Mark A. Lewis

Adverse Prognostic Impact of Intratumor Heterogeneous HER2 Gene Amplification in Patients With Esophageal Adenocarcinoma

PURPOSE There is increasing recognition of the existence of intratumoral heterogeneity of the human epidermal growth factor receptor (HER2), which affects interpretation of HER2 positivity in clinical practice and may have implications for patient prognosis and treatment. We determined the frequency and prognostic impact of heterogeneous HER2 gene amplification and polysomy 17 in patients with esophageal adenocarcinoma (EAC). PATIENTS AND METHODS HER2 amplification (by fluorescence in situ hybridization) was examined in surgical EAC specimens (n = 675). HER2 heterogeneity was defined according to consensus guidelines as gene amplification (HER2/CEP17 ratio ≥ 2.0) in more than 5% but less than 50% of cancer cells. No patient received neoadjuvant or HER2-targeted therapy. Cox models were used to assess disease-specific survival (DSS) and overall survival (OS). RESULTS Overall, 117 EACs (17%) demonstrated HER2 amplification, of which 20 (17%) showed HER2 heterogeneity. All HER2-heterogeneous tumors were amplified. Among HER2-amplified tumors, heterogeneous tumors had significantly higher frequency of poor histologic grade and polysomy 17. In multivariable models that included number of metastatic lymph nodes, grade, tumor stage, and polysomy 17, only HER2 heterogeneity and node number were prognostic among HER2-amplified tumors, with heterogeneity showing worse DSS (hazard ratio, 2.04; 95% CI, 1.09 to 3.79; P = .025) and OS (P = .026). Among HER2-nonamplified EACs, polysomy 17 was independently associated with worse DSS (P = .012) and OS (P = .023). CONCLUSION Among HER2-amplified EACs, 17% show HER2 heterogeneity, which independently predicts for worse cancer-specific death. Among HER2-nonamplified EACs, polysomy 17 is independently associated with worse survival. These novel findings demonstrate aggressive subgroups in HER2-amplified and -nonamplified EACs that have important implications for HER2 analysis and determination of benefit from HER2-targeted therapy.

Prognostic impact of body mass index stratified by smoking status in patients with esophageal adenocarcinoma

PURPOSE Given that smoking affects body mass index (BMI) and survival, stratification by smoking status may be required to determine the true prognostic impact of BMI. Although obesity increases risk for developing esophageal adenocarcinoma (EAC), the prognostic influence of obesity and its potential modification by smoking status is unknown in this disease. PATIENTS AND METHODS All patients (N = 778) underwent potentially curative esophagectomy. BMI was calculated using measured height and weight at surgery and categorized as obese (≥ 30 kg/m(2)), overweight (25 to 29.9 kg/m(2)), or normal (18.5 to 24.9 kg/m(2)). Cigarette smoking was categorized as never or ever. The association of BMI with disease-specific survival (DSS), disease-free survival (DFS), and overall survival (OS) was determined by Cox regression. RESULTS Excess BMI was significantly associated with DSS in a manner that differed substantially by smoking status (P for interaction = .023). Among never smokers, obesity was significantly associated with adverse DSS (hazard ratio [HR] = 2.11; 95% CI, 1.31 to 3.43; P = .002), DFS (HR = 2.03; 95% CI, 1.30 to 3.18; P = .002), and OS (HR = 1.97; 95% CI, 1.24 to 3.14; P = .004), as compared with normal weight, after adjusting for covariates. By contrast, among ever smokers, obesity was not prognostic, and overweight status was significantly associated with favorable survival in univariate, but not multivariate, analysis. CONCLUSION Obesity among never smokers was independently associated with two-fold worsening of DSS, DFS, and OS after surgery for EAC, after adjusting for known prognostic factors. These data, in one of the largest reported resected EAC cohorts, are the first to show an adverse prognostic impact of obesity in EAC.

Social Media and Oncology: The Past, Present, and Future of Electronic Communication Between Physician and Patient

The relationship between patient and physician is in flux with the advent of electronic media that are advancing and enhancing communication. We perform a retrospective, current, and forward-looking examination of the technologies by which information is exchanged within the healthcare community. The evolution from e-mail and listservs to blogs and the modern social networks is described, with emphasis on the advantages and pitfalls of each medium, especially in regard to maintaining the standards of privacy and professionalism to which doctors are held accountable. We support the use of contemporary platforms like Twitter and Facebook for physicians to establish themselves as trustworthy online sources of medical knowledge, and anticipate ongoing collaboration between researchers, patients, and their advocates in trial design and accrual.

Neuropathic Symptoms and Their Risk Factors in Medical Oncology Outpatients With Colorectal vs. Breast, Lung, or Prostate Cancer: Results From a Prospective Multicenter Study

CONTEXT Few studies have examined the prevalence and severity of treatment-induced neuropathic symptoms in patients across different cancer types. OBJECTIVES This study aimed to report the prevalence of numbness/tingling (N/T) and neuropathic pain in patients with colorectal cancer (CRC) vs. other cancers, describe the prevalence of moderate-to-severe N/T by specific clinical variables, and examine factors associated with the presence of these symptoms. METHODS A total of 3106 outpatients with colorectal (n = 718), breast (n = 1544), lung (n = 524), or prostate (n = 320) cancer were enrolled at any point in their treatment. Assessments were conducted at the initial visit and 28-35 days later. Patients reported pain and N/T; clinicians reported mechanism of pain and ranked the top three symptoms causing difficulties. RESULTS Moderate-to-severe N/T was higher in patients with CRC relative to other cancer types (25.8% vs. 17.1%, P < 0.001); 25% vs. 10.5% of clinicians rated N/T as a top three symptom for patients with CRC relative to other cancers (P < 0.001). The prevalence of neuropathic pain was comparable between patients with CRC and other cancers (P = 0.654). Patients with CRC, longer duration of cancer, prior therapy, on current therapy, older patients, and patients of black race experienced worse N/T. CONCLUSION Patients with CRC experience significantly higher rates of N/T but comparable neuropathic pain, relative to patients with other cancers. Awareness of the prevalence and severity of neuropathic symptoms and their associated risk factors in this patient population is critical for both clinicians and patients.

Central nervous system relapse in patients with untreated HER2-positive esophageal or gastroesophageal junction adenocarcinoma

Although HER2‐positive breast cancers demonstrate a propensity for central nervous system (CNS) metastasis, it is unknown whether other HER2‐positive tumors, including adenocarcinomas of the esophagus/gastroesophageal junction (EAC), share this characteristic. Insight into this association may inform the development of HER2‐targeted therapies that penetrate the blood‐brain barrier. We examined HER2 overexpression and gene amplification in 708 patients with EAC who underwent curative‐intent surgery during a time period (1980–1997) when no patient received HER2‐targeted therapy. We identified patients whose site of first cancer recurrence was CNS and those who had a CNS relapse at any time. After a median follow‐up of 61.2 months, 3.4% (24/708) of patients developed CNS relapse (all involved the brain). Patients with HER2‐positive (vs ‐negative) primary tumors showed a higher 5‐year cumulative incidence of CNS relapse as first recurrence (5.8% vs. 1.2%; p = 0.0058) and at any time (8.3% vs. 2.4%; p = 0.0062). In a multivariable model that included covariates previously associated with HER2 or with CNS relapse in breast cancer, HER2 positivity was the only variable that was statistically significantly associated with shorter time to CNS relapse as first recurrence (p = 0.0026) or at any time (hazard ratio 4.3 [95% confidence interval 1.8 to 10.3]; p = 0.001). These are the first data in a non‐breast cancer to demonstrate an association between HER2 positivity and higher CNS relapse risk after surgery, and suggest that HER2‐positive EACs have a predilection for CNS metastases.

From Victim to Victor: “Breaking Bad” and the Dark Potential of the Terminally Empowered

As treatments for malignancies have improved incrementally over the preceding decades, patients with cancer have been encouraged to reject an attitude of hopelessness and to choose instead the role of fighters. The recasting of the cancer patient as warrior and winner, upheld through the Livestrong movement, reaches its monstrous apotheosis in the form of Walter White, the central figure in the AMC television series “Breaking Bad.” The story begins with Walt as the protagonist, but the arc of this conversion narrative transforms him into the antagonist, exploring the darkest potential of his post-diagnosis empowerment. His awareness of his own mortality enables him to take risks that his more rational, pre-cancer self would have avoided. Rather than being rendered impotent by fear of an impending death, he finds himself emboldened, liberated from behavioral norms, capable of heretofore-unthinkable violence and even murder. As Walt moves from victim to victor, the viewer realizes the perils of a survive-at-all-costs mentality and is forced to question their own, initially sympathetic perception of Walt. The series subverts the notion of the cancer patient made noble through struggle by portraying a man betrayed by his own body who then becomes willing to betray everything else in the amoral service of his pride.

Gain of Function: Empathy for the Uncertain Patient With Cancer

DOI: 10.1200/JCO.2011.35.1122 The news came from the embassy. Chest x-rays were performed on all immigrants to screen for tuberculosis, and my father’s was markedly abnormal. A theologian with a masterful command of language, he would later write of “a tumorous excrescence of considerable magnitude and menace.” But he described it to me, in terms comprehensible to his 8-year-old child, as a pineapple-sized cancer occupying much of his right lung. I never saw the chest x-ray, but its opacity cast a long, unwelcome shadow over my family’s new life in the sunny American South. We were unsettled in every sense, inhabiting a sparsely furnished house while we awaited my father’s curative-intent resection. Like his unseen metastases, our belongings were still in transit. The discovery of a lung tumor in a lifelong nonsmoker (indeed, a teetotaler) elicited the expected gasps of surprise from his peers and, from his son, misplaced moral outrage. It was simply “unfair” that a devout, clean-living man of God could develop cancer, especially a type so causally linked to the vice of smoking. To his credit, my father never asked “Why me?” but rather “Why not me?” Amid all the accidents and disasters that gave no premonition of impending death, why was he permitted the luxury of a warning? He used his time—the time between initial diagnosis and recurrence, between surgery and palliative chemotherapy—wisely. He died after finishing his life’s work, relishing his friendships, and sharing his love with his wife and son. Directing his energies so purposefully, he gave only fleeting thought to the etiology of his disease. But questions of cause and effect lingered in my mind. It would be dishonest to say that my own faith was not profoundly shaken by my father’s disease, all the more so when my clumsy prayers for an explanation received no discernible reply. Even while he was still alive, I searched for answers in science. My first literature search involved desperately rifling through encyclopedias in the school library, hoping to find some crucial fact that his doctors were overlooking. Such an epiphany was not to be found in the Britannica’s elementary account of cancer. During the years of medical training that eventually followed, I came to appreciate the difficulties of establishing a diagnosis, let alone definitive causation. The analogy between medicine and detective work resonated strongly with me as I assembled the facts of a clinical presentation to identify the culprit illness. In my own private investigation, I had no access to my father’s discarded records, but I accrued details of his case from my mother’s recollections. I learned that he had been inexplicably hypercalcemic throughout adulthood. I discovered that the histology of his lung tumor had not been adenocarcinoma but atypical bronchial carcinoid. And then my paternal uncle died from complications of a pituitary macroadenoma, another cruel twist of fate that I could not rationalize. At the beginning of medical school, I succumbed to the same temptations of hypochondria as every other student awoken suddenly to the frightening vastness of human pathology. While joking with my friends that every nosebleed presaged some exotic hemorrhagic fever, it never occurred to me that the explanation for my father’s disease would arrive through a process so prone to folly as self-diagnosis. The answer first appeared not in front of my nose, but on it. During residency, small fleshy papules erupted on my face, which a wise dermatologist diagnosed as angiofibromata. These red spots remained merely an affront to my vanity until the day before I started oncology fellowship, when I awoke with right lower quadrant pain so severe I was convinced I had appendicitis. In fact, there was no surgical emergency, and a subsequent visit to the internist revealed that I too was hypercalcemic. Suddenly, the hereditary connection became clear to me. I beseeched my internist to order the necessary tests and consultations to confirm my suspicion. Adenomas were found in every parathyroid gland, islet cell tumors were seen on endoscopic ultrasound of my pancreas, and I had a frameshift mutation in chromosome 11q13. Multiple endocrine neoplasia type 1 explained everything. My interest in oncology long predates my diagnosis with a familial tumor syndrome. I had been drawn to the field ever since watching my father’s doctors tend to him. Neutrophil counts rebounded as if by magic when they administered filgrastim. My father’s intractable nausea—until then, the bane of our tense, hushed meals around the family dinner JOURNAL OF CLINICAL ONCOLOGY A R T O F O N C O L O G Y VOLUME 29 NUMBER 22 AUGUST 1 2011

Incorporating Digital Tools to Improve Clinical Trial Infrastructure: A White Paper From the Digital Engagement Committee of SWOG

Progress toward improvement in cancer therapy relies on clinical trials. Yet, only a minority of eligible patients with cancer enroll as a result of multiple barriers at the patient, investigator, center, and national level. However, the rise of the Internet and mobile technology has created a slew of tools with medical applications, from Web sites to apps to social media platforms, all of which may aide clinicians in our quest to improve the clinical research enterprise. SWOG is one of five members in the National Cancer Institutes National Clinical Trials Network-the nations oldest and largest publicly funded cancer research network-and is taking a leadership role in exploring and testing the promise of digital engagement through the empaneling of the Digital Engagement Committee. This article outlines the mission, principles, and priorities of the Digital Engagement Committee and proposes how this work may inform the use of digital tools for the cancer research community and, hopefully, translate to improved outcomes for our patients.

A Balancing Act

DOI: 10.1200/JCO.2011.41.5695 My day began and ended at the bedside of a woman who was suffering. For nearly a year, cells had been proliferating inside her body, growing imperceptibly at first but crescendoing to a loud, agonizing climax. Now, to describe her as “an uncomfortableappearing female” in the written medical record would have been a gross understatement. And there was not just one woman; rather, there were two women, in different rooms, in different hospitals, each approaching the end of her pain. The two women had undergone physical transformations in parallel, their abdomens swelling synchronously as I had observed them both during the preceding months. The first woman was my patient, in the last throes of being ravaged by an aggressive angiosarcoma, a tumor that had replaced her liver and flooded her belly with ascites until her skin was drum tight. The second woman was my wife, giving birth to our second child after a difficult pregnancy. These two women had shared my longstanding worry about their outcomes, but as I stood beside each of them in turn, moving from the intensive care unit in the morning to the obstetric ward in the evening, I witnessed their paths diverge to the diametric opposites of death and life. The timing was eerily simultaneous. While my son’s first moments were spent having a nuchal cord disentangled, elsewhere the noose of end-stage cancer tightened mercilessly around a young mother. He entered the world at almost the exact instant when she made her premature exit. I had seen my patient’s death approaching from a distance. From our initial encounter, I was sure that her tumor, having already metastasized at presentation, was going to kill her. With the inevitability of Chekhov’s gun, it was only a matter of time before the final disaster, and my sense of foreboding progressed along with her advancing disease. But I could not have predicted that her tragic end would intersect precisely with a blissful beginning, posing the ultimate challenge to my notion of work-life balance. Throughout fellowship, I had struggled to maintain a great divide between my occupation and my family. Then that awful, wonderful day, with its tumult of feelings and obligations, proved to me that the barrier was permeable. The horrors encountered in my job were not so easily contained, and they could coincide with my most precious joys. The addition of a healthy child to our family would in no way compensate for the wrenching loss levied against another. This was not an equation to be solved like emotional algebra, yet I had to acknowledge a deep conflict. I was ecstatic yet grieving, apparently in danger of betraying two commitments. In mourning, was I guilty of not properly celebrating the birth? Or was I rejoicing despite my patient’s death? Never before had I experienced such extreme feelings at the same time, and the dissonance was unsettling. Perhaps my immaturity as an oncologist in training was partly to blame for the confusion. I have since been told by more experienced physicians that I do not actually mourn when a patient under my care succumbs to their disease; removed from the true despair reserved for a loved one, I should feel only an attenuated sadness, and to claim otherwise would be maudlin. To a certain extent, this distinction seems semantic, but I can acknowledge the difference. Admittedly, I do not experience a patient’s demise with the same devastation that accompanied, for instance, the death of my father, a sorrow that can still feel acute decades later. A series of equally profound losses would surely cripple me as a caregiver, given the distressing frequency with which my patients are terminally ill. That said, I fear becoming insensate more than I fear the pain of repeated bereavement. If I do not feel somber at the time of each patient’s death, then I will have become unresponsive to the mortal struggles that drew me to this precarious specialty in the first place. Even in my inexperience, I can already foresee a career spent reconciling resilience with compassion, donning emotional armor while hoping to retain some modicum of vulnerability. As my patient’s death threatened to loom over the joy of my son’s arrival, I was reminded that the human toll of the terrible diseases we treat can haunt us. In his magnificent chronicle of cancer, The Emperor of All Maladies, Siddhartha Mukherjee includes an account of his own daughter’s birth, which he rhapsodizes as “perhaps the singularly transformative event” of his life. Mukherjee describes JOURNAL OF CLINICAL ONCOLOGY A R T O F O N C O L O G Y VOLUME 30 NUMBER 15 MAY 2

Association of body mass index (BMI) with smoking and prognosis in 721 patients with esophagogastric adenocarcinoma.

e14635 Background: While obesity is an established risk factor for developing adenocarcinomas of the esophagus, gastroesophageal junction or gastric cardia (all 3 referred to as EAC), it is unknown whether BMI is associated with prognosis in this disease. Also, smoking (ever vs never) alters the relation between BMI and death, but its effects in EAC are unknown. METHODS All patients (n = 721) had surgery with curative intent for EAC at our institution (1980-97); none received neoadjuvant therapy. BMI (kg/m2) was calculated using height and weight at surgery, and categorized as obese (30+), overweight (25-29.9), vs normal (18.5-24.9). Cigarette smoking was categorized as ever vs never. Cox models were used to examine the association between variables and outcome (ie, disease-specific survival [DSS] and disease-free survival [DFS]). RESULTS Overall, 146 (21%) patients were stage IA-IIA; 575 (79%) were IIB-IIIC. For BMI, 160 (22%) were obese, 323 (45%) were overweight, and 238 (33%) were normal. There were 220 (30%) never, 337 (47%) ex-, and 164 (23%) current smokers (median 35 pack-years among ever smokers). Higher BMI (vs normal) was marginally associated with earlier stage (p=.08) and not with smoking. Smoking modified the effect of BMI on DSS (p=.055 for interaction), supporting stratification by smoking status (Table). CONCLUSIONS Obesity at the time of surgery was an independent predictor of adverse DSS and DFS compared to normal weight (adjusted for stage, grade, and age) in never smokers, but not in ever smokers. These data, in one of the largest reported resected EAC cohorts, are the first to show an adverse prognostic impact of obesity in EAC patients. [Table: see text].