WASHINGTON, DC—Some of the new gadgets and techniques used to diagnose and treat lung cancer look like they may have a role in a science fiction movie, but many can improve diagnostic speed and accuracy and prolong the life of people with lung cancer. Speakers at a symposium here at Washington Adventist Hospital reviewed some of the most promising advances.
Staging non-small-cell lung cancer (NSCLC) by mediastinal lymph nodes is critical, and if the nodes are negative, the patient is a good candidate for surgery, noted Paul De Leyn, MD, PhD, Consultant Thoracic Surgeon for Lung Cancer at University Hospitals Leuven in Belgium. If the nodes are positive, the patient should receive multimodality treatment, with or without surgery.
Although cervical mediastinoscopy is the most effective technique in such situations, it is invasive. When used with computed tomography-positron emission tomography (CT-PET), the positive predictive value increases, but findings still need to be proven histologically.
Some physicians prefer other, less invasive techniques such as transbronchial needle aspiration (TBNA) and endoscopic ultrasound needle aspiration. However, their sensitivity and negative predictive value are lower than mediastinoscopy, the diagnostic gold standard, he continued. The specificity of mediastinoscopy is 85% to 90% and sensitivity, upwards of 90%.
The positive predictive value of CT-PET in evaluation of the mediastinum is still fairly low—about 80%, Dr. De Leyn said. “This means that CT-PET can be falsely positive 20% of the time. Before denying a patient a potential cure of surgery, positive findings should be verified with fine needle aspiration.”
Mediastinoscopy carries such a low risk that it is done on an outpatient basis. However, if a complication occurs, it can be catastrophic: severe hemorrhage; pneumothorax (which occurs 20% to 30% of the time); air embolism; or damage to the esophagus, recurrent laryngeal nerve, or tracheobronchial tree.
Despite these benefits, there has been some controversy, noted Marc Margolis, MD, Chief of Thoracic Surgery at George Washington University—specifically about when to perform the procedure.
“Some groups perform it routinely on all patients irrespective of whether there are suspicious lymph nodes. Others do it if there are suspicious nodes that need to be evaluated preoperatively. Also, many institutions don't have practitioners who can perform TBNA. And if TBNA or transesophageal biopsy does not yield sufficient information, then mediastinoscopy is appropriate.”
Eric D. Anderson, MD, Associate Professor of Medicine in the Division of Pulmonary, Critical Care, and Sleep Medicine at Georgetown University, explained that transbronchial needle aspiration allows biopsy of subcarinal, paratracheal, and hilar lymph nodes. It is relatively noninvasive but has a diagnostic yield only about 71% of the time.
He noted that at Georgetown, pathologists perform Rapid Onsite Evaluation (ROSE) of the tissue gleaned from a TBNA—right at the time it is being performed. “Most of the time we know right away if the node is malignant so treatment can begin forthwith,” Dr. Anderson said.
Improved imaging modalities may be combined with traditional TBNA to increase the yield of diagnostic tissue. With the addition of real-time CT imaging during bronchoscopy, the exact location of a biopsy can be targeted, especially in suspected recurrences, treated neoplasms, or large lesions with necrosis. The positive predictive value then is about 98%. CT-guided TBNA carries minimal risk of minor bleeding, infection, and pneumothorax, but, Dr. Anderson noted, it is time consuming, creates a significant amount of radiation exposure for the physician, and involves a good deal of equipment and patient movement.
Another technique, endobronchial ultrasound, which has been extensively studied for examination of the lymph nodes, uses a small convex probe attached to the distal tip of a flexible bronchoscope. It allows evaluation of the submucosal and peribronchial structures and has a number of advantages over conventional TBNA, he said:
- A clearer image of mediastinal anatomy.
- Can evaluate local invasion of early endobronchial disease and sample the lymph nodes by guiding biopsy location, assessing the depth of invasion and evaluating tumor invasion of other structures.
- Allows real-time TBNA under direct ultrasound guidance.
- Sensitivity and specificity for detecting malignant lymph nodes are superior.
- Is more reliable for biopsies of small nodes and those in difficult locations.
- Requires fewer aspirates to make a diagnosis.
Endobronchial ultrasound (EBUS) has no side effects, but the equipment is expensive and only a few physicians know how to perform it, Dr. Anderson said. However, in centers where the technique is performed, the number of mediastinoscopies usually declines. “In fact,” he said, “there's reason to believe that TBNA and EBUS done together may replace mediastinoscopy as the diagnostic gold standard.”
Most lung cancer is radiologically occult until it is so far advanced that cure is almost impossible, said Timothy C. Kennedy, MD, Clinical Professor at the University of Colorado Health Sciences Center. But before it gets to that point, autofluorescence can highlight dysplasia, premalignant cells, or superficial carcinoma in situ.
Irregular surfaces are enhanced (fluoresce) at varying intensities upon exposure to a blue light—three or four times better than white light, he continued. Why blue? “We don't really know, but we think it has to do with something in the cells.”
Moreover, during autofluorescence, malignant cells glow in a different color than normal cells do, thus directing the physician to changes in the bronchial epithelium. These changes are particularly common in current or former heavy smokers. And, he added, “About 30 percent of all lung cancers start in large central airways, so this is a technique that is particularly well suited to relatively early diagnosis.”
The smaller the lesion, the more likely it can be diagnosed by fluorescence, as compared with white light.
The other advantages of autofluorescence are: Real-time tracking of a lesion is possible as it moves with the patient's respiration; Choice of route that avoids the ribs and transverse processes; Easy adjustability of approach if position of lesions changes; Ability to perform biopsy at same time; Can also determine extent of surgery.
“When you combine autofluorescence with a bronchoscopy, you can kind of cruise through the area and see what's going on in the surface of the mucosa of other cells,” Dr. Kennedy said. Fluorescence is safe and effective with CT guidance and can substantially reduce the procedure time, thus reducing the amount of radiation patients and physicians are exposed to.
The best candidates for the procedure, Dr. Kennedy said, are people suspected of having NSCLC—for example, those who have had a sputum test that reveals suspicious cells.
Virtual bronchoscopy is a real-time, three-dimensional trip down the airways without instrumentation, although often done with an actual bronchoscopy, Dr. Anderson explained. The procedure involves electromagnetic tracking, with the patient lying on a pad in which sensors are imbedded, and simultaneous visualization of a CT reconstruction using thin (1–3 mm) cuts. The more cuts, the better the detail.
Many lesions occur in parts of the bronchial tree and lung that are not reachable by a traditional bronchoscope, but increasingly sophisticated techniques have improved the resolution and accessibility of fiberoptics so that complex thoracic pathology, as well as nodules as small as 5 mm, can be displayed in three dimensions.
Virtual bronchoscopy extends the reach of the bronchoscope by introducing a disposable probe covered by a flexible sheath extended through the working channel of a flexible bronchoscope. The steerable probe contains a transmitter that is tracked by the electromagnetic pad. Thus, the operator always knows exactly where the tip is.
The probe can then be guided to a peripheral lung lesion using the virtual bronchoscopy images. It can be removed and replaced by various endoscopic tools so a biopsy can be taken.
The method is noninvasive and can visualize the airway beyond what one can see with traditional instrumentation, such as the distal side of obstructions—valuable in patients with airway stenoses secondary to bronchogenic carcinoma, for example. Virtual bronchoscopy also can identify candidates for stenting or transbronchial biopsies of small peripheral lesions.
At Georgetown, research is under way to improve virtual bronchoscopy applications for mediastinal lymph node sampling, Dr. Anderson noted. By varying the transparency of the bronchial wall, lymph nodes can be visualized in real time on the virtual bronchoscope while TBNA is performed, thus potentially improving the diagnostic yield.
But virtual bronchoscopy is not perfect, he cautioned. It may not see subtle flat mucosal lesions, and cannot put in a stent or debulk tumors. Still, this technique might be used in the future for placement of brachytherapy or focal treatments of chemotherapy.
Video-Assisted Surgery (VATS)
Some of these new therapies can be curative, but most are palliative: laser therapy, electrocautery, argon plasma coagulation, cryotherapy, brachytherapy, photodynamic therapy, and stents.
Dr. Margolis explained that many of these procedures are minimally invasive and can be performed by video-assisted surgery, a robot-like way to enhance what the human eye can see. VATS, which involves advanced video technology, computers, and high-tech electronics, requires skills that not every surgeon has mastered, but some of the more basic procedures (fluid drainage, biopsies, pleurodesis, decortication) are relatively easy. “However, a lobectomy performed by means of VATS or the da Vinci robot is more technically demanding and difficult.”
He added that most surgeons agree that VATS confers benefits similar to those of laparoscopy: i.e., less postoperative pain, smaller incisions (and no rib spreading), less inflammatory response, and shorter hospital stays.
VATS can be controversial for cancer surgery: There are no studies that compare its benefits to lobectomy via thoracotomy, and tumor seeding is possible. A 2006 study in Annals of Thoracic Surgery by Robert McKenna, MD, et al, reviewed 1,100 consecutive lung cancer patients (ages 39 to 85) who underwent VATS lobectomy with lymph node dissection between 1992 and 2004. The VATS procedures were standard for NSCLC, and did not compromise surgery for patients with poor performance status.
There were nine deaths (none intraoperative or due to bleeding), 932 had no postoperative complications, and 45 required blood transfusion. The patients stayed in the hospital for an average of three days, but 180 were discharged on the first or second postoperative day. Twenty-eight patients required conversion to a thoracotomy.
Dr. McKenna said that in his experience none of the concerns that caused controversy are warranted. The duration of air leaks, chest tube insertion, and hospital stays were all shorter compared with thoracotomy, and patient deaths and cases of pulmonary impairment were the same for both procedures. Still, VATS is not appropriate for all lobectomies or all surgeons: “It requires excellent video skills and knowledge of thoracic anatomy, and because there is a risk of bleeding, the surgeon should be prepared for the possibility that the surgery might turn into an open thoracotomy.”