| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Eur J Cardiothorac Surg 2004;26:165-172
© 2004 Elsevier Science NL
a Department of Thoracic Surgery, University of Torino, 3, Via Genova I-10126, Torino, Italy
b Department of Pathology, University of Torino, Torino, Italy
Received 23 October 2003; received in revised form 1 March 2004; accepted 19 March 2004.
* Corresponding author. Tel.: +39-11-6335919; fax: +39-11-6960170
e-mail: enrico.ruffini{at}unito.it
 |
Abstract |
|---|
 Top Abstract 1. Introduction2. Materials and methods3. Results4. DiscussionAppendix A. Conference...References |
|---|
Key Words: Pre-invasive lesion • Lung neoplasms • Atypical adenomatous hyperplasia • Carcinoma in situ • Diffuse idiopathic pulmonary neuroendocrine cell hyperplasia
|
1. Introduction |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. DiscussionAppendix A. Conference...References |
|---|
In 1999 the World Health Organization [1] published a tumor classification system that defines three different pre-invasive lesions for the human bronchial epithelium: (1) atypical adenomatous hyperplasia (AAH) which may be the precursor to adenocarcinoma/bronchioloalveolar carcinoma (BAC) (particularly peripheral); (2) squamous dysplasia/carcinoma in situ (CIS) which may be the precursor to squamous cell carcinoma; (3) Diffuse idiopathic pulmonary neuroendocrine cell hyperplasia (DIPNECH), which may progress to carcinoid. Although a number of reports [2] have investigated the molecular and pathologic aspects of the pre-invasive lung lesions, and have basically confirmed the carcinogenic sequence adenoma–AAH–adenocarcinoma and hyperplasia–metaplasia–dysplasia–CIS with either circumstantial or genetic evidence, very few studies have addressed the issue in a clinical setting. Even fewer studies have been published looking at the outcome of patients with resected primary lung carcinoma and associated pre-invasive lesions.
The purpose of the present study was to evaluate the occurrence and significance of pre-invasive lesions (AAH, CIS and DIPNECH) incidentally detected in the surgical specimen following lung resection for primary neoplasms.
|
2. Materials and methods |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. DiscussionAppendix A. Conference...References |
|---|
2.1. Pre-invasive lesions: classification criteria
For the present study, the 1999 WHO classification of pre-invasive lesions of the bronchial epithelium into 3 groups was adopted: (1) Atypical Adematous Hyperplasia (AAH); (2) Squamous dysplasia/CIS; (3) Diffuse Idiopathic Pulmonary Neuroendocrine Cell Hyperplasia (DIPNECH). Figs. 1 and 2
illustrate microscopic characteristics of AAH and CIS.
|
|
Conversely, distinction between AAH and subtypes of adenocarcinoma, particularly the non-mucinous variant of Bronchiolo-alveolar Carcinoma (BAC) is difficult. For the present study, the following criteria have been adopted for definition of AAH: (1) Well-defined lesion less than 5 mm in greatest diameter, comprised of proliferation of non-invasive single-layered atypical epithelial cells along the alveolar walls without central scar formation or collapse; (2) The cytoplasm is abundant, and the cells have a rounded or domed appearance resembling Clara cell/type II pneumocytes/columnar cell; (3) the atypical cells in AAH have hyperchromatic nuclei and prominent nucleoli, although atypias are less marked than in adenocarcinoma. No distinction was made between low and high grade AAH.
Finally, diagnosis of DIPNECH requires the presence of hyperplasia of airway neuroendocrine cells, organized also in small nodular aggregates, without invasion beyond the basement membrane and in association with chronic inflammation or fibrosis. In this setting, but in absence of airways inflammation or diffuse interstitial fibrosis, DIPNECH is regarded as a precursor to the development of carcinoid tumorlet (in case of diameter <5 mm) or carcinoid tumor (in case of diameter >5 mm).
2.2. Statistical analysis
Comparison between variables was made using chi-square test (with Fisher exact test for small sample size) for categorical variables and student t-test for continuous variables. Correlation between variables was performed using logistic regression analysis with a binary dependent variable (the presence or absence of the pre-invasive lesion) and different independent variables. Wald test was used for statistical significance. Odds ratios with associated 95% Confidence intervals were also provided as estimates of the association between the independent and dependent variables. Survival analysis was performed using the Kaplan–Meier method; differences between survival curves were evaluated using the log-rank test. A P-value less than 0.05 was considered for hypothesis testing of significance.
|
3. Results |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. DiscussionAppendix A. Conference...References |
|---|
|
Histology of the associated primary tumor was BAC in 8 patients, adenocarcinoma in 17, adenosquamous carcinoma in 3.
Among BAC, there were 6 well-differentiated and 2 moderately differentiated BAC. Among adenocarcinomas, there were 3 well-differentiated, 7 moderately differentiated and 6 poorly differentiated tumors. Nineteen patients had tumors in the upper lobes, 1 patient in the middle lobe and 10 in the lower lobes; among these, 3 patients had centrally-located tumors extending in more than one lobe. Three patients had multiple AAH.
Thirteen patients were at Stage I (3 Ia and 10 Ib); 7 patients were at Stage II (4 IIa and 3 IIb), 6 patients were at Stage IIIa-N2 and 2 patients were at Stage IIIb-T4.
Three patients had a history of previous malignancy: one had an occult cervical mucinous adenocarcinoma treated with primary chemo-radiotherapy; the other two had colic adenocarcinoma successfully treated with primary resection.
We compared patients with AAH (AAH+) to our total population of resected lung cancer patients (AAH–) with respect to sex, age, median tumor dimension, histology, histologic grading, histologic evidence of invasiveness (vascular invasion, perineural invasion), peritumoral lymphocytic infiltrates, pathologic (p)TNM, lobe location of the tumor and the presence of associated synchronous or metachronous malignancies. Logistic regression analysis was used considering AAH as the dependent variable and the all aforementioned variables as independent (predictive) variables (Table 1).
Sex, age, median tumor dimension, histologic evidence of invasiveness, peritumoral lymphocytic infiltrate and Stage were not correlated with AAH.
A strong correlation was found between AAH and histology: all patients with AAH presented either BAC or adenocarcinoma or combined or mixed tumors comprising adenocarcinoma (P=0.00002).
A trend, although not a significant correlation, was found between histologic grading of the tumor and AAH: patients with AAH presented, as compared to our total population of resected lung neoplasms patients, a greater histologic differentiation (G1 and G2).
We further evaluate whether a correlation exists between the location of the invasive tumor and the presence of associated AAH lesion, as suggested by some authors. In our series the difference among the frequencies in upper, middle and lower lobes was not significant in the AAH+ and AAH– groups.
Finally, we did not find any significant correlation between a history of previous malignancy and the presence of AAH, as suggested by other authors.
Survival rates in the two groups were not significantly different (P=0.3) (Fig. 3) ; the lack of significance persists (P=0.7) even if the AAH group is compared with the population of resected adenocarcinoma+BAC (Fig. 4) : indeed, patients with AAH had a slightly better prognosis than the total group.
|
|
|
Histology of the associated primary tumour was squamous cell carcinoma in 39 patients, adenocarcinoma in 2 and BAC in 1 patient. Eight patients had well-differentiated (G1), 27 moderately differentiated (G2) and 7 poorly differentiated (G3) tumours.
At pathologic examination, vascular invasion was found in 19 patients, perineural invasion in 8, and peritumoral lymphocytic infiltrate in 10.
Twenty-seven patients were at pathologic Stage I (9 Ia and 18 Ib); 9 patients were at Stage II (5 IIa and 4 IIb), 6 patients at Stage IIIa (1 T3 and 5 N2). Six patients had a history of previous malignancy, including 2 laringeal, 1 bladder and 3 lung carcinomas.
The 5 patients with brCIS had all squamous cell carcinoma as associated primary; there were 1 T1N0M0, 3 T2N0M0 and 1 T2N1M0. Post-operative management included bronchoscopic surveillance in 3 cases and external radiotherapy in 2. Two patients are alive and disease-free at 2 and 5 years from surgery, and 3 patients died of distant metastases at 1, 2 and 3 years.
We compared patients with CIS (CIS+) with our total population of resected lung carcinoma patients (CIS–) using the same variables as for AAH: sex, age, median tumor diameter, histology, histologic grading (G1 through G3), vascular and perineural invasion, peritumoral lymphocytic infiltrate, pTNM, location of the tumor, a history of previous malignancy. (Table 2).
A strong correlation resulted between CIS and squamous cell carcinoma as associated primary tumour (P=0.009). None of all the remaining variables showed a significant correlation with the presence of CIS.
Fig. 5 shows the survival curves of the CIS and total population groups. The difference was not significant (P=0.2). Indeed, patients in the CIS group presented a slightly survival advantage over the total patient population which becomes more evident if CIS group is compared with the population of resected squamous cell carcinoma (Fig. 6) , although it did not reach statistical significance (P=0.1).
|
|
3.3. Diffuse idiopathic pulmonary neuroendocrine cell hyperplasia (DIPNECH)
DIPNECH was found in 3 patients with resected lung neoplasms. Prevalence among all resected patients with neuroendocrine neoplasms (typical and atypical carcinoids) was 5.4% (3/55). There were 1 man and 2 women with a mean age of 62 years (48–77 years). All patients had an associated well differentiated neuroendocrine tumor (typical carcinoid). pTNM was T1N0M0 in 1 patient and T2N0M0 in 2 patients. None of the patients had history of previous malignancy. None developed local or distant relapse at follow-up. All patients are alive and well at 30, 35 and 37 months from surgery. Despite the small number of cases which do not permit to draw any statistical conclusion, patients with resected neuroendocrine neoplasms and associated DIPNECH seem to behave similarly to those with corresponding neuroendocrine tumors without DIPNECH.
|
4. Discussion |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. DiscussionAppendix A. Conference...References |
|---|
Prevalence of AAH in clinical study varies widely, ranging from as low as 2.8% in autopsy cases [4] to as high as 23% in resected lung specimens containing adenocarcinoma [5,6]. Our prevalence was 2.8% out of the total population of resected lung carcinoma, and 5.8% in resected adenocarcinomas. The difference may be partly due to the difficulty to distinguish between AAH and multicentric BAC or a pulmonary metastasis from adenocarcinoma. Some authors claim that the more atypical AAHs are almost indistinguishable from BAC and suggest that they may actually be the same condition [5]. Other authors [6,7] have suggested using size (5 mm) to differentiate between AAH and BAC. Indeed, very often, the distinction between the conditions can only be made based on the outcome. Some reports pointed out that many patients have multiple AAH, ranging from less than 10 to as many as 100 [8–10]. The series reporting multiple AAH also observed that these patients are at increased risk to develop multiple synchronous primary adenocarcinoma (from 2 up to 6 for each patient) [11]. None of our patient had synchronous primary adenocarcinoma and only 3 cases presented multiple AAH. Indeed, according to a recent review [12], it appears that, to some extent, the number of AAH lesions found would be proportional to the enthusiasm of the pathologist to look for them, the amount of lung tissue blocked, the state of lung fixation, and so on. It is therefore difficult to draw reproducible conclusions from series from different Institutions using different methods of pathological analysis, lung fixation and number of blocks.
Prevalence of CIS is more difficult to determined since few clinical reports have been published. In a recent report, Pierard and colleagues [13] using fluorescence bronchoscopy found CIS (which they call Roentgenographically Occult Lung Carcinoma, or ROLC) in 4 out of 43 resected lung carcinomas (9.3%). Other less recent reports included similar prevalence rates [14–16]. Our prevalence of 4% (8.7% among all resected squamous cell carcinomas) seems to be in the reported range.
Prevalence of DIPNECH is even more difficult to calculate, since definition criteria for this condition have not appeared until very recently and the issue still remains quite controversial. Neuroendocrine proliferations are a spectrum of histologic patterns ranging from neuroendocrine cell hyperplasia to tumourlets and carcinoid tumours, and distinction among these lesions are often based on arbitrary criteria. Customarily, DIPNECH is confined to airway mucosa, without penetration through the basement membrane and appear in a diffuse pattern. Tumourlets represent micronodular neuroendocrine cell proliferations, extending beyond the basement membrane and forming aggregates less than 0.5 cm in diameter; arbitrarily, every neuroendocrine proliferation greater than 0.5 cm is defined as a carcinoid tumour. Essentially, the only evidence that DIPNECH is a pre-invasive lesion is circumstantial, based on isolated case studies which reported concurrent carcinoid tumors and DIPNECH [17,18]. In the present study, DIPNECH was found in 3 patients out of 55 resected neuroendocrine neoplasms (typical and atypical carcinoid tumors), indicating a 5.4% prevalence. All patients had typical carcinoids and all basically behaved similarly to our total population of resected carcinoids.
In order to investigate the clinical and prognostic significance of the presence of associated pre-invasive lesions in resected lung carcinoma patients, we performed a differential statistical analysis on AAH and CIS (the DIPNECH group was too small and we limited ourselves to a descriptive analysis) analizing whether a correlation exists among different clinico-pathologic variables and the presence of the pre-invasive lesion.
The results of the analysis indicate that a strong correlation exists between AAH and either adenocarcinoma or BAC or mixed tumors containing adenocarcinoma (es. adenosquamous carcinoma) and between CIS and squamous cell carcinoma. This finding confirms previous reports, supporting further evidence of the carcinogenic pathways hyperplasia–metaplasia–dysplasia–squamous cell carcinoma [19], and adenoma–adenomatous hyperplasia–adenocarcinoma (or BAC). Unlike other reports, however, we were unable to demonstrate any other significant correlation, particularly with grading, histologic evidence of aggressiveness and pTNM, suggesting that actually the two populations (those with and without preinvasive lesion) are basically similar.
In a recent report, Nakahara and colleagues [6] found that AAH lesion were more frequently detected in patients with multiple primary carcinomas; they further pointed out that a significant correlation exists in their series between a history of preceding malignancy and multiple AAH. A recent case report [10] describing a patient with synchronous four primary lung adenocarcinoma associated with multiple AAH provides further evidence to this hypothesis. In our series we did not find similar results: none of our patients had multiple primary carcinomas and only 3 patients had a history of previous malignancy (all adenocarcinomas), but the difference with the total population was not significant. From our data, therefore, we cannot conclude that patients with AAH may be at an increased risk to develop malignant neoplasms. Similar findings have been reported by other authors [20]. Even for the CIS group, contrary to Massard and colleagues [21], we failed to demonstrate any correlation between a history of previous malignancy or the occurrence of metachronous neoplasms and the presence of CIS.
Survival rates of patients with associated pre-invasive lesions were compared with those of the entire population of resected lung carcinoma and with those of the corresponding invasive primary (adenocarcinoma for AAH and squamous cell carcinoma for CIS). We failed to demonstrate any significant difference between the two groups. Indeed, patients with associated pre-invasive lesion had a slight survival advantage, more evident in the CIS group. This finding parallels other reports from the literature both for CIS [21,22] and for AAH [23,24]. In a very recent report, Takigawa and colleagues [18] identified a subgroup of patients with AAH and early stage Ia in whom survival was even better than the corresponding adenocarcinomas without AAH. Survival results from our and other series are not easy to interpret. Since pre-invasive lesions are found incidentally in the resected specimen, they might presumably be in the remaining lobe(s) and by definition they will invariably progress in a definite period of time to invasive neoplasms. Thus it may be hypothesised that patients with associated pre-invasive lesions are at increased risk to develop second primary lung carcinomas in the remaining lobes. This, however, is not the case. Among all patients with pre-invasive lesions in the present series, only 9 had progression of the disease in the remaining lobes at follow-up: this was more evident among AAH patients (6 cases) than CIS (3 cases). The difference with the total population of resected lung carcinoma was not significant. A possible explanation of these unexpected findings might be that pre-invasive lesions occur as an accompanying condition only in the affected lobe and are therefore radically taken out by surgery.
In conclusion, associated pre-invasive lesions in patients resected for primary invasive lung carcinoma do not adversely affect long-term survival. AAH is strongly associated with adenocarcinoma, BAC or combined adenocarcinoma-containing tumors; CIS is associated with squamous cell carcinoma; DIPNECH is associated with typical carcinoid. The presence of associated pre-invasive lesions in the resected specimen does not seem to pose the patient at increased risk of developing a second primary lung carcinoma as compared to the total population of resected lung carcinoma. Based on these observations, therefore, it does not seem to be justified any modification of follow-up surveillance or any adjuvant treatment in this patient population with the possible exception of CIS at the bronchial resection margin.
|
Footnotes |
|---|
|
Appendix A. Conference discussion |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. DiscussionAppendix A. Conference...References |
|---|
Dr Ruffini: Yes. Adjuvant therapy may be offered in selected cases when the carcinoma in situ is at the bronchial resection margin. We had 5 of these patients. Of these, one received brachytherapy; the other one, external radiotherapy; and the other 3, only surveillance. This is the only strategy we used.
Dr Rea: Do you believe that this kind of adjuvant radiotherapy can have an influence in your results when you compared the group?
Dr Ruffini: No, I do not think so.
|
References |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. DiscussionAppendix A. Conference...References |
|---|
This article has been cited by other articles:
|
|
 |
|
  E. C. Johney, J. Pfannschmidt, R. J. Rieker, P. A. Schnabel, G. Mechtersheimer, and H. Dienemann Diffuse idiopathic pulmonary neuroendocrine cell hyperplasia and a typical carcinoid tumor J. Thorac. Cardiovasc. Surg., May 1, 2006; 131(5): 1207 - 1208. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
