From Moore’s Law to Molecular Alterations

The landscape of patient care has undergone significant changes in recent years, particularly with regard to new diagnostic technologies. Highly effective treatments have become available for many difficult-to-treat diseases but, to benefit, patients must first be tested for specific, clinically relevant molecular alterations. The growing need for molecular testing drives the development and implementation of technologies that make the process as quick, efficient and accurate as possible.

After a tissue sample undergoes morphological and immunohistochemical assessment, the remaining material is used for molecular testing (the details of which vary depending on tumor type, tissue of origin, and available treatments). To make the best use of the available sample, we need to look at various different types of molecular alterations, which means that we need to use multiple different platforms.

Many targets, many platforms

Full profiling of a tumor requires assessment of multiple targets of different types and often involves looking at aspects of gene mutations, chromosomal alterations and protein expression. We can’t achieve all of that by using just one technology – so we need to choose the combination that best meets our needs. The selection of technologies to assess these targets requires careful consideration to ensure that the techniques are timely, accurate, and suitable for (often extremely small) real-life samples. Turnaround time, cost and accreditation also need to be taken into account. The best example of this is lung cancer, in which there are now several targetable molecular alterations: ALK translocation (a chromosomal feature), EGFR mutation (a genomic feature) and PD-L1 expression (a protein expression feature). All of these results are dependent on each other, so oncologists need all three at once to know how to treat a given patient – and it has to be done using a very tiny biopsy, with the results available very quickly, so that treatment can be initiated without delay.

The reality is that a laboratory needs to have multiple platforms accredited and ready to use. We need platforms for DNA and RNA assessment; in our lab, for instance, we have three real-time PCR platforms, pyrosequencing technology and next generation sequencing facilities, which are all used every day on both tissue and plasma to fulfill our service’s daily requirements. We also use two types of automated immunostainers and fluorescence in situ hybridization technology to assess for protein and chromosomal alterations. The molecular profile demands integration of the information coming from all these platforms, performed in parallel. Consequently, the logistical implications are enormous – but the selection and preparation of samples is key. It’s possible that multiplex assessment of both gene and chromosomal alterations on real-time PCR or NGS, and protein/chromosomal alterations using multiplex algorithms on digitized slides, could become standard in the future, but we feel that this will be a gradual process and that, for the foreseeable future, we will need to maintain a wide variety of equipment.

Plasma testing allows rapid, noninvasive assessment of particular alterations and should be regarded as complementary to tissue testing in certain clinical circumstances, rather than a full surrogate. It will not replace the morphological diagnosis of cancer and the assessment of protein expression, but could become increasingly useful in gene and chromosomal alterations as long as the clinical sensitivity is acceptable. (At the moment, because we can’t distinguish tumor from non-tumor circulating DNA, the lack of detection of a targetable alteration on plasma cannot be regarded as a final result.) The possibility of plasma testing is huge in the context of monitoring patients to assess treatment efficacy and to detect early recurrence; however, clinical studies are needed for every marker in every type of tumor to validate them in clinical contexts.

With these new technologies, I have great hope that we will be able to further improve patient care. By looking for variants beyond those currently known to be clinically relevant, we could potentially identify new markers for more accurate prognoses and better predictions of treatment response.

Slide scanning is an example of a technology that – while already making a difference – is still developing. Nonetheless, it opens up the possibility of multiplex assessment of protein expression (rather than just gene alterations), which will likely become increasingly important in the future, especially in selecting patients for immunotherapy. Paired with sequencing, which has become a routine process in clinical laboratories, these complementary technologies allow us to deliver optimal treatment. As the healthcare landscape continues to evolve, technological advances will continue to make it easier for us pathologists to do the best for our patients. However, there is one aspect that’s more valuable than technology when it comes to ensuring optimum outcomes.

Collaboration is key

It’s well-established that collaboration between multiple clinical teams (including pathologists, surgeons, oncologists, and radiologists) is essential for optimal patient care. The pathologist’s work increasingly involves integrating complex information from multiple sources – morphology to protein expression to molecular alterations. Integration of this information, particularly in the context of molecular changes, demands extremely close work with laboratory scientists. Likewise, the synthesis of this information with clinical and radiological data demands that we work very closely with clinicians. It’s a group effort that extends beyond laboratory or clinic work and into multidisciplinary meetings.

The case of unknown primary carcinoma is a good example of the effectiveness of such cooperation. Recently, a patient presented to physicians with a mass in the liver. A biopsy showed adenocarcinoma, but immunohistochemistry gave no specific clues about the origin. The discussion at a multidisciplinary meeting revealed that there was also a lung lesion, which could not be biopsied. The team decided that the liver biopsy should be assessed for molecular alterations that might link to lung cancer (for instance, EGFR mutations, ALK translocations, or PD-L1 expression). Ultimately, the team identified a mutation conferring sensitivity to anti-EGFR drugs, and the patient was able to commence treatment for presumed metastatic lung cancer.

It’s worth pointing out that discovering an EGFR mutation in a malignant tumor does not necessarily convey any useful information on its own. The alteration only has predictive use if it is identified in lung cancer. That’s because the value of any specific alteration in a particular type of tumor must first be demonstrated in clinical trials. No mutation is agnostic for tumor type because of the enormous and poorly understood complexity of signaling pathways in cancer cells. It’s also worth mentioning that the presence of an actionable mutation is never a guarantee of response to a particular drug; it conveys only a statistically significant increase in likelihood of response.

Managing a technology-driven future

As pathologists, we do not regularly see patients ourselves – but by attending multidisciplinary meetings, we receive regular feedback about patients we’ve tested. We’re also increasingly seeing those patients return for secondary resistance mutations months or years later – so, in a sense, that’s a kind of feedback, too. Systematic auditing and data collection is essential in monitoring these patients, and that data is always worth publishing to keep the community informed.

Pathologists sit at the interface between clinical practice and testing technologies. As technology develops and there’s a tendency to investigate as much as possible, I think the pathologist’s role will become increasingly focused on utilization management – using their clinical knowledge to limit testing to what is actually going to inform clinicians and help patients. All in all, pathologists are applying their medical background – in tandem with their technical knowledge – to make a real impact on patient care in a way that has never been done before.

As the future becomes more technology-driven, I don’t think molecular testing will not take place in all pathology departments; instead, I believe it is likely to be limited to large, centralized hubs. However, the advent of digital technologies allows every pathologist to access all of the clinical and molecular data required to personalize patient care – even from distant locations – and, as a result, pathologists working today and in the future will need to move to a more patient-centric approach. No individual pathologist can lead this move, because no one person can have all of the answers. A change this big requires the involvement of the profession as a whole; everyone must get on board.

Newly available technologies open up more opportunities and create the potential for the field to grow – but, if we are to achieve the ultimate goal of personalized medicine, it is as vital as ever that we play an increasingly active role in patient care.