Next-generation sequencing has enabled unprecedented advances in oncology: personalized therapies are becoming possible, pathologists have gained unparalleled insights into the nature of tumors, and artificial intelligence can arrive at reliable diagnosis based on the immense data created by sequencing. Prof. Paul Hofman, a pathologist at the University of Côte d’Azur, France, wonders if doctors like him will even be needed in the future. We visited him at Nice Hospital.
Prof. Paul Hofman began his professional career as an oncologist. “I experienced it as a frustrating discipline: we always did the same thing, chemotherapy. And all the time we knew beforehand that it wouldn’t work for most of our patients.” It was just that nobody could say for whom it wouldn’t work.
At the age of 30, Prof. Hofman switched definitively to pathology. He assumed that he would have a more varied job and at the same time provide greater value to his patients.
If he could not fight cancer, he at least would strive to better understand it. He wanted to ensure that patients received the appropriate therapy. “Today, however,” Paul Hofman says with a smile, “I could imagine switching back to oncology. Since next-generation sequencing revolutionized the fight against cancer, there are so many new approaches and possibilities; for example, immune therapy. It’s a really exciting time for medicine.”
Today, his switch to pathology is more than several decades old. Prof. Hofman is currently in charge of the Surgical Pathology and Tumor Molecular Biology Department and of the Biobank Unit at the University Hospital of Nice.
The clinic is an architectural patchwork blanket that grows uphill and, due to the lack of space, has spread over the years into every imaginable gap on the site. At the top is a tower whose white facade is slowly graying. “Laboratoire” is spelled out vertically in large letters, one letter for each floor. Inside, its construction is as intricate as it is outside. Narrow and winding corridors, tight offices. A place in which it is easy to get lost.
Prof. Hofman looks like an image of the building, lean and tall, his lab coat no longer brilliantly white, his eyes tired – but in the corners of his mouth he has a friendly, gentle smile.
He is considered one of the leading lung cancer pathologists in Europe: a disease in which positive treatment outcomes depend on very early diagnosis. Prof. Hofman is studying, for instance, if liquid biopsy could become an alternative screening approach, allowing the disease to be detected months or even years earlier than with traditional imaging tools. In addition to his research work and managing the institute, Prof. Hofman holds functions in more than a dozen professional societies. A 70-hour week is the rule, not the exception.
The telephone rings incessantly during the conversation; often students from all over the world want to work with him. Employees stick their heads through the door to ask a question. The tone is conversational and friendly. When asked how he endures this day after day, Prof. Hofman shrugs and grins, “With the help of 15 espressos a day, at least.”
In reference to the new fascinating possibilities in oncology, of course, there is also a touch of coquetry. Next-generation sequencing (NGS) is a game changer in his specialty too. The revolution in the war against cancer starts in pathology and understanding how liquid biopsy samples can be best used for screening.
From Prof. Hofman’s point of view, this upheaval starts on the third floor of his institute. While traditional laboratory medicine is practiced on the ground floor where tissue samples from cancer patients are prepared, dissected into wafer-thin slices and dyed, up here are various sequencing machines that analyze the samples genetically. Recently, a QIAGEN
GeneReader NGS System was installed to help Prof. Hofman obtain the insights he seeks in his quest to advance research into better prevention and treatment of cancer.
If oncologists want to deploy precision medicine, i.e., treat patients individually based on their genomes and that of their tumors, Prof. Hofman and his colleagues will have to deliver precision pathology.
Because every tumor has genetic and immunological differences that affect how immunotherapy works, tumor sequencing makes sense in pathology. It can provide valuable data on the genetic profile of a specific form of cancer. Together with the interpretation of a patient’s exome and genome, it can result in the discovery of new biomarkers.
One such biomarker, which has been associated with immunotherapy response in multiple disease types, is tumor mutational burden (TMB). TMB measures the number of mutations within a tumor genome, and tumors that harbor more mutations have been shown to have a greater likelihood of an immunotherapy response. Now, Prof. Hofman is working toward developing methodologies to assess TMB in lung cancer.
NGS and new therapies such as immuno-oncology are weapons that bring victory in the fight against cancer closer to reality than ever. The use of NGS is also changing the nature of medicine. More and more, doctors must mediate between technologies. Today they need to be as familiar with algorithms as they are with anatomy.
In addition, sequencing technologies generate enormous amounts of data, the evaluation of which has long since overwhelmed a single person. Doctors face the fact that artificial intelligence sometimes makes a more precise diagnosis out of the terabytes of data than humans could ever do.
Should the integration of all patient-related data one day become reality, could software take care of all analysis in the hospital? That would mean integrating “big data” in clinical care. Prof. Hofman says: “If we want to combine all types of diagnostics, we need to look at big data and deep, or machine, learning.” After a short pause, he laconically adds, “That leads to the question: Do we want to have a doctor or are we better off just using a machine with a giant database?”
The GeneReader NGS System is the world’s first complete Sample to Insight solution for next-generation sequencing (NGS). It enables high sensitivity detection in liquid biopsy specimens and high throughput sample processing through compatibility with the QIAsymphony automation platform. The new GeneRead QIAact gene panels are compatible with FFPE and liquid biopsies. Integrated bioinformatics provide optimized, cloud-based analysis and interpretation of NGS data. In 2017, QIAGEN released new studies at the Association for Molecular Pathology (AMP) annual conference highlighting the outstanding analytical performance and easeof-use of the GeneReader NGS System.
In 2017, QIAGEN made great strides towards supporting greater use of molecular diagnostics for immuno-oncology therapies. The company has licensed novel biomarkers for microsatellite instability (MSI) and mismatch repair (MMR), and entered into a collaboration agreement with Bristol-Myers Squibb to develop gene expression profiles to help guide treatment with novel immuneoncology molecules currently under development. QIAGEN is also working on the application of its QuantiFERON Monitor (QFM) to monitor immune function. As part of the diagnosis of patients eligible for or treated by I-O therapy, QFM has the promise to provide important clues about the state of a patient’s immune system.
TMB is the hypothesis that an increased mutation rate leads to an increased number of mutated proteins, or neo-antigens, on the surface of tumor cells, capable of eliciting an immune response. This pre-existing immune potential is a major factor that will determine whether patients derive benefit from immune-oncology treatments.
Scientists at the Simons Center for Quantitative Biology at Cold Spring Harbor Laboratory (CSHL), New York, U.S., estimated that up to two billion genetic sequences could be available by 2025. That means the technology of gene analysis will generate reams of data, surpassing all other data-producing giants – including accelerator physics, astronomy, and the current recordholder, YouTube. While YouTube and astronomy produce as many as one to two exabytes a year, the amount of stored genetic information may be as high as 40 exabytes. One exabyte is one million terabytes, about the size of one million hard drives.