they don’t understand well can make conversations both more effective and more efficient.
Second, really listening to patients talk about both the emotional and cognitive aspects of care (“What are you most afraid of?” “How do you think you’re doing emotionally?” “What are you having the hardest time processing?”) inspires the sense that you really care, and thus leads to trust and respect in profound ways.
4. Start every important conversation with “What do you understand about… ?”
When I meet a patient for the first time, after reviewing records, taking a history, and doing a physical examination, I try never to launch right into my “speech” about the summary of their condition, prognosis, treatment options, etc. Instead, I ask them what they understand about their cancer and overall situation, what they are hoping to get out of our encounter, and I try to get a sense of who they are and what pre-conceptions they have based on research they may have done coming into the encounter.
Although this takes five to 10 minutes to listen carefully to, I find that it helps me to frame my remarks much more effectively. By the time I have listened to the patient for these few minutes, I can usually tell if they are starting with almost no information or whether they have researched their situation extensively and sought other opinions already. I can tell if a person is afraid of chemotherapy or desperately wants to begin as soon as possible. I also can usually uncover any profound misconceptions which many people walk in with, and then I know what major hurdles we will have in reaching a mutual understanding.
5. Check your anxiety at the door
I am convinced that many of our own anxieties interfere with our ability to offer therapeutic presence to patients and families. It is indeed highly anxiety-provoking to give terrible news to someone whom you already know as highly anxious and emotional. But I have found that having a “toolbox” for such interactions allows me to operate on a more solid footing. I remind myself that the interaction is not about me, but about the patients and their families.
I always give a warning that important, negative, or scary information is coming (e.g., somber demeanor and expression, a verbal warning that I have some “bad news” -- or equivalent). If I don’t know the patient well, I ask explicitly if it is okay to share negative information with them right now, and offer a later time if they don’t feel up to it; I ask if it is okay to share it as plainly and honestly as I can. They rarely refuse this, but my belief is that this gives people a great sense of being respected and makes them feel a little more in control.
When patients begin to weep or tear up, I stop talking and acknowledge their emotion verbally or non-verbally. I scoot my chair a little closer and sit with them in silence until they are ready to talk again. If family members become uncomfortable and try to break the silence, I nod toward them but continue to focus on the patient. I talk only after the patient has said something. This rarely lasts longer than a minute or two, but I once sat silently for 15 minutes.
If you do this, the patients will know that you will not run from them because of hard times and will be grateful that you did not become anxious, run away, fill in the silence, or say something glib (e.g., “You have to take every day one day at a time.”) I do not offer tissues unless the patient looks around or asks for them, because I feel it sends a message to clean up and stop crying.
6. Talking about prognosis is one of the most important and most difficult tasks for cancer clinicians
Broadly speaking, there are really two types of prognosis conversations: The first (and less challenging) is with a patient who has potentially curable disease and where the focus is therefore on the chances of cure (or, conversely, the risks of recurrence). This type of conversation can be highly anxiety-provoking, but there is always a built-in source of hope for patients, and so in my mind they are “easier.”
The second is the more classical conversation, so common in oncology, in which patients are really concerned with how long they will live given that their disease is incurable. These conversations are difficult in myriad ways, but mostly because they require finding other sources of hope.
Here are a few pearls I have learned. Patients sometimes ask for this information in roundabout ways: “I was wondering about, you know, time frames.” If I think they are asking about their prognosis (i.e., a numerical estimate of life expectancy), I ask them if that is what they are asking and if they truly feel ready to talk about it. I think this step is very important because sometimes they are asking a different question, sometimes they ask but don’t actually feel ready to talk about it fully, and because taking this step to clarify their preferences is experienced as a respectful gesture.
I ask them if they would like numbers or qualitative estimates (e.g., not very much time, weeks or months vs. years). Next, I tell them I have a long answer. I make sure they know I don’t actually know, but rather can make an educated estimate based on knowledge and experience. It is important to acknowledge that we are sometimes wrong (because we are). The further away from death patients with life-ending diagnoses are, the more inaccurate clinicians are, and I try to acknowledge the amount of uncertainly I have at any given phase of the illness. Then I try to give them an average life expectancy surrounded by a range of best- and worst-case scenarios.
I try to have a careful conversation about the fact that the disease is incurable and to clarify the actual goals of therapy. I acknowledge how hard such information is, and then try to process the emotional impact of the discussion with them.
There are a number of other important principles and techniques I and others employ and would encourage readers to take in some of the rich literature in this area. There are also other highly challenging conversations the average oncologist must engage in: discussion of adjuvant therapy, making complex decisions involving major trade-offs of quality and quantity of life or risks and benefits of therapy, informed consent for clinical trials, and many more.
I am convinced that many important tools of effective communication are both teachable and learnable, so all of us have the opportunity to become better at this most important clinical skill.
Sunday, August 11, 2013
Daniel Morgensztern, MD, is Associate Professor and Director of Thoracic Oncology Clinical Research at Washington University School of Medicine in St. Louis.
Squamous cell carcinomas (SQCC) represent approximately 30 percent of all cases of non-small cell lung cancer (NSCLC). These tumors are strongly associated with cigarette smoking and are typically centrally located. Most of the recent progress in the treatment of NSCLC, including the remarkable benefit from epidermal growth factor (EGFR) tyrosine kinase inhibitors (TKIs) for patients with activating EGFR mutation and crizotinib for patients with ALK or ROS1 translocations has been restricted to patients with non-squamous histology, particularly adenocarcinomas. There are currently no clearly effective targeted therapies approved for the treatment of SQCC.
The standard therapy for metastatic SQCC is platinum-based combination chemotherapy. Since several randomized clinical trials reported between 2001 and 2003 showed no significant differences in outcomes among the chemotherapy regimens, the choice of therapy had been based mostly on personal preferences and toxicity profile. Histology did not play a role in the selection of chemotherapy regimen until the results of the JMDB trial, which compared cisplatin plus pemetrexed (CP) with cisplatin plus gemcitabine (CG). Although the median overall survival (OS) was identical for both arms at 10.3 months, CG was associated with a significant improvement in OS among patients with SQCC (10.8 vs. 9.4 months), whereas CP was associated with a significant benefit in non-squamous histology (11.8 vs. 10.4 months).
The decreased efficacy of pemetrexed in patients with SQCC has been attributed at least in part to the higher expression of thymidylate synthase. Therefore, pemetrexed is not indicated in patients with SQCC, with the treatment options restricted mainly to the combination of platinum plus gemcitabine, vinorelbine or a taxane.
In the initial randomized Phase II study comparing chemotherapy with or without bevacizumab, there were six life-threatening pulmonary hemorrhages among the 66 patients receiving bevacizumab, including four deaths. All six patients with severe hemorrhage had centrally located tumors close to the major blood vessels, five had tumor cavitation, and four had SQCC.
With four out of 13 SQCC patients treated with bevacizumab developing severe hemorrhagic events, this histology was excluded from the two subsequent randomized trials that led to the approval of bevacizumab in NSCLC with non-squamous histology. Although the BRIDGE study showed that delaying the use of bevacizumab until the third cycle of chemotherapy and excluding patients with intrathoracic cavitation or recent hemoptysis may be associated with decreased risk of severe pulmonary hemorrhage in SQCC, the use of bevacizumab in this population remains experimental and should not be pursued outside of a clinical trial.
Two randomized clinical trials comparing chemotherapy with or without cetuximab showed discordant results. While the FLEX study using cisplatin and vinorelbine showed a significant improvement in OS for patients in the cetuximab arm (11.3 vs. 10.1 months), this benefit was not observed in the BMS099 study using carboplatin plus paclitaxel, where the numerically higher OS in the cetuximab arm did not reach statistical significance (9.6 vs. 8.3 months).
A subsequent analysis of the FLEX study showed that patients with high EGFR expression, defined as immunohistochemistry score ≥ 200, had a significant benefit from the addition of cetuximab to the chemotherapy, whereas there were no benefits from the triplet in patients with low EGFR expression.
Although high-expression of EGFR was found in only 31 percent of patients, there were significant differences according to histology (25% in adenocarcinomas and 38% in SQCC). Furthermore, among all patients evaluated, the highest benefit from cetuximab was observed in patients with SQCC and high EGFR expression.
If the predictive value for EGFR expression is validated, the addition of cetuximab to chemotherapy may be considered during first-line therapy, particularly for SQCC, where the options remain limited. The ongoing SWOG 0819 trial, comparing chemotherapy with carboplatin plus carboplatin (and bevacizumab for eligible patients), with or without cetuximab, may validate high EGFR expression as one of the predictive factor for response to cetuximab and clarify whether the benefits can be also observed in a better tolerated carboplatin-based chemotherapy regimen.
Since the JMEN study showed improved survival for switch maintenance using pemetrexed after four cycles of a platinum-based regimen compared with no maintenance, several randomized clinical trials have been conducted, mainly to compare different maintenance strategies with pemetrexed and bevacizumab alone or in combination. However, these studies are not applicable for SQCC patients, since the benefits from JMEN were observed mainly in non-squamous, and neither pemetrexed nor bevacizumab are indicated in SQCC.
Therefore, the only recent maintenance study applicable for SQCC is SATURN, which compared erlotinib with placebo in patients without tumor progression after four cycles of chemotherapy. In patients with SQCC, maintenance erlotinib was associated with a significant improvement in progression-free survival (PFS) but not OS compared with placebo. However, when stratified by response to first-line chemotherapy, patients achieving stable disease had a significant improvement in OS with the use of maintenance erlotinib. In contrast, there was no OS benefit from maintenance erlotinib in patients who responded to prior therapy.
The current approved therapies for second-line therapy in NSCLC include docetaxel, pemetrexed, and erlotinib. Docetaxel was approved based on two randomized clinical trials, showing response rates between 5.5 and 10.8 percent and median PFS between 1.9 and 2.4 months. Since pemetrexed, which was shown to be equivalent to docetaxel, has no role in the treatment of SQCC, the current alternative to docetaxel is erlotinib.
Although earlier randomized studies comparing gefitinib with docetaxel showed no significant differences in outcome, patients were unselected by EGFR mutation status, preventing the comparison of the two therapies in wild type EGFR, which is less likely to benefit from EGFR TKIs. The TAILOR study, comparing docetaxel with erlotinib in patients with wild type EGFR, showed that docetaxel was associated with a significant improvement in response rates, median PFS, and OS compared with erlotinib.
Since reaching a plateau in efficacy with standard platinum-doublets, a growing percentage of patients with adenocarcinomas have achieved a significant improvement in survival with the use of new targeted therapies. Unfortunately, this improvement has not been shared with SQCC, where activating EGFR mutations and ALK translocations are either rare or absent.
Improving outcomes for patients with SQCC can only be achieved through a better understanding of the pathophysiology and molecular characterization of this heterogeneous group of malignancies. A significant step forward was The Cancer Genome Atlas (TCGA) research network study, which showed that approximately two thirds of patients with SQCC have somatic alterations of a potentially targetable gene, including EGFR, fibroblast growth factor receptors (FGFRs), and PI3K pathway. Furthermore, studies using monoclonal antibodies targeting the programmed cell death protein 1 (PD1) or its ligand (PDL1), have shown remarkable results regardless of the histology subtype.
Metastatic Squamous Cell Lung Cancer
My first treatment option for patients with metastatic SQCC is enrollment into a clinical trial. For patients who are not eligible for a trial, the first-line therapy is guided mostly by the performance status, co-morbidities, and patient preferences. For patients with good PS and no significant co-morbidities, my favored first-line treatment is the combination of carboplatin plus gemcitabine, with carboplatin plus paclitaxel as an alternative choice. Although cisplatin has been associated with a slight benefit compared with carboplatin in meta-analyses, individual trials have not shown significant differences, and carboplatin is usually better tolerated.
In case of poor performance status, I typically use single-agent docetaxel. Although the survival improvement from maintenance erlotinib is modest in patients with wild type EGFR, I often review the risks and benefits from maintenance erlotinib, particularly for patients with stable disease after chemotherapy.
My second-line therapy of choice is docetaxel. For patients with good performance status and progression after second-line docetaxel, I usually offer erlotinib. The replacement of a mostly empirical chemotherapy by the use of new drugs directed at specific driver mutations or abnormal pathways is expected to provide better treatment options in the near future, likely resulting in improved survival for patients with metastatic SQCC.
Friday, July 26, 2013
Jeffrey A. Sosman, MD (left), is Professor of Medicine and Director of the Melanoma and Tumor Immunology Program at Vanderbilt-Ingram Cancer Center; and Douglas B. Johnson, MD, is a Clinical Fellow in the Division of Hematology/Oncology at Vanderbilt-Ingram Cancer Center.
Metastatic melanoma traditionally has had limited options and almost universally poor outcomes. Recent advances in our understanding of melanoma biology and novel melanoma therapeutics have radically altered treatment paradigms.
Approximately 40 to 50 percent of advanced melanomas harbor a BRAF V600 mutation. Selective BRAF inhibitors (vemurafenib and dabrafenib) and a MEK inhibitor (trametinib) are now available and effective treatment options in this genetically defined cohort of melanoma patients. However, the remaining 50 to 60 percent of melanoma patients without BRAF wild-type mutations (BRAF WT) do not benefit from BRAF inhibitors and remain a challenge to treat. We will review the treatment options for this population, which largely consist of immune-based therapies.
Prior to determining that a melanoma is BRAF WT, accurate molecular profiling should be performed. BRAF V600E Cobas testing is commonly used in clinical practice and is highly sensitive and specific for BRAF V600E mutations. However, many of the 20 percent of alternative BRAF mutations may not be identified with this testing platform (V600K, V600D, V600R, V600M, V600E’). Since these mutations are generally sensitive to BRAF inhibitors, extended molecular testing should be considered to determine if a melanoma is truly BRAF wild type.
Assessment of CKIT mutation status should also be performed in some cases, especially if the primary was from mucosal and acral surfaces or cutaneous sites with chronic sun exposure (head and neck; see “Imatinib” section below).
Ipilimumab is a monoclonal antibody directed against the immune checkpoint CTLA-4, which activates the immune system by blocking this key negative T-cell regulator. Ipilimumab is the only agent that has shown an overall survival benefit in the BRAF WT population, and demonstrated increased overall survival in pre-treated melanoma patients compared with use of an experimental vaccine in a Phase III study (10 vs. 6.4 months).
A trial in therapy-naïve metastatic melanoma also demonstrated a survival benefit compared with chemotherapy (DTIC). Most impressive is the finding from both trials that at 24 and 36 months the overall survival had nearly doubled (from 10-15% to 20-25%). Only approximately 10 percent of patients achieve classic tumor regression, although other patients may have delayed responses or even apparent progression prior to tumor shrinkage.
In view of these delayed responses, defining whether a patient will benefit is difficult to assess prior to 18 to 24 weeks after treatment initiation. Patients with rapidly progressive disease, elevated LDH, and visceral organ involvement appear to be less likely to benefit from ipilimumab.
Side effects are primarily caused by aberrant immune stimulation and auto-immune toxicity. Colitis, hepatitis, dermatitis, and endocrinopathies are the most frequent immune-related adverse events, although more uncommon side effects have been described as well. Severe toxicities should be treated promptly with corticosteroids.
We use this agent commonly both in treatment-naïve or pre-treated patients with BRAF WT melanoma. This agent is generally better tolerated than interleukin-2 (see below) and appears to cause prolonged remissions in a similar number of patients.
High-dose IL-2 has been a cornerstone of advanced melanoma therapy for many years. The overall response rate is in the range of 15 to 20 percent, with durable complete responses occurring in only six to eight percent of treated patients. Severe acute multiorgan toxicities and shock due to the vascular leak syndrome may occur during treatment. Therefore, we consider IL-2 only in young patients with normal organ function. Age greater than 65, untreated brain metastases, and poor performance status are generally exclusion criteria.
Despite the advent of newer therapies, we continue to use IL-2 in selected patients. There are several reasons to consider IL-2 as a first-line agent as opposed to ipilimumab. It appears that ipilimumab is equally effective following IL-2 use as in IL-2 naïve patients. However, there is significant theoretical concern to using IL-2 following ipilimumab, as the extreme immune activation caused by IL-2 may unleash severe delayed ipilimumab toxicities.
Therefore, if IL-2 is an option, it should be given upfront, with ipilimumab reserved for second-line therapy. Another benefit to using IL-2 first is that responses are generally evaluable sooner than in patients treated with ipilimumab (8-10 weeks vs. 18-24 weeks). Finally, the long-term complete remission rate of IL-2 is better established than for ipilimumab.
Activating CKIT mutations, primarily in exon 11, predict sensitivity to imatinib. Although infrequent in cutaneous melanoma (<2%), they do occur in up to 20 percent of mucosal or acral (nail beds and soles of feet) melanomas. In these rare melanoma subtypes, molecular profiling for CKIT mutation should be strongly considered.
Unfortunately, the clinical activity of imatinib in melanoma does not appear to be as high as in gastrointestinal stromal tumor (GIST), the other major CKIT-mutant tumor. Imatinib may cause objective responses in 16 to 23 percent of CKIT-mutant melanoma with some patients achieving disease control for more than a year.
The clinical benefit is considerably better if patients with exon 11 mutations in CKIT are selected. Cases demonstrating responses to sunitinib and dasatinib have also been reported.
Chemotherapy was traditionally a mainstay in melanoma therapy although response rates are very low (less than 10%) and no improvements in overall survival have been demonstrated. Dacarbazine and temozolomide have been shown to be equivalent, with occasional durable disease control. We occasionally use cytotoxic chemotherapy in selected patients with no other treatment options.
Some patients may present with a limited number of metastases, which may be amenable to surgical resection. We favor an aggressive surgical approach in these cases as approximately one third of patients with complete resections may experience long-term survival.
This decision is influenced by the number of metastases, the organ site, time from prior surgery or disease-free period, and the number of prior surgeries. In some cases we will give systemic therapy for a short period of three to six months prior to the surgical resection. This may provide an opportunity to make certain that other lesions are not likely to occur rapidly, and to treat microscopic disease while treating measurable or evaluable disease.
Surgery may also play a role in patients treated with immune therapy who experience regression of all their sites of disease other than one or two lesions. Resection of these persistent lesions may cause “surgical complete responses.”
Anti PD-1/PD-L1 Therapy – Future Directions
Novel immune checkpoint inhibitors targeting the programmed cell death-1 receptor and its ligand (PD-1/PD-L1) are currently in clinical trials (nivolumab, lambrolizumab, MPDL3280A). These promising agents have similar mechanisms of action to ipilimumab and have demonstrated objective response rates in the range of 30 to 50 percent, many of which appear durable.
Anti-PD-1/PD-L1 therapies are well tolerated, with pneumonitis emerging as the most concerning toxicity. Although these treatments are not currently approved, patients should be considered for referral to a clinical trial after failure of approved agents or even for first-line therapy.
Wednesday, July 10, 2013
BY ALLAN LIPTON, MD
Professor of Medicine
Penn State Hershey Cancer Institute
The skeleton is the most common site of metastases in patients with epithelial malignancies. Worldwide, it is estimated that 1.5 million patients with cancer have bone metastases. In the U.S. alone, approximately 400,000 patients are diagnosed with bone metastases annually.
Bone metastases are commonly associated with breast, prostate, lung, and renal cancers. Postmortem examinations of patients dying of cancer confirm that 75 percent of patients with breast or prostate cancer and nearly 40 percent of patients with lung and renal cancers have evidence of metastatic bone disease. Common sites of bone metastases are the ribs, spine, pelvis, skull, and proximal limbs.
Apart from hypercalcemia, patients with bone metastases are also at risk for potentially debilitating skeletal-related events (SREs). Those associated with bone metastases include pathologic fracture, use of palliative radiation therapy, need for surgery to bone to treat or prevent an impending pathologic fracture, and spinal cord compression.
Approximately two-thirds of patients with malignant bone lesions develop at least one SRE if they do not receive bone-directed therapy. Although the rate of SREs is highly variable, patients generally experience a larger number of SREs as their disease progresses. Moreover, patients who have one SRE are at increased risk for subsequent SREs.
Patients with metastatic bone disease live for a significant time after their first SRE. Cancer patients who experience a pathologic fracture are at significantly increased risk of death in the next 12 months compared with patients who do not experience a pathologic fracture.
Healthy bone is in a state of equilibrium in which the processes of bone deposition and bone resorption are normally balanced. In normal bone development, remodeling is an ongoing and ordered process regulated by two different cell types with opposing functions: osteoclasts, which resorb bone; and osteoblasts, which deposit bone.
The interaction of receptor activator of nuclear factor kappa B ligand (RANKL) with the RANK receptor is pivotal for the regulation of osteoclast-mediated bone resorption during normal bone remodeling. The binding of RANKL to its receptor, RANK, activates downstream signaling pathways that drive the differentiation, survival, and function of osteoclasts. The interaction of RANKL and RANK is blocked by osteoprotegerin (OPG), an endogenous decoy receptor of RANKL, secreted by osteoblasts. By interfering with the interaction of RANKL/RANK, OPG limits osteoclastogenesis, resulting in an increase in bone density.
At the site of bone metastases there is locally increased bone remodeling and increased osteoclast activity. The “vicious cycle” hypothesizes that tumor cells interact with the bone marrow microenvironment to drive bone destruction and tumor growth in a symbiotic relationship. The tumor cells secrete various factors that stimulate production of RANKL. In turn, the increased expression of RANKL in the tumor environment leads to increased formation, activation, and survival of osteoclasts and resulting osteolytic lesions. Osteolysis then leads to the release of growth factors derived from bone (e.g., TGF-beta) and these factors promote tumor growth.
Bisphosphonates have a high affinity for bone and readily bind to bone. After binding, bisphosphonates are taken up by actively resorbing osteoclasts. The bisphosphonate induces apoptosis of osteoclasts by inhibition of key pathways for osteoclast survival. Multiple studies have demonstrated the efficacy of several bisphosphonates in patients with metastatic bone disease.
Two potent, nitrogen-containing bisphosphonates are approved for the treatment of bone metastases. Pamidronate, first approved on the basis of two placebo-controlled trials, shown to reduce skeletal complications in patients with breast cancer and SREs in patients with multiple myeloma.
Zoledronic acid was evaluated in a head-to-head comparative trial versus pamidronate in patients with breast cancer or multiple myeloma. The primary endpoint was the proportion of patients with at least one SRE, and zoledronic acid was found to be non-inferior to pamidronate for this endpoint.
Analyses of protocol-defined patient strata revealed that in breast cancer patients receiving hormonal therapy, treatment with zoledronic acid significantly prolonged the time to first SRE (415 days) compared with pamidronate (370 days). In this patient population, zoledronic acid reduced the skeletal morbidity rate from 1.37 to 0.83 SREs/year, and the risk of developing any SRE, including hypercalcemia, by an additional 20 percent versus pamidronate in exploratory multiple event analyses in the subset of patients with breast cancer.
Phase III trials of zoledronic acid vs. placebo were also conducted in prostate cancer and other solid tumors (excluding breast and prostate cancer). In both studies, zoledronic acid decreased the risk for all types of SREs, as well as reducing the skeletal morbidity. Also, in both studies zoledronic acid significantly delayed the time to first SRE.
Despite the availability of bisphosphonates, approximately 40 percent of patients with bone metastases continue to experience skeletal complications. Bisphosphonates have also been associated with deterioration of renal function.
Denosumab is a fully human monoclonal antibody to RANKL, the key mediator of osteoclast formation, function, and survival. In patients with bone metastases denosumab (at 120 mg) is administered as a single subcutaneous injection once every four weeks. As denosumab is a fully human monoclonal antibody, it is cleared via the reticuloendothelial system and does not require renal monitoring.
Denosumab was directly compared with zoledronic acid in three identically designed Phase III trials. Results from the breast cancer trial showed that denosumab was superior to zoledronic acid in delaying both the time to a first on-study SRE as well as the time to a first-or-subsequent SRE. Similar results were also observed in the prostate cancer trial and in the other solid tumor trial (not including the myeloma patients).
The identical design of these three trials allowed the patient level data to be combined to assess the efficacy of denosumab across a broad range of tumor types. Denosumab was superior to zoledronic acid in delaying the time to first on-study SRE (HR = 0.83; p value less than 0.001), with a median delay of more than eight months in the time to first on-study SRE for patients who received denosumab versus those who received zoledronic acid. Furthermore, denosumab was also superior to zoledronic acid in delaying the time to first and subsequent SREs (HR = 0.82; p value less than 0.001).
Safety data from the three denosumab clinical trials demonstrated a decrease in acute phase reactions and less renal toxicity in patients treated with denosumab. Hypocalcemia occurs more frequently with denosumab -- especially in patients who did not take the recommended daily supplements of calcium and vitamin D.
Osteonecrosis of the jaw (ONJ), which was monitored and adjudicated in the three Phase III denosumab trials, was found to occur infrequently, with a similar incidence in the treatment groups -- 1.4 percent in the patients receiving zoledronic acid and 1.8 percent in those receiving denosumab.
In practice, my approach is to treat all solid tumor patients with bone metastases with denosumab, and all myeloma patients with bone involvement with zoledronic acid.
In summary, malignant bone lesions from solid tumors or multiple myeloma are associated with significant morbidity, decrease in quality of life, and survival. Over the past two decades zoledronic acid has allowed us to successfully treat hypercalcemia. The incidence of skeletal-related events has been decreased by approximately two-thirds by the use of zoledronic and subsequently denosumab. However, since neither bone-targeted agent has demonstrated an effect on progression-free or overall survival, there is still much that needs to be accomplished in our treatment of these patients.