| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Departments of Medical Oncology (Drs. Ang, Tan, and Leong), Clinical Trials and Epidemiological Sciences (Ms. Koh), and Surgical Oncology (Dr. Agasthian), National Cancer Center, Singapore; and Departments of Respiratory and Critical Care Medicine (Dr. Eng) and Diagnostic Radiology (Dr. Cheah), Singapore General Hospital, Singapore.
Correspondence to: Eng-Huat Tan, MD, Department of Medical Oncology, National Cancer Center, 11 Hospital Dr, Singapore 169610; e-mail: dmoteh{at}nccs.com.sg
 |
Abstract |
|---|
 TOP Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Study objectives: Patients must fulfill the following criteria before a diagnosis of PIME can be made: clinical presentation dominated by pleural/pericardial effusion; histologic proof of malignancy obtained from the pleura and/or pericardium; no definite primary site in the lungs or elsewhere from CT scan of the chest, chest radiograph, or physical and endoscopic examination; no history of malignancy; and no history of asbestos exposure. Exposure to environmental tobacco smoke (ETS) among the nonsmokers was examined in a case-control setting.
Methods: We conducted a retrospective search of our database of patients who were referred to the Department of Medical Oncology with a diagnosis of pleural/pericardial effusion from January 1993 to January 2000.
Results: Seventy-one of 200 patients from our database met the criteria. A significant majority of the patients were women (65%) and nonsmokers (72%). All patients had adenocarcinoma shown on biopsy. The majority of patients (63%) had disease localized to the intrathoracic serosal surfaces; the rest had distant metastases involving the lung (50%), bone (27%), liver (19%), brain (8%), and skin (4%). Six patients had two or more sites of distant metastases. There was a significant association with ETS exposure when compared to a control group comprised of patients with colonic cancer, matched for sex and age. The median survival was 10 months for patients with disease localized to the pleura/pericardium and 7 months for those with distant metastases. Thirty-eight patients (54%) received chemotherapy. All had platinum-based chemotherapy, except for three patients. The median survival for patients treated or not treated with chemotherapy was 12 months and 5 months, respectively. This difference in survival was statistically significant (p = 0.003).
Conclusions: PIME should be viewed as a distinct entity. Its etiology remains largely unknown, although exposure to environmental tobacco smoke may play a part. Platinum-based chemotherapy may have a positive biological effect on this disease. More studies are required to elucidate the epidemiology, possible etiologic factors, and treatment options for this group of patients.
Key Words: chemotherapy • women • nonsmokers • pericardial effusion • pleural effusion
|
Introduction |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
It is estimated that 6 to 15% of all malignant pleural effusions are from an unknown primary site.1 2 The search for the primary site is generally symptom guided. An exhaustive search (usually radiologic) is, on the whole, unproductive, costly, and does not influence the treatment decision significantly. The treatment addressed in published studies1 3 focused mainly on local treatment to afford symptomatic relief and to reduce the risk of recurrence with either sclerosing agents or creation of shunts. One study3 looked at the use of chemotherapy, and this was deemed effective in relieving symptoms in 78% of patients for a median period of 6 months. The median survival of this heterogeneous group of patients in this study was 12 months. The use of chemotherapy in PIME specifically has not been addressed before.
The paucity or absence of literature pertaining to this group of patients who present with PIME initiated this retrospective review of such patients referred to the Department of Medical Oncology since 1993. This retrospective review primarily aims to shed some light on the characteristics of patients with this form of presentation. It is also our hope that this study will stimulate the medical community at large to look more closely at patients who presented in like manner, and to help to confirm or refute our suggestion that this could be a distinct entity.
|
Materials and Methods |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
All patients routinely had a complete history taken and underwent a detailed physical examination, which particularly included the breast, thyroid, pelvis, and rectum. Tumor-related symptoms and physical signs were recorded. Most of the patients had at least a CT scan of the chest done as part of the workup. Mammography and CT scan of the abdomen and pelvis were not routinely done unless symptoms and signs indicated their use. In selected patients with pleural effusion, video-assisted thoracoscopy or open thoracotomy were required for diagnostic purposes, after failure to obtain histology by pleural tap and biopsy. Clinical records were also updated to see if a primary tumor that had been missed initially had surfaced subsequently. Patients were excluded if a new primary site was identified. Death records were obtained from the Singapore death registry.
After identification and documentation of the patients, a telephone follow-up of their smoking habits was made where smoking history was incomplete. If the patients were lifetime nonsmokers, the degree of exposure to environmental tobacco smoke (ETS) was ascertained. In our study, we defined ETS exposure as significant if there was a parent or relative staying in the same household for > 5 years who had smoked, a spouse living together who had smoked for > 5 years, or if the patient himself considered ETS exposure at the workplace to be significant.
We conducted a case-control study to assess the significance of ETS exposure in this group of patients. We selected nonsmoking patients with colorectal cancer who were followed up at the GI outpatient clinic at our center who were matched for age and sex as our control group. Patients with colorectal cancer were selected as control subjects for two reasons: (1) the lack of established association of colorectal cancer with tobacco smoke,4 and (2) the easy availability of these patients for the interviews.
Statistical Analysis
Factors affecting outcome were assessed with the Cox
proportional-hazards regression model, which is incorporated into the
statistical software program (SPSS; Version 10.0; Chicago IL).
Factors investigated were age, sex, performance status, and extent of
disease. Estimate of the effect of chemotherapy on overall survival
among patients who receive or did not receive chemotherapy was also
incorporated into this model. Survival curves were generated from this
program by the Kaplan-Meier method.5
Direct comparisons of
survival curves were performed with the log-rank test.
For the analysis of exposure to ETS among study and control group,
three nonsmoking control subjects matched for sex and age were required
for each case in order to have 80% power of detecting an odds ratio of
3.5, setting 
at 0.05. Fisher’s Exact Test (one sided) was used
to determine the significance of association.
|
Results |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Most of the patients had pleural effusion only (86%). Right-sided effusion was more common (54%). Bilateral effusions of similar volume were seen in five patients. Three patients presented with pericardial effusion only, and seven patients had both pericardial and pleural effusion. Pleural fluid cytology and/or pleural biopsy were able to confirm the diagnosis in most of the patients (59%). Fifteen patients, however, required thorascopic biopsy (11 patients) or open thoracotomy (4 patients) for diagnosis. Diffuse pleural-based disease was seen in all of these patients at the time of procedure. All patients had adenocarcinoma.
Most of the patients had disease localized to the intrathoracic serosal surfaces at the time of the diagnosis (63%). The lungs (50%), bone (27%), and liver (19%) were the three common sites of distant metastases. Performance status was good (less than or equal to ECOG performance status of 1) in most of the patients (86%).
Local treatment for patients requiring symptomatic drainage initially included chest tap for 35 patients (55%) and chest tube drainage for 17 patients (27%). Twenty-eight patients (44%) required repeated drainage procedures for reaccumulated effusions. Pleurodesis with either talc or bleomycin was performed in 28 patients for large symptomatic effusions. It was difficult to assess the efficacy of pleurodesis alone in preventing reaccumulation, as a significant number of the patients also received systemic treatment with chemotherapy.
Thirty-eight patients (54%) received chemotherapy. All had platinum-based chemotherapy except for three patients. Response to chemotherapy in these patients was difficult to assess for the following reasons: (1) tumor response according to the World Health Organization criteria was not possible in these patients due to the lack of measurable disease; and (2) most of the patients had pleurodesis as well, making it difficult to differentiate the efficacy of this local procedure from chemotherapy. With these limitations, we defined a positive response as the absence of reaccumulation of effusion and/or symptomatic improvement/stabilization based on records in the patient charts. The degree of symptomatic relief is difficult to assess in this retrospective review. Based on these criteria, 13 of 32 evaluable patients had a positive response to chemotherapy. These patients received a median of six cycles of chemotherapy. The nonresponders received a median of only two cycles of chemotherapy.
At a median follow-up of 16 months for all surviving patients, the median survival was 10 months for patients with localized disease and 7 months for those with distant metastases. This difference was, however, not significant (Fig 1 ). There was, however, statistically significant difference in survival between those who received chemotherapy (median survival, 12 months) and those who received supportive care alone (median survival, 5 months) [Fig 2 ]. Chemotherapy remains as an independent factor predicting for improved survival, with a hazards ratio of 0.4 (confidence interval, 0.2 to 0.8).
|
|
|
|
Discussion |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The medical literature and textbooks tend to classify these patients as having metastatic carcinoma from an unknown primary site, and no effort was made to further define this subgroup of patients. This is the first study to have done so, albeit with the limitations characteristic of a retrospective study. From this study, the predilection of this illness toward the female gender and Chinese race, and the uniform histology, seem to suggest that this form of presentation could be a distinct entity. Possibilities include a primary from the pleura and/or pericardium, and a small subserosal site that could breach the serosal lining early in the course of the disease. The latter is more likely and would be difficult to locate in the presence of diffuse serosal disease at presentation. This could also explain the uniform histology in this group of patients, as adenocarcinoma tends to occur in the lung periphery.
The likelihood that these patients had undiagnosed breast, GI, or gynecologic malignancy is probably low, as the symptoms referable to the primary site should be manifest on follow-up. Only one patient in this study developed vague induration in her right breast, which was on the same side as the pleural effusion on follow-up, but this was proven to be negative on trucut biopsy.
Management of this condition remains ill defined. The use of drainage or shunt, and sclerosing therapy generally provided only short-term palliation with a high probability of reaccumulation of the fluid and progression of symptoms. Moreover, a significant number of patients developed neuralgic, band-like pain over the ipsilateral chest and/or upper abdomen, which was not reduced significantly with drainage treatment alone. The use of chemotherapy is usually met with difficulty in assessing the tumor response due to lack of measurability. However, chemotherapy appeared to have a positive impact on the overall survival, which was independent of factors such as age, sex, stage, and performance status.
Tobacco smoke contains approximately 60 known or suspected carcinogens, and animal bioassays have shown that sidestream tobacco tar is more carcinogenic per unit weight than mainstream tar. Since 1979, numerous epidemiologic studies have examined the association between passive smoking and cancer. While the evidence is less convincing for nonbronchogenic cancers, most studies6 7 8 9 10 on lung cancer showed a positive association with ETS exposure especially in lifetime nonsmoking women. This association, however, is substantially weaker than that between active smoking and lung cancer due probably to the substantially lower dose-rate for passive smokers. In our study, patients with PIME showed significantly increased exposure to ETS at home when compared to a control group with colorectal adenocarcinoma. While the limitations of a case-control study do not permit us to attribute a cause-effect relationship between passive smoking and PIME, it does point to the possibility that the primary site originates from the intrathoracic region.
We are keenly aware that the retrospective nature of this study limits the interpretation of the results significantly. The danger of bias in selecting the population for study does exist. Hence, there is a need for a prospective study to characterize this group of patients more thoroughly. This prospective study should also include more intensive workup, such as mammography and GI endoscopy, to exclude an asymptomatic primary. Prospective collection of clinical and molecular data from such a cohort of patients would be useful to better define the possible etiologic factors and the optimal therapy for such a condition.
|
Footnotes |
|---|
Received for publication May 25, 2001. Accepted for publication January 24, 2001.
|
References |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. L. Gross, R. N. Younes, D. Deheinzelin, A. L. Diniz, R. A. da Silva, and F. J. Haddad Surgical Management of Symptomatic Pericardial Effusion in Patients with Solid Malignancies Ann. Surg. Oncol., December 1, 2006; 13(12): 1732 - 1738. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Woenckhaus, U. Grepmeier, B. Werner, C. Schulz, F. Rockmann, P. J. Wild, G. Rockelein, H. Blaszyk, M. Schuierer, F. Hofstaedter, et al. Microsatellite Analysis of Pleural Supernatants Could Increase Sensitivity of Pleural Fluid Cytology J. Mol. Diagn., October 1, 2005; 7(4): 517 - 524. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
