This month's Journal of Thoracic Oncology includes a landmark article, written by a multidisciplinary group of lung cancer experts that features several of the leading lung cancer pathologists in the world, that is attempting to do no less than present a new categorization of the pathology of lung cancer, focusing primarily on adenocarcinomas, but also touching on other lung cancer subgroups. Among the concepts included is a proposal that the term bronchioloalveolar carcinoma (BAC) be discontinued and recategorized based on whether patients have any invasive component to their lung cancer or not, and whether is is mucinous or non-mucinous. As significant as it is to redefine an entire disease of BAC, the concepts it introduces have implications that actually lead to a fundamentally different way of thinking about lung cancer, or at least NSCLC.

The volume of material covered in this paper is far too comprehensive to cover in a post or two, or even a few more. Still, a few central themes emerge. The over-arching concept is that the general view of "NSCLC" as a category is being replaced by multiple distinct subtypes that may be defined by details of how they appear under a microscope and the molecular markers they do or don't feature. This started with the idea of squamous and non-squamous, or perhaps adenocarcinoma and large cell neuroendocrine carcinomas as their own entities, but there is a growing sense that there are relevant differences in the natural history (the way a cancer behaves on its own) and responses to treatment that may be revealed by looking for greater granularity in the subtypes that have historically been aggregated as NSCLC.

The new classification system also highlights the need for more tissue by actually offering different classification systems depending on whether the tissue obtained is from a resected, surgical specimen or from small biopsies, such as a core biopsy or fine needle aspirate. This reflects the fact that a small biopsy can only tell us about the appearance of the small piece of cancer sampled, while the cancer may have different features only detectable by reviewing a larger area. The authors make a very directed plea to obtain enough tissue to provide more information than just "NSCLC not otherwise specified", with no additional refinement of a subtype of lung cancer. Also discussed in the context of the size of the biopsy specimen is the importance of molecular marker studies to help re-classify subtypes such as the subpopulation with an EGFR mutation, KRAS mutation, or ALK rearrangement.

It is also no accident that the large group of authors includes specialists from a wide range of disciplines, as an emerging feature of the brave new world of lung cancer is that different subtypes of lung cancer may be defined in dimensions of clinical behavior without therapy, radiologic appearance, molecular marker patterns, and response to systemic therapies.

And then, specifically regarding the subtypes that have historically been defined as BAC, the new system makes a fundamental distinction between non-invasive adenocarcinoma in situ (meaning "in place") (AIS), which is predicted to have a 100% five-year survival from lung cancer, and invasive forms that represent the predominant life-threatening form of lung cancer. Though there hasn't been a lot of research dedicated just to non-invasive BAC, the implication of this work is that it may well be appropriate to follow different algorithms for management compared with the invasive form that has the capacity to metastasize throughout the body. The paper makes the key point that AIS almost exclusively tends to apply to non-mucinous adenocarcinoma; what we've historically called "mucinous BAC" would generally now be considered mucinous adenocarcinoma, which would be subclassified as having a "predominantly lepidic (scale-like) pattern" if it has very little invasive component.

The implications of this new classification are very significant, one being that defining an individual's lung cancer is about to become far more complex. This will be very unsettling to physicians and patients alike, since the breakdown of a much simpler system that we felt we understood is replaced by a new system in which there are gaping holes highlighting what we don't yet know. We are transitioning from a time, just a few years ago, when almost the only division you needed to consider is whether a lung cancer is SCLC or NSCLC, to a new era in which patient prognosis and optimal treatments are defined by a specific histologic subtype and pattern of molecular markers that we are only beginning to appreciate. Future developments are likely to include molecular markers that we may or may not have even begun to focus on, and other factors, such as patterns of metastatic spread, may well emerge as future points of refining our categorization and treatment patterns far more than they do today.

As unsettling as this is, as we realize how much we don't yet know, I can also foresee that this new, far more refined system will almost certainly allow us to follow an optimized management algorithm for each person's lung cancer. It's as if we're beginning to recognize that there are individual colors that just come out as brown because we have historically mixed them all together. We can see a much clearer picture by keeping the individual colors distinct.