It's not uncommon for a question here to be about the a pathologist's terminology on a report that equivocates about whether a lesion is bronchioloalveolar carcinoma (BAC) or another form of adenocarcinoma, perhaps "well-differentiated adenocarcinoma", especially if it has a radiographic appearance of a hazy infiltrate or many small ground glass opacities. Meanwhile, there's a new classification of lung cancer subtypes that obliterates the term BAC, instead favoring a definition of adenocarcinoma in situ, classifying small non-invasive lesions previously called BAC as a pre-malignant condition. How have the changes over the past few years changed how we should approach BAC?

I would have to say that the new reassignment of BAC as adenocarcinoma in situ hasn't taken the lung cancer world by storm and that I still think of the clinical entity as BAC. For the preceding decade, the definition the pathologist's used technically excluded a lesion with even 1% or 5% of the lesion being invasive as being called BAC, even if it acted for all the world like BAC. Clinicians learned not to be too hung up on a pathologist's technical definitions and tended to define BAC more functionally/operationally. General oncologists and expert lung cancer specialists alike managed BAC based on the overall picture of how it behaved if it looked like a BAC pattern.

A few years later, I think the biggest change has been the ongoing transition from clinical/histologic (pathology-based) decision-making to more molecular decision-making. Three years ago I wrote about my own "conversion" to being convinced that molecular rather than clinical factors reign supreme, based on the results of the IPASS trial that showed that even Asian never-smokers with an adenocarcinoma demonstrated a remarkably better progression-free survival and response rate with first line chemotherapy than with an EGFR tyrosine kinase inhibitor (TKI) like Iressa (gefitinib) or also probably Tarceva (erlotinib) if they don't have an EGFR activating mutation, while they were far better served with first line EGFR TKI-based therapy if they have an EGFR mutation.

In my mind, these results extend beyond Asian never-smokers and also serve to change the landscape of NSCLC in general to an increasingly molecularly-defined paradigm. BAC isn't and shouldn't be treated with one systemic therapy or another just because it's called BAC: instead, BAC is a subtype of adenocarcinoma that has a higher probability of being associated with an EGFR mutation, and perhaps also an ALK rearrangement (the latter is not known with certainty at this point), but in the absence of either of these molecular signals of high utility for targeted therapy, chemotherapy is still the strongest option for initial systemic therapy. And because Alimta (pemetrexed) is also a particularly helpful chemotherapy for patients with an adenocarcinoma, I also favor using it as a compelling choice for chemotherapy, combined with a platinum, when treating BAC (although we haven't seen results of a small completed trial that tested Alimta specifically in patients with BAC).

Interestingly, many lung cancer experts, myself included, also think that our current use of NSCLC histologies to prioritize chemotherapy choices is also a less ideal interim point than using molecular markers, perhaps like thymidylate synthase (TS), to determine why patients with an adenocarcinoma tend to respond well to Alimta (and why some with an adenocarcinoma don't). Our expectation is that in several years the molecular profile will be a more accurate predictor than histologic subtype of what systemic therapy will work best (and perhaps we can identify a rare patient with a squamous NSCLC who is likely to benefit from Alimta, just as we occasionally find a patient with a squamous NSCLC who has an activating EGFR mutation and can respond very robustly to an EGFR TKI).

Otherwise, my approach to the treatment of BAC is really the same as my approach to other lung adenocarcinomas, except that I am inclined to be very judicious about when to recommend starting therapy if it appears to be a very slowly growing process. I think it's very important to get as good a sense as possible of the pace of the cancer (such as by assessing how slowly or quickly it's changing between two or more scans over time), and if it's changing very slowly (i.e., you need to squint to see a millimeter of growth over a 6-12 month period), I am very disinclined to recommend rushing into a systemic therapy for which the side effects of treatment may be worse than the disease. In a setting in which we may realistically hope to be following patients with few or no cancer-related symptoms for years and even potentially beyond a decade, I don't favor rushing through treatment just for treatment's sake, long before it's needed, and then going through years of investigational therapies because you've exhausted every therapy with a track record.

As I indicated above, the clinical picture remains hugely important in how I think of an individual patient with BAC. I feel a little hypocritical trying to focus on the evidence as often as possible but then relying on experience and clinical judgment in predicting and managing BAC, but the reality is that there still isn't a lot of data on what to expect in the very clinically heterogeneous world of BAC, and patients with very different natural histories (clinical patterns of progression in the absence of treatment) are still lumped together. Some patients with BAC will have an extremely indolent process that progresses at a glacial pace over years to decades, even when it's multifocal and incurable (and I would argue best thought of as if you're managing diabetes), while others with something that is also called multifocal, stage IV BAC can exhibit a much more aggressive pattern of fulminant progression. I sometimes feel that referring to the "art of medicine" is code for "I can do whatever I want without any evidence to justify it", but I do believe that BAC is a situation in which the rule-based approach is not as helpful as considering what the individual characteristics of a person's cancer are telling you.