Future Developments in Esophageal Cancer Research

The progress that has been made against esophageal carcinoma is limited. Many relevant issues remain. Herein the author discusses his outlook for the treatment of this disease, which has a steadily rising incidence.

Each year esophageal carcinoma accounts for more than 460,000 new cases and approximately 375,000 deaths worldwide. The incidence has steadily risen in the United States over the past 17 years. Squamous cell histology is more frequent in the endemic areas, whereas adenocarcinoma histology is more frequent in the nonendemic areas. An alarming as well as intriguing aspect is the increase in the incidence of adenocarcinoma in the West and the shift in the location to the lower third of the esophagus and gastroesophageal junction. Although Scotland and England have the highest incidence of adenocarcinoma of the esophagus in the world, there are no definite clues as to why. Squamous cell carcinoma has been associated with smoking and alcohol, whereas adenocarcinoma is associated with obesity, gastroesophageal reflux disease, and Barrett’s metaplasia. Whereas previously the diagnosis of adenocarcinoma has often been made in executive-type Caucasian men in their late 50s and early 60s, there is now a rise in the incidence of adenocarcinoma in Caucasian women and blacks. It would appear that adenocarcinoma of the esophagus is not a gender- or ethnicity-specific disease but predominantly related to lifestyle factors and that susceptibility has a major role.

Because early detection strategies for esophageal cancer have rarely been implemented, it is commonly diagnosed in an advanced stage resulting in an increasing number of deaths. It is hoped that heightened awareness of the consequences of Barrett’s metaplasia might lead to a higher fraction of individuals undergoing surveillance and early detection of adenocarcinoma. A parallel strategy would be to identify individuals who are at high risk for developing adenocarcinoma through molecular epidemiologic studies. Molecular studies hold considerable promise. Various genetic polymorphisms have already been correlated with an increased risk of adenocarcinoma of the esophagus, and such findings (when mature) may facilitate targeted surveillance. Much work remains to be done to develop a validated risk model. With the advent of emerging technology and a greater understanding of genetic and epigenetic changes in cancer and patient genetics, we will have a better understanding of the reasons why some individuals have a higher risk and why inherent heterogeneity occurs in the clinical biology of esophageal cancer that is commonly encountered.

Localized carcinoma of the esophagus is a potentially curable condition, but to provide optimum and effective care it is necessary to establish an experienced infrastructure that exercises a multidisciplinary approach. A multidisciplinary approach requires accurate staging and review of medical data by all involved disciplines (gastroenterology, radiology, medical oncology, surgery, radiation oncology, pathology, and nutrition) before rendering any therapy. Newer techniques to stage cancer have facilitated better patient selection for complex approaches. Among the many new techniques to accurately stage cancer, endoscopic ultrasonography (with an ability to perform fine-needle aspiration of suspected nodes) and positron emission tomography (PET) represent advancements. Changes in serial PET are of particular importance because PET not only provides staging information but also characterizes the clinical biology of esophageal cancer undergoing therapy. The future of cancer imaging lies in using more sophisticated technology with new (and specific) imaging targets.

The major source of frustration for patients, their relatives, and the medical team has been the selection of therapy that is likely to be effective and the avoidance of ineffective ones. There is some clarity for early stage cancer. For example, a consensus is emerging for endoscopic mucosal resection as the preferred therapy for patients with T1a cancers and for surgery for those with T1b cancers. There is lack of convincing data when dealing with cancers that are stage II or III; however, in the West, for the commonly diagnosed stages of localized esophageal adenocarcinoma (II or III), preoperative therapy has become increasingly popular despite the lack of convincing data in randomized trials. One tradition has been to group patients in a certain stage category (eg, combining all patients with stages II and III) and offer everyone the same therapeutic approach. Some patients are cured by this strategy but many are not, and some do not benefit at all. Clearly, all patients sustain considerable toxicities from chemoradiation and complications from surgery. Unfortunately, clinical parameters are not helpful in selecting therapy for patients with stage II or III cancer. It is increasingly clear that residual cancer found in the resected surgical specimen after preoperative therapy (particularly chemoradiation) is a determinant of the patient’s long-term outcome. The class of cytotoxic agents or the use of induction chemotherapy also does not seem to influence the pathologic response. Fluoropyrimidine should be administered with radiation, but the second cytotoxic agent can be a platinum compound, taxane, or camptothecin. The value of the addition of biologics to chemoradiation is being investigated. We should realize that this is simply an extension of an investigative tradition that promotes empiricism.

When asking the question of what tools could be employed to optimize therapy for patients with esophageal cancer, one should first recognize the three subtypes of esophageal cancer that drive the clinical biology and therefore patient outcomes. These subtypes are based on the degree of chemoradiation resistance inherent in the primary tumor. Each subtype could be treated differently and expected to result in a different patient outcome.

Subtype I includes patients whose tumors are extremely chemoradiation sensitive and in whom no cancer cells are found in the surgical specimen (pathologic complete response). Approximately 25% of patients (irrespective of whether the tumor is clinical stage II or III) achieve a pathologic complete response. Their 5-year survival rate often exceeds 50%. One could argue that if we could identify these patients before therapy is initiated, observation after chemoradiation would be a strong consideration (ie, an attempt at esophagus preservation with surgery used as salvage). Although some biomarkers are associated with pathologic response, currently these are not well refined, and the specificity is too low for implementation.

Subtype II includes patients whose tumors are partially sensitive to chemoradiation. Approximately 1% to 50% have residual cancer found in the surgical specimen. These patients (if identified before any therapy is delivered) seem to benefit from chemoradiation and surgery. They have a moderate lifespan and moderate metastatic potential.

Subtype III includes patients whose tumors are extremely resistant to chemoradiation. More than 50% have residual cancer found in the surgical specimen. Clearly, these patients need to be identified as soon as possible (preferably before therapy is initiated, but even early during therapy would be better than what is done today) to avoid chemoradiation and proceed directly to surgery. Their prognosis is very poor.

The progress that has been made against esophageal carcinoma is limited and certainly not impressive. Many relevant issues remain, including whether to deal with squamous cell carcinoma differently than adenocarcinoma, the dose of radiation and the fractionation scheme, how to orchestrate a multidisciplinary approach, the therapy of T4 and M1 cancers, specific surgical techniques, the development of a risk model for surgical therapy and chemoradiation therapy, the frequency of follow-up and with what staging methods after definitive therapy, the programmatic approach to molecular biologic studies, and so on. These issues are beyond the scope of this article. Table 1 summarizes my thoughts on what the future might hold. Fortunately, the future appears promising. Nevertheless, the startling increase in the incidence of adenocarcinoma suggests that esophageal adenocarcinoma is the disease of future and in 20 years might surpass colon cancer. Although this sounds unlikely, we will see. I hope that the National Institute of Health is paying attention, because it has not yet set aside dedicated funds to study adenocarcinoma of the esophagus or gastroesophageal junction.

Table 1

Future of esophageal cancer research

Area of Research	Future Possibilities	Comments
Epidemiology	Molecular epidemiology will dominate this arena.	To study large cohorts of patients, collaboration among institutions is necessary.
Etiology	Better understanding of specific causes of adenocarcinoma is highly likely.	Knowledge will fuel prevention strategies.
Prevention/early detection	Better understanding of susceptibility factors (genetic or lifestyle related) will arise. Creation of a multifaceted risk model will follow.	Identification of high-risk populations and tailored surveillance are necessary.
Imaging	More specific imaging to identify and guide therapies will be developed.	Biomarkers will have a greater role in complementing imaging advances.
Staging	Improved imaging and other technologies will allow more sophisticated staging that will be able to predict biologic behavior of cancer.	Patients will be subgrouped in a more meaningful manner than is possible today. Metabolic and target-specific imaging will define biology.
Therapy for localized cancer	Individualization of therapy will be possible by studying molecular biology of cancer and patient genetics.	Selection of optimal therapy might become the norm and will reduce cost and morbidity.
Radiotherapy	Further evolution of radiation techniques to reduce normal tissue toxicities is likely.	Sophisticated tumor imaging will have an increasing role in radiation planning for esophageal carcinoma.
Surgery	Minimally invasive surgery will evolve and will be employed in select cases.	Evolving technology might allow image-guided surgery and also identification of involved nodal groups.
Personalized medicine	Integration of clinical and molecular information could lead to validated individualized approaches.	This effort will reduce cost, frustration, morbidity, and uncertainties but at the same time might identify new exploitable therapeutic targets.