Skip Navigation LinksHome > Blogs > How Do I Treat…?
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.

Friday, May 29, 2015

Professor of Pediatrics and Molecular Pharmacology,

The Albert Einstein College of Medicine of Yeshiva University;

Vice Chairman of Pediatrics and Division Chief of Pediatric Hematology, Oncology and Marrow and Blood Cell Transplantation,  The Children’s Hospital at Montefiore,  Bronx, NY


All cancers in children and young adults are fortunately rare. Even in their period of most frequent incidence, during the second decade of life, bone tumors are not the most common malignancy of childhood. For sarcoma oncologists, most patients referred to your practice have a high enough probability of having cancer that everyone can be more thoroughly assessed, which is not the case for primary care providers.


It can be very challenging for the pediatrician and orthopedic surgeons to consider a bone cancer as a possible diagnosis in the myriad patients who present complaining of pain or a mass more often resulting from benign etiologies. Trauma, growing pains, and osteomyelitis are all more common than bone cancer.


The advice given to primary care providers is that in patients with persistence of pain or the association of the pain with a mass, particularly when symptoms are becoming more severe over time, further evaluation with a plain radiograph of the site is warranted.


The two most common bone cancers in younger patients are osteosarcoma and Ewing sarcoma family tumors, with approximately 400 and 200 new pediatric cases per year, respectively, in North America. Although they can occur in any bone in the body, approximately half of osteosarcomas occur in the region around the knee. The majority of Ewing sarcoma family tumors occur in the appendicular skeleton even though this entity’s tropism away from the axial skeleton is not as pronounced as for osteosarcoma. This, coupled with the age distribution of these cancers, dictates a higher level of concern for malignancy in teenagers with symptoms and physical exam findings around their knees.


Once a plain radiograph is obtained, bone tumors usually have a dramatic appearance, and a completely negative radiograph is reassuring that a cancer is not present. Osteosarcomas often have soft tissue extension with a periosteal reaction that is classically described as having a “sun-burst” appearance in the metaphyseal region of the bone.


Ewing sarcoma family tumors are permeative, leading to a periosteal reaction that is lamellar or “onion skinning” in appearance in the diaphyseal or metaphyseal region of bones. Occasionally patients with bone cancer will present more dramatically with pathological fractures, but even in some of these cases, a reported history of trauma may muddy the picture, leading to confusion and misdiagnosis, so a high index of suspicion always needs to be maintained to avoid delays in diagnosis.


Once a diagnosis of a bone cancer is suspected prompt referral to a group with expertise in sarcomas is strongly recommended as these patients require complex multidisciplinary care.


Multidisciplinary Sarcoma Care

All bone cancers require multidisciplinary care by physicians with expertise in their management. For this purpose, all new patients seen at our sarcoma center are seen right from the first encounter by both a pediatric/medical oncologist and an orthopedic oncologist, at the same time, in shared clinic space.


A history and physical examination along with plain radiographs dictates further evaluation. One needs to consider the benign entities that can mimic more aggressive lesions as well as variant osteosarcoma and Ewing sarcoma family tumor appearances, which can misleadingly suggest a benign lesion.


In most patients at least additional imaging is warranted, which is typically magnetic resonance imaging of the site, with and without gadolinium contrast including the entire bone to capture skip lesions that may be present. This is best done prior to the biopsy to assist in decision making and biopsy placement and so that the appearance of the lesion prior to its disruption by the biopsy can be appreciated.


In patients strongly suspected of having osteosarcoma or a Ewing sarcoma family tumor, high-resolution chest computed tomography is obtained prior to biopsy to avoid issues of postoperative atelectasis creating challenges in interpretation of findings within the lungs, as this is the site in which metastases are most likely to occur.


In all patients being biopsied at least a chest radiograph is warranted. Elevated alkaline phosphatase and lactate dehydrogenase supports a probable diagnosis of a bone cancer.


I am a strong advocate of open biopsies, but many institutions have core needle biopsies performed by interventional radiologists or orthopedic surgeons with this topic being immensely controversial.


Universally endorsed is the belief that the biopsy needs to be performed by someone with experience in resecting bone cancer, with the critical issue being consideration of the surgical plan for definitive resection. If the pathology reveals a cancer of bone, the needle, or incision tract will need to be removed along with the tumor en block, to abrogate the risk of seeding the site with tumor cells.


Osteosarcoma is diagnosed by a bone tumor pathologist based on the histologic appearance of a malignant spindle cell tumor that produces osteoid. Ewing sarcoma family tumors are small round blue cell tumors which typically express CD99 and possess one of a number of recurrent chromosomal translocations most commonly involving the EWS and FLI genes on chromosomes 11 and 22 detected via FISH or PCR and are disease defining.


Once the diagnosis is made, the staging workup is completed with a technetium bone scan, and additionally, in the case of Ewing sarcoma family tumors multiple site bone marrow aspirates and biopsies.


The role of PET scans in routine clinical practice for bone sarcomas remains undefined. A central venous catheter is required for the types of systemic chemotherapy that are typically utilized, and additional pre-treatment assessments of organ function are performed appropriate to the planned treatment. Fertility preservation methods need to be discussed with all patients.


Systemic and Local Therapy

The standard of care for all patients with high-grade osteosarcoma and Ewing sarcoma family tumors is the use of neoadjuvant chemotherapy, followed by local control and subsequently additional adjuvant chemotherapy. Systemic therapy is needed because of the high frequency of micro-metastases in patients with radiographically localized disease.


Routine clinical practice for younger patients with bone cancers is heavily dictated by Children’s Oncology Group clinical trials. If a Phase III trial is active for newly diagnosed patients offering participation is routine. In the absence of a front-line study the superior arm of the last Phase III trial is used typically as standard therapy.


For osteosarcoma patients, front-line studies are not active at present, with the standard arm of the last international Phase III study, the EURAMOS study, being superior to the experimental arms. As such, the standard induction and adjuvant chemotherapy is comprised of cisplatin and doxorubicin alternating with two consecutive weeks of high-dose methotrexate in accordance with the schema in the EURAMOS study.


Local control requires a complete surgical resection of the tumor as osteosarcoma is relatively radiation resistant. These resections typically do not require amputation, and considerable advances have been made in the surgery as well as the typically metallic internal prostheses yielding numerous functional enhancements.


The field of orthopedic oncology is highly specialized, interesting, and beyond the scope of the present discussion. The degree of necrosis in the resected tumor samples is prognostic but thus far has been shown to be of no clinical value for tailoring subsequent therapy.  In patients with osteosarcoma and pulmonary metastases the nodules need to be resected. In our institution this is performed as bilateral staged thoracotomies but considerable variation in practice exists.


For patients with newly diagnosed Ewing sarcoma family tumors, active Children’s Oncology Group clinical trials are ongoing. All patients are given intensively timed cyclophosphamide, doxorubicin, and vincristine alternating with ifosfamide and etoposide as this resulted in superior survival in the last Phase III trial and represents the standard of care.


In patients with localized disease the randomized question is whether the addition of cyclophosphamide, topotecan, and vincristine improves survival. In patients presenting with metastatic disease the randomized question is whether the addition of ganitumab, an antibody to IGF-1R, improves survival. Local control is performed after neoadjuvant chemotherapy but, unlike osteosarcoma, can involve either surgery or radiation therapy or both. Whether a patient receives surgery or radiation therapy frequently is decided based on the functional consequences and late effects of each. Considerable variation exists in institutional practice. In patients with metastatic disease radiation therapy is more frequently utilized.


Care of Patients with Recurrent Osteosarcoma and Ewing Sarcoma

Approximately 60 to 70 percent of patients presenting with localized high-grade osteosarcoma or Ewing sarcoma family tumors will have long-term disease-free survival with the aforementioned therapy. For the 20 percent of patients who present with radiographically visible metastatic disease, the prognosis is markedly worse and further influenced by the extent of the metastatic disease. Similarly the prognosis of patients with recurrent disease is poor.


In patients with recurrent osteosarcoma, metastectomy has been shown to be of clinical benefit with a small percentage of patients having durable disease-free survival with resection alone. The value of salvage chemotherapy, typically ifosfamide and etoposide is controversial. In Ewing sarcoma family tumors radiation therapy is employed for local control with cyclophosphamide, topotecan, and vincristine or temozolomide, irinotecan, and vincristine being the most typical salvage therapies.


Numerous clinical trials of novel agents are being performed in patients with recurrent osteosarcoma and Ewing sarcoma family tumors, and participation in these trials can be offered. It is hoped that some of these ever-changing menu of agents will prove effective in the therapy of patients with recurrent bone cancer, and will improve survival when added to upfront therapy of newly diagnosed patients, which will need to be proven in the context of future randomized Phase III trials.

Tuesday, February 24, 2015

BY Sandy Srinivas, MD

Associate Professor of Medicine (Oncology)

Stanford University Medical Center


Prostate cancer spans a spectrum like all cancers, ranging from localized disease to metastatic disease. As with breast cancer, prostate cancer is exquisitely sensitive to hormonal manipulation, and from a practical point of view is divided into hormone-naïve versus castrate-resistant disease.


Recurrent Prostate Cancer Post-localized Disease

Recurrent post-localized prostate cancer is usually manifested by a climbing serum prostate-specific antigen (PSA) after therapy for localized disease. Typically, I stage patients with a CT scan and bone imaging, either with a bone scan or a F18 PET scan. When PSA doubling time (PSA DT) becomes less than six months, androgen-deprivation therapy (ADT) is initiated.


In general, if scans show no metastases, I start patients on LHRH analogs such as leuprolide or gosserlin. There are LHRH antagonists commercially available that can be used as well. I tend to offer patients intermittent ADT as long as patients appear to be reliable and motivated for follow-ups.


I do not use a specific PSA level to reinitiate therapy but rather look at the rate at which the PSA rises. Every time it takes a sharp rise, I tend to reimage, and do not do scans on a set schedule.


Metastatic Prostate Cancer

Due to the current controversies about PSA screening, we are seeing more patients with newly diagnosed metastatic prostate cancer now than five years ago. Based on the CHARTERED trial, which showed a significant survival advantage to the use of chemotherapy along with ADT, I do offer docetaxel x 6 cycles in addition to ADT for these patients with bone or visceral metastases.


For patients with recurrent prostate cancer following localized disease, I start them on ADT. I offer intermittent ADT even in patients with metastatic disease after having a discussion with patients regarding their goals and tolerability to ADT.


Castrate-Resistant Disease

Ultimately, castrate resistance--defined as a rising serum PSA despite low testosterone--develops in almost all patients starting ADT. I will reimage them to set a new baseline, and if the patient remains non-metastatic, I would use drugs such as bicalutamide first. In general responses are variable, and if serum PSA rises, I will initiate anti-androgen withdrawal and then reimage again. If patients are non-metastatic, I would pursue older drugs such as nilutamide or even ketoconazole/hydrocortisone--surveillance alone may also be appropriate in certain cases.


Metastatic Castrate-Resistant Prostate Cancer

Management of patients with metastatic castrate-resistant prostate cancer has become challenging, as there are several FDA-approved drugs for this group of patients. These include sipuleucel-T, abiraterone, enzlautamide, radium 223, as well as chemotherapy. I use Provenge or sipuleucel-T when patients have slow-rising PSA with long PSA DT and when they are asymptomatic and not too concerned about PSA values.


I use enzalutamide if patients have issues with diabetes or poorly controlled hypertension and abiraterone with prednisone if they have pain and would benefit from the prednisone use. Typically I do not use these drugs back to back, as there is a very low chance of response.


After failure of either enzalutamide or abiraterone, I re-scan. If patients have bone-only disease, I would recommend alpharadin (radium-223, Xofigo) for six cycles. If patients have bulky nodal disease or visceral disease, I would recommend chemotherapy with docetaxel.


I typically stop chemotherapy after six to eight cycles and give patients a chemotherapy holiday. I will treat with additional hormonal agents such as abiraterone or enzalutamide if patients have never received that drug. For patients progressing after this, I will re-treat with docetaxel if it has been more than a year since the last treatment; otherwise I recommend cabazitaxel.


While mitoxantrone is an old drug, I still use it in patients who have progressed on taxanes or those who have had an allergic reaction to taxanes.


Bone Health

Bone health is a vital part of the care of patients with prostate cancer. I use bone-modifying drugs in patients with castrate metastatic prostate cancer but not in those with newly diagnosed bone metastases. A randomized trial from CALGB did not support the use of zoledronic acid in newly diagnosed metastatic prostate cancer patients. I will use zoledronic acid or denosumab every three months along with a GNRH analog, which I continue as well.

Wednesday, July 09, 2014

By Elena Elimova, MD, MSC; Brian Badgwell, MD; Prajnan Das, MD, MS, MPH; Jeannelyn Estrella, MD; Aurelio Matamoros Jr., MD; and Jaffer A. Ajani, MD


Gastric cancer represents a serious health problem on a global scale. It is the second leading cause of cancer-related death worldwide. Novel therapeutic targets are desperately needed because the meager improvement in the cure rate of about 10 percent realized by adjunctive treatments to surgery is unacceptable as more than 50 percent of patients with localized gastric cancer succumb to their disease.


Either postoperative chemoradiotherapy (in the United States), pre-and post-operative chemotherapy (in Europe), and adjuvant chemotherapy after a D2 resection (in Asia) can all be regarded as standards of care in the localized management of the disease. For patients with metastatic disease, the addition of trastuzumab to chemotherapy is standard of care in HER2-positive disease. In the HER2-negative population, the treatments remain limited.


In the first-line setting, the standard of care is a combination of fluoropyrimidine and platinum-containing chemotherapy, with or without epirubicin or docetaxel. Finally there is a minimal overall survival benefit in treating patients with metastatic disease in the second-line setting, with either irinotecan, docetaxel, or ramucirumab with or without chemotherapy.


Our approach to the treatment of patients with gastric cancer begins with appropriate clinical staging to determine if the cancer is localized or advanced. This involves full imaging, including CT of the chest/abdomen/pelvis, PET-CT, endoscopic ultrasound (EUS), and finally a staging laparoscopy.


EUS is the most reliable nonsurgical method to evaluate the depth of invasion, with concurrent evaluation of regional lymph nodes of primary gastric cancers and is therefore instrumental; however, things are not as simple as doing an EUS because this technique is really only useful in the hands of a skilled operator.


The PET/CT scan is most useful in detecting occult distant metastasis, thereby helping avoid high morbidity surgery in a sub-group of patients. Laparoscopy should be considered for patients who appear to have locoregional disease (other than stage IV, Tis or T1a stage) after conventional radiographic and EUS staging. However, because it is sometimes difficult to differentiate T2 and T3 lesions on EUS, it is reasonable to perform a laparoscopy for any medically fit patient who appears to have more than a T1 lesion on EUS, no histologic confirmation of stage IV disease, and who would not otherwise require a palliative gastrectomy because of symptoms. This is because 20 to 30 percent of patients with greater than T1 EUS disease will be found to have peritoneal metastasis despite having negative CT and PET scans.


In our center the clinical staging is followed by a multidisciplinary discussion in all localized gastric cancer cases:


Localized Gastric Cancer

In terms of localized gastric cancer, a curative resection (R0) offers the best chance of cure, and is best managed at high-volume centers and by high-volume surgeons.  We strongly believe that a multidisciplinary approach and preoperative therapy is the cornerstone of management in the West.


Although gastrectomy is the recommended treatment in relatively early-localized gastric cancer (T1b), in more advanced disease (T2N0, T1aN+, or T1b-T3N+) we recommend adjunctive therapy in addition to gastrectomy. As previously mentioned, postoperative chemoradiotherapy (United States), pre-and post-operative chemotherapy (Europe), and adjuvant chemotherapy after a D2 resection (Asia) can all be regarded as standards of care in the management of localized gastric cancer. However, at our institution, we use a combination of these approaches in the pre-operative setting, because in our experience post-operative therapy is much harder to deliver. 


Our general approach is to start with pre-operative chemotherapy with a platinum-based doublet or triplet (depending on the performance status of the patient) for two to three months, followed by chemoradiation also given preoperatively with 5-fluorouracil ± taxanes or platinum, finally followed by surgery. This approach is based on Phase II study data and results in R0 resection in 70 to 78 percent of our patients and five-year overall survival rates at least comparable to those reported in the Intergroup 0116 and MAGIC clinical trials.


In terms of post-treatment surveillance, there is no data to provide guidance, and arbitrary surveillance strategies are used.


Metastatic Gastric Cancer

In terms of our approach to metastatic gastric cancer, clearly in the context that this is no longer a curative situation our approach is to the palliation of symptoms and prolongation of life:

Otherwise, we would treat differently based on HER2 status. Clearly in HER2-positive gastric cancer there is an overall survival benefit to the addition of anti-HER2 therapy to first line chemotherapy. 


It is our practice to typically use trastuzumab and not lapatinib because of the negative results of the lapatinib trial in combination with platinum-based doublet. Although no convincing data exists as to the benefit of the addition of HER2 therapy in gastric cancer, we extrapolate from the breast cancer trials and continue anti-HER2 therapy beyond progression, typically switching to an alternative agent such as pertuzumab.


In the context of HER2-negative metastatic disease, our options continue to be limited. In a select subgroup of patients who have small-volume disease and who are asymptomatic, a careful watch-and-wait strategy is reasonable as long as the patient is comfortable with this approach. In symptomatic patients, a reasonable option in the first-line setting is a platinum-based doublet with the addition of docetaxel or epirubicin, depending on the performance status of the patient or clinical trials.


In the second-line setting, we often use irinotecan-based doublets, but with the recent approval of ramucirumab, this agent in combination with chemotherapy will have a role in our practice.   


Genetic Profiling

Genetic profiling of tumors is becoming a more widely used tool in the treatment of gastric cancer, as it is in other cancers. Patients are often found to have multiple and even more often non-targetable mutations. Even when a potentially targetable mutation is found and the patient is treated with a given drug, we have found that responses are rare--likely because of our poor knowledge of driver mutations. Therefore we do not consider a genetic evaluation a critical part of treatment, but rather emphasize the enrollment of patients into available clinical trials.


In Summary

In summary, we strongly feel that all patients with localized gastric cancer in a potentially curable situation should be evaluated in a multidisciplinary way and in a high-volume center, so that they can have the benefit of a surgery performed by a high-volume surgeon.  In metastatic disease at our institution we put emphasis of enrollment of patients on clinical trials in hopes of improving outcomes. 



The authors are all from the University of Texas MD Anderson Cancer Center: Elena Elimova, MD, MSC, Department of Gastrointestinal Medical Oncology; Brian Badgwell, MD, Department of Surgical Oncology; Prajnan Das, MD, MS, MPH, Department of Radiation Oncology; Jeannelyn Estrella, md, Department of Pathology; Aurelio Matamoros Jr., md, Department of Diagnostic Radiology; and Jaffer A. Ajani, md, Department of Gastrointestinal Medical Oncology.

Saturday, July 05, 2014


BY Geoffrey R. Oxnard, MD

Assistant Professor of Medicine

Dana-Farber Cancer Institute

Harvard Medical School


T790M is a point mutation in the EGFR gene that is associated with resistance to epidermal growth factor receptor (EGFR) kinase inhibitors like erlotinib and gefitinib. Some of the most exciting results presented at this year’s American Society of Clinical Oncology Annual Meeting have to do with new inhibitors targeting the EGFR T790M mutation. Given that this appears to be an emerging biomarker in the treatment of lung cancer patients, it is worth reviewing the management of patients carrying this mutation. Importantly, it can be seen in several different clinical circumstances where it can mean different things.


Baseline EGFR T790M

The EGFR T790M mutation is rarely seen in a lung cancer at initial diagnosis, prior to treatment with an EGFR kinase inhibitor. The prevalence of baseline T790M is debated in the scientific literature, but using conventional testing methods it is generally though to occur in one to two percent of all EGFR-mutant lung cancers. When seen in this setting, it is most commonly identified in addition to a second drug-sensitive EGFR mutation. Despite that, lung cancers with baseline EGFR T790M are unlikely to respond to standard EGFR kinase inhibitors and should be treated in the first-line setting with standard chemotherapy. Interestingly, the presence of baseline T790M indicates a high likelihood of an underlying germline T790M mutation, discussed further below.


Acquired EGFR T790M

The more common setting where the T790M mutation is seen is as an acquired mutation in EGFR-mutant lung cancer after treatment with an EGFR kinase inhibitor. A repeat tumor biopsy in this setting can identify a new T790M mutation more than half the time. Previous data has suggested that T790M-mediated resistance can be associated with an indolent growth and a relatively favorable prognosis when compared with other types of resistance. Interestingly, there are some types of that resistance less likely to carry T790M – such as progression in the brain only, or recurrence after stopping adjuvant erlotinib – suggesting that these situations may not be fully resistant, and that further treatment with an EGFR kinase inhibitor might make sense.


While there was hope that irreversible EGFR kinase inhibitors like afatinib or dacomitinib might inhibit T790M, response rates to these drugs have been low in patients with resistance to erlotinib or gefitinib. A higher response rate of 30 percent was reported with afatinib plus the EGFR antibody cetuximab, though this activity was seen both in T790M positive and negative resistance.


At ASCO this year, data were presented regarding a new class of drugs called “mutant-selective irreversible EGFR kinase inhibitors” which target T790M-mediated drug resistance. These drugs potently inhibit mutant EGFR protein without inhibiting wildtype EGFR, aiming to induce responses while avoiding EGFR-related toxicities. Data from three trials of three different drugs in this class (AZD9291, CO-1686, and HM61713) were presented, and each drug reported dramatic tumor responses in patients with EGFR-mutant lung cancers after resistance to standard EGFR kinase inhibitors.


In the largest trial, studying AZD9291, a striking difference in activity was seen between tumors with T790M-mediated resistance (a 65% response rate) and those with T790M-negative resistance (a 22% response rate). Additional phase II data will be needed to better understand any differences in activity between these drugs, but T790M does appear to be an emerging biomarker suggesting drug sensitivity.


Clinical trials are ongoing around the world (NCT01802632; NCT01526928; NCT01588145) and are an attractive alternative to standard chemotherapy for these patients.


Germline EGFR T790M

The rarest setting where EGFR T790M can be seen is as a germline mutation where it has been found to be associated with familial lung cancer, particularly in non-smokers. However, the risk of lung cancer in healthy individuals carrying such an inherited mutation is not well understood. Given how rare this condition is – associated with less than 1 in 1000 lung cancers – germline testing is not widely available and is not part of standard practice. In my practice, I only test for germline EGFR T790M when a patient presents with baseline EGFR T790M, a setting where the prevalence of germline mutations is estimated at approximately 50 percent.


To better understand this condition, and to offer patients free genetic counseling and germline testing, my institution has teamed up with the Addario Lung Cancer Medical Institute (ALCMI) to open a prospective trial titled INHERIT: Investigating Hereditary Risk from T790M. Individuals can present to the study website for more information: If a lung cancer patient harboring baseline T790M undergoes germline testing and is found to be positive for an inherited mutation, they can then invite their relatives to be tested, allowing study of entire families. If we can demonstrate that these families are at a high risk of lung cancer, then perhaps they should be undergoing CT-screening much as is recommended for individuals with a significant smoking history.

Sunday, May 11, 2014




Globally esophageal cancer is the eighth leading cause of cancer-related death. Although squamous cancer is more common in the East, in Western Europe and the United States adenocarcinoma has emerged as the most common histology. The rate of increase of adenocarcinoma of the esophagus and gastroesophageal junction (GEJ) has plateaued in recent years. These diseases continue to increase in incidence, and will overtake gastric cancer as the more common cancer of the upper gastrointestinal tract in the West.


Potential explanations for this increase include Western population trends of obesity, which may increase the rate of esophageal reflux disease; tobacco use; and the disappearance of Helicobacter pylori (HP) infection from Western populations. The decline in HP infection likely explains the decline in incidence of gastric cancer.  Paradoxically, because HP infection may lead to atrophic gastritis and reduced stomach acid production, the disappearance of HP may actually lead to a population-wide increase in gastroesophageal reflux disease.


Screening for esophageal cancer is not feasible given that it occurs at less than 10 to 20 percent the incidence of more common cancers such as breast, prostate, and colorectal cancer. Endoscopic assessment is recommended for patients with chronic reflux, and patients identified with Barrett’s esophagus should undergo some form of regular surveillance endoscopy.


Recent population-based series indicate that, even in higher-risk patients with chronic reflux and Barrett’s esophagus, the rate of developing a cancer over 10 years may be even less than the reported one to two percent of patients. Only five percent of cases of esophageal cancer are detected by screening. Aspirin in conjunction with proton pump inhibitor therapy is the subject of ongoing trials as a cancer preventive therapy in Barrett’s esophagus. Use of life-long proton pump inhibitor therapy, however, is recommended in patients with Barrett’s esophagus as this may slow progression to dysplastic Barrett’s. 


For Barrett’s esophagus with high-grade dysplasia, a population at high risk to develop esophageal adenocarcinoma in the short term, radiofrequency ablation (RFA) has replaced surgical resection as the therapy of choice. For early in situ cancers or early stage T1a cancers in the setting of Barrett’s esophagus, RFA can be combined with endoscopic mucosal resection as primary therapy.


Recent studies suggest there may be some familial predilection for both Barrett’s esophagus and esophageal adenocarcinoma, accounting for up to 10 to 15 percent of cases. Candidate genetic biomarkers, however, require further study and validation prior to routine clinical application. Esophageal or GEJ adenocarcinoma seen in the context of familial colorectal, uterine, or ovarian cancer should raise suspicion for Lynch Syndrome, which requires the identification of gene carriers who require specialized screening and follow-up.


Advanced Disease Therapy:  Molecular Targets and the Emergence of Targeted Therapy

Chemotherapy for advanced esophageal and GEJ cancers has modest benefit, with responses in 30 to 40 percent of patients and a median survival of nine to 10 months. Standard chemotherapy combines a fluorinated pyrimidine and a platinum agent. The combination of (1) capecitabine with oxaliplatin or cisplatin, or (2) 5-FU and oxaliplatin on the mFOLFOX6 regimen, has emerged as global standards.  


My practice in most patients is to use FOLFOX, given the chronicity of chemotherapy and the cumulative cutaneous toxicity of capecitabine. Triplet therapy adding a taxane to 5-FU and cisplatin adds modest benefits in response and survival at the cost of escalated toxicity; its use should be reserved for younger, higher functional status patients willing to tolerate greater treatment-related toxicity. Epirubicin added to flourinated pyrimidine and platinum therapy may be no better than FOLFOX, as seen in the recent CALGB 80403 Trial. Second line chemotherapy has now been validated to modestly improve survival, with data supporting the use of taxanes or irinotecan after disease progression on first-line chemotherapy.


The only validated molecular target in esophagogastric adenocarcinoma is HER2, with the recent TOGA trial showing enhanced response and survival when trastuzumab was combined with first-line chemotherapy in HER2-positive patients. All patients at diagnosis of esophagogastric adenocarcinoma should have documentation of HER2 status.


Standard practice is to perform IHC testing and if 3+, we declare the patient HER2 positive. If IHC 2+, we perform FISH testing and if positive, we consider HER2 positive. We consider IHC 0-1+ patients HER2 negative and do not perform FISH testing. Preliminary reports of phase III trials of the EGFr/HER2 tyrosine kinase inhibitor lapatinib, combined with first- and second-line chemotherapy in HER2-positive esophagogastric cancer, failed to improve overall survival.


Ongoing trials of novel HER2-targeted therapies include a randomized trial in first-line comparing chemotherapy and trastuzumab with or without the HER2-3 targeted agent pertuzumab (the JACOB Trial). Second-line paclitaxel is being compared with the trastuzumab conjugate agent TDM-1 after progression on trastuzumab-based first-line chemotherapy (the GATSBY Trial).      


The positive results for the TOGA trial indicate the need to select patients with a biomarker predictive of a potential greater benefit from a new agent. This has been made painfully clear by recent large phase III trials of new agents in unselected patient populations. Multiple trials of EGFr targeted agents, including first-line trials combining chemotherapy with either cetuximab or panitumumab, or later-line therapy with gefitinib, failed to improve outcome. 


Unlike colorectal cancer, where RAS mutation is commonly present and is predictive of resistance to EGFr therapy, no such biomarker for EGFr has been identified in esophagogastric cancer. A recent genomic analysis of a large series of esophagogastric adenocarcinomas indicated that, in contrast to colorectal cancer and non-small cell lung cancer, mutations in RAS, BRAF, and EGFr were rarely seen. On the other hand, a key role for gene amplification was seen, with nearly 40 percent of cancers having amplification in at least one of five important pathways: EGFr, HER2, MET, FGF, and KRAS.   


Two large phase III trials are now evaluating the addition of MET-targeted therapies to first-line chemotherapy in MET-positive patients by IHC, including rilotumumab (which targets the MET receptor ligand hepatocyte growth factor) and onartuzumab (which blocks the MET receptor).   


The large AVAGAST trial adding bevacizumab to first-line chemotherapy failed to improve survival, despite improvements in progression-free survival and anti-tumor response rate. An improvement in overall survival was limited to Western patients with no improvement in Asian patients. The higher utilization of second- and third-line chemotherapy in Asian patients may have undercut any survival benefit seen.


Resurgence in interest in VEGF-targeted therapy has now occurred with positive results reported for ramucirumab, a monoclonal antibody with blocks the VEGFr2 receptor. The REGARD Trial compared supportive care alone versus supportive care plus ramucirumab in second-line treatment of esophagogastric cancer. All endpoints were improved for ramucirumab, including progression-free and overall survival, leading to FDA approval of ramucirumab as monotherapy. Equally compelling results were recently reported for the RAINBOW trial, which compared second-line therapy with paclitaxel versus paclitaxel plus ramucirumab. Response rate, progression-free, and overall survival were improved with the combination of ramucirumab plus chemotherapy. 


These positive results will likely change the standard of care for second-line chemotherapy in esophagogastric cancer, with the option of ramucirumab monotherapy or the combination with taxane-based chemotherapy. 


Neoadjuvant Chemotherapy or Combined Chemoradiotherapy?  

Endoscopic ultrasound (EUS)-staged T2-3 or node-positive patients are candidates for combined-modality therapy. The predominant approach in the U.S. is combined chemotherapy and radiotherapy followed by surgery, whereas in Europe, preop chemo alone is preferred. The Dutch CROSS trial reported in 2012 treated over 360 patients with EUS-staged esophageal squamous cell and adenocarcinoma.


Combined chemoradiotherapy, using a modern regimen of weekly carboplatin, paclitaxel, and 41.4 Gy of radiotherapy improved median overall survival by nearly a two-year increment over surgery alone. Other positive endpoints included an improved rate of R0 resection from 67 to 92 percent, a pathologic complete response rate of 27 percent, and an improved five-year overall survival of 13 percent. I feel that the CROSS trial established a new therapy standard for preoperative treatment in esophageal cancer.


Recent data from European studies indicate that early response observed on PET scan during preoperative chemotherapy may predict response at surgery and improved survival, and that PET scan non-responders can have preoperative chemotherapy discontinued and referral to immediate surgery without a detriment in outcome.  Based on these observations, CALGB/Alliance Trial 80803 uses early PET scan to assess response to induction chemotherapy in esophageal and GEJ cancer. Patients are assigned randomly to receive either carboplatin/paclitaxel, or FOLFOX treatment. PET scan responding patients continue the same chemotherapy during subsequent radiotherapy; PET non-responding patients cross over to the other chemotherapy regimen. 


The primary endpoint is to increase rates of pathologic complete response in PET scan non-responding patients who change chemotherapy during radiation. 


Support also exists for the use of preop chemo without radiotherapy, but mixed positive and negative results of phase III trials have led to lesser acceptance of this approach in esophageal and GEJ cancers in the U.S.  Positive preop chemo trials include the British MAGIC trial, employing perioperative chemo with ECF, and the French FFCD/FNLC trial using perioperative cisplatin/5-FU. 


These trials, treating both esophageal and gastric adenocarcinoma, demonstrated a 13 to 14 percent improvement in five-year overall survival. The large British MRC OEO2 trial gave preop cisplatin/5-FU to 800 patients with squamous cell and adenocarcinoma of the esophagus. A six percent improvement in five-year overall survival was achieved with preop chemo. The largest negative trial of preop chemo was the U.S. Intergroup 113 trial, which failed to show improvement in any endpoint for preop chemo versus surgery alone. In my practice, I reserve the use of preop chemo for distal gastric cancer, and prefer the use of combined chemoradiotherapy in esophageal and GEJ adenocarcinoma.



DAVID H. ILSON, MD, PHD, is Attending Physician and Member of the Gastrointestinal Oncology Service at Memorial Sloan Kettering Cancer Center and Professor of Medicine at Weill Cornell Medical College; and Chairman of the Intergroup Esophageal and Gastric Cancer Task Force Committee.






About this Blog


Practical perspectives on cancer treatment by thought leaders, explaining how they approach the treatment of a patient in their area of expertise.

As the blog’s Editor, RAMASWAMY GOVINDAN, MD, OT’s Clinical Advisory Editor for Oncology -- Co-Director of the Section of Medical Oncology and Professor of Medicine at Washington University School of Medicine, Alvin J. Siteman Cancer Center -- notes, “While all of us want to see more patients enrolled in well-designed clinical trials, this series is all about how one treats patients “off-protocol” in routine clinical practice. Practice patterns vary, since we do not always have firm data for every single clinical scenario.”

Let us know what you think! Add your comments, both about individual treatment scenarios as well as to suggest future questions.