Prospective Collection of Health Information and Biospecimens in Esophageal Cancer

Trial Information

Prospective Collection of Health Information and Biospecimens in Esophageal Cancer

Prospective databases have contributed substantially to the advancement of clinical medicine
and surgery. Examples of contributions from databases include an improved understanding of
the treatment and outcome for rare pathologies, the long term outcomes, and incidence of
uncommon complications from more common disease, and the repercussions of infrequently
performed complex procedures. (1-3) Problems identified by database review have led to
improvements in patient care and have guided the planning of numerous prospective studies.
The value of these databases incrementally increases with the number of patients enrolled
and with the length of time the databases have been maintained. Recently databases have been
used to create and test nomograms, which are increasingly being implemented to further
refine patient prognosis and to stratify patients for clinical trials. (4) We anticipate
that a prospective database will become more valuable as we look to the future.
Increasingly, molecular markers are identified which are thought to be important to patient
outcome or treatment response. When linked to pathologic tissues, clinical databases can be
used as the initial test of these newly identified markers and to determine which warrant an
independent prospective review. An independent prospective analysis performed for each
marker is costly and time consuming. A prospective database is more comprehensive,
accurate, and ultimately less time consuming than retrospective review for testing each set
of markers as novel questions arise.

This study will augment a prospective database of patients with esophageal cancer who
undergo investigation and /or surgery at Washington University School of Medicine. This
database will be periodically updated from the patient records and the computerized medical
record as patients continue routine follow-up care. These data will be used to address
questions regarding treatment and/or disease-specific outcomes.

In addition, we seek to acquire, store and analyze tissue and blood samples on these
patients. Samples will be processed and stored at the Tissue Procurement Facility of
Washington University School of Medicine. We will be using this material to support ongoing
efforts to identify serum markers with applications for early diagnosis, prognosis or
treatment efficacy. Markers identified will be linked to the clinical outcomes data as
captured in the database. Using the database, we will be able to link patient and treatment
outcomes to analyses performed on previously collected pathologic specimens or research
specimens and correlate these results with the patients' treatment response and outcome in a
time efficient and cost effective way.

The incidence of esophageal adenocarcinoma is on the rise. This has been a trend that has
been noted over the last decade and has gone hand in hand with a decrease in the incidence
of esophageal squamous cell carcinoma. Overall the number of patients with esophageal cancer
being diagnosed annually in the United States is steadily increasing. Esophageal cancer
afflicts approximately 19,000 patients annually in the United States. Broadly speaking,
esophageal cancer can be diagnosed at an early stage where there is no metastatic disease in
distant organs or at a relatively advanced stage when there is evidence of metastatic
disease in distant organs. Unfortunately patients with metastatic disease can receive only
palliative therapy and survival is quite limited. Stage I - III esophageal cancer is often
treated either by primary surgery or by a combination of preoperative or postoperative
chemoradiation therapy along with surgical resection. For stage I esophageal cancer,
surgical resection usually suffices and patients have upwards of 70-80% five year survival.
For patients with stage III esophageal cancer, the patients usually undergo induction
chemoradiation therapy followed by surgical resection. Induction chemoradiation usually
involves two cycles of % Fluorouracil and cisplatin with 5040 Gy of external beam radiation
therapy. This aspect of treatment has been well standardized (5). For stage II esophageal
cancer, surgery forms the backbone of therapy, however, there is no clear consensus among
treating physicians about the efficacy and role of chemoradiation therapy in these patients.

In patients who undergo chemoradiation therapy prior to surgery for esophageal cancer, it is
noted that 15-30% of patients have had a complete pathologic response to therapy. Thus no
tumor is detected in the resected specimen in these patients. Therefore, theoretically,
these patients have undergone an unnecessary resection. There is currently no way to
predict which patients would have had a pathologic response short of resecting the esophagus
and thus exposing the patients to the significant morbidity and mortality associated with
this major operation.

The most significant impact of next-generation sequencing on cancer genomics has been the
ability to re-sequence, analyze and compare the matched tumor and normal genomes of a single
patient. With the significantly reduced cost of sequencing and tremendously enhanced
throughput, it is now within the realm of possibility to sequence multiple patient samples
of a given cancer type. Genomic analyses over the last few years, with significant
contributions from our own Genome Center at the Washington University in St. Louis, (6-7)
have led to significant advances in understanding the behavior of certain tumors including
leukemias and glioblastoma. Specifically, in acute leukemias, our center has shown that
certain genotypes of the tumor predict favorable or unfavorable outcomes in the patients
analyzed. These findings now have been replicated in at least 3 independent studies of
separate AML cohorts. Our goal with this project would be to develop an understanding of
the genomic profile of esophageal carcinomas and changes with treatment. Furthermore with
availability of tissue both before and after chemoradiation therapy, we would be able to
identify changes in the genome of the malignancy with induction therapy. Also with
resection specimens we will be able to identify which patients have had a significant
response to therapy (including complete pathologic response) and others who have not had a
significant response to therapy or have had progressive disease despite therapy. This may
help us in identifying genomic features of the spectrum of esophageal carcinomas that
predict response to therapy after having had the chance to longitudinally study the disease
or the clinical course of the patients. Specifically this would be achieved by obtaining
tissue specimens from the malignancy prior to any therapy at the time of the initial
diagnosis or initial endoscopy for these patients. Subsequently the patients would undergo
routine therapy which would be based upon their clinical stage of disease. Eventually when
patients come back for their definitive resection, we would be able to access the pathologic
profile to assess response to induction therapy and also to study the genomic profile of the
tumor tissue after induction therapy.

A prospective database which would be maintained in the Division of Cardiothoracic Surgery
would be able to link clinical outcomes to processed tissues (pathologic specimens, research
specimens and plasma) enabling higher level correlation studies to be performed. Proper use
of such a resource will aid considerably in our ability to advance our field and care for
our future patients with esophageal cancer.

Our eventual aim with the analyses of the genome of esophageal cancer would be to identify a
patient population that has a complete pathologic response to chemoradiation therapy for
carcinomas of the esophagus. We will be able to identify genomic differences, if any,
between such tumors and other tumors where there is only a partial response/no
response/progressive disease with induction chemoradiation therapy. If analyses are
sufficiently predictive and efficiently dichotomize patients into these categories based
upon genomic profiles, this can be the lead groundwork for a clinical trial where patients
with certain specific genomic profiles would be treated only with chemoradiation therapy for
carcinomas of the esophagus, thus precluding unnecessary surgery with its significant
attendant morbidity and mortality. As we already know, about 15-30% of patients undergoing
induction chemoradiation therapy clinically fall into such a category. The well known
morbidity of an esophagectomy includes significant pulmonary problems, anastomotic leaks,
cardiovascular problems, chyle leak, venous thromboses, amongst other major problems. The
rate of morbidity after an esophagectomy varies from 20-40%. Also there is 3-14 % risk of
mortality from esophagectomy. (8-9) Therefore, a tool that can effectively predict which
patients do not require an esophagectomy for treatment of carcinomas of the esophagus would
be extremely useful clinically.