As cancer remains among the leading causes of death worldwide, with an estimated 1.8 million new cases diagnosed in 2020 (American Cancer Society), scientists and doctors are fighting against the clock to find new therapeutic approaches to beat this rife disease.  Today, millions of people are living with—or are survivors of—cancer.
The evolution of cancer management remained unchanged for decades, from ancient Egypt and Greece to the beginning of the 20th century, where standard treatment was radical surgery—often ineffective and fatal. Following World War II, with the discovery of cytotoxic antitumor drugs and the birth of chemotherapy, patients began going through a “one-size-fits-all” treatment regimen, usually including chemotherapy, radiation, and surgery. Cancer treatment was viewed from a macro perspective, with the hope that the chemotherapy’s strength would ultimately kill the tumor.
But this method came with a price, as chemotherapy isn’t selective to cancer cells; healthy cells are killed as well, resulting in the adverse events we are all familiar with – loss of hair, bone marrow suppression resulting in life-threatening anemia, infections, loss of blood and more. So, although this approach significantly reduced cancer-related death rates, it remained non-targeted, inaccurate, with very limited directionality towards a specific condition or type of cancer, and, most importantly, not patient-specific.
In the 1980s, targeted therapies were discovered. As cancer is very often driven by genetic mutations to the DNA in our cells, this method would target specific mutated proteins within the tumor. This approach played a key role in cancer treatment development and progression by identifying that DNA mutations could serve as cancer biomarkers, which could then be utilized for diagnosis, prognosis, and prediction of the cancer’s progression.
A good example of a targeted treatment approach in oncology can be seen in breast cancer cases, a leading cause of mortality among women (about 1 in 8 U.S. women will develop invasive breast cancer over the course of her lifetime.) We currently use targeted biological treatments aimed specifically at HER2+ cancers—cancers that test positive for a protein called human epidermal growth factor receptor 2, which promotes the growth of cancer cells—including trastuzumab (Herceptin®; Genentech/Roche), pertuzumab (Perjeta®; Genentech/Roche), and others. These targeted therapies, also called biologic therapy, use the body’s immune system or hormonal system to fight breast cancer cells.
Targeted therapies represent a landmark in cancer management and to this day, promising new drugs are being discovered and developed. However, with these advances come new challenges, including the optimization of therapy, recognizing and dealing with side effects, and, most importantly, the development of resistance. Targeting a specific mutated protein results in a rapid “evolutionary pressure” towards cancer cells that do not harbor the specific mutation, causing a tumor cell clone that is resistant to treatment.
The New Kid on the Block – Immunotherapy
For many years, investigators studied the interaction between our immune system and cancer. Evidence that our own body holds a powerful weapon against the cancer has accumulated over the last 40 years, but we struggled to find an effective drug that would harness this internal force. This changed when pioneers like James Allison, an American immunologist and Nobel laureate, discovered what would eventually become the first modern immunotherapy that was approved by the FDA in 2011.
Immunotherapy is probably of the greatest discoveries in the history of cancer management. We finally found a way to harness the immune system against this disease. Some believed that this was the ‘silver bullet’ that would allow us to overcome cancer – here we have a treatment that is potentially universal against all cancers, has a good safety profile, and, most importantly, demonstrates dramatic clinical efficacy.
With time we understood that while the immunotherapy was literally lifesaving for some patients, others were not responding, and for a small subset of patients, the new therapy even worsened their disease.
The Blind Spot
One of the challenges that have baffled oncology experts for years is: why is treatment effective in some cases but not in others? The search for predictive biomarkers began, but to date, there are still very few – most of them are not very predictive and if they are, it’s only for a small subset of patients. Why is that?
The traditional search for biomarkers focused on the interaction between the treatment and the tumor. For example, if the drug targeted a specific receptor-like PD-1, kits to measure PD-1 levels were developed. However, we learned that this approach was relevant only for select tumors and treatments. In recent years, investigators started to look into the tumor microenvironment, trying to decipher the complex biological interaction between the tumor, the treatment, and the human body (or the host.) Our body, a complex, sophisticated biological system, was completely overlooked for many years when it came to understanding resistance mechanisms to treatment, leaving physicians and oncologists with a glaring blind spot.
As scientists have worked to understand the complexity of the biological system, it has led us to the realization that although responders and non-responders look the same clinically, the difference between them is their body’s response to the treatment, otherwise known as the host response.
The Power of Host Response
A growing body of evidence demonstrates that cancer therapy can induce host-mediated local and systemic responses, many of which shift the delicate balance within the tumor microenvironment. This issue, known as “resistance to cancer treatment,” is a major topic of concern in oncology today.
On the one hand, cancerous cells are foreign to the body and should be seen as ‘cell invaders.’ But on the other hand, the origin of the tumor is our own cells and, as such, harbors the potential to be recognized by the body as a part of itself. This results in the immune system staying non-reactive to cancer, as it doesn’t view the tumor as a threat. Instead of fighting off the malignancy as it would a virus or bacteria, the body fights off the treatment, creating new blood vessels and protecting the cancer cells. This counterintuitive process leads to changes in the tumor microenvironment, ultimately supporting the growth of the tumor and potentially leading to disease metastasis (spread of cancer.)
In order to overcome this self-contradictory progression, it is crucial to take a deeper look at the bigger picture—the complex biological system. In the last 20 years, we made momentous progress in understanding the tumor DNA, mapping out the different mutations and the role of the tumor-related mutated proteins in cancer evolution. Host-related factors, specifically host-related proteins – the building blocks and “engines” of the host-related biological process – remained an uncharted territory for a long time.
Proteomic Profiling – Revealing the Blind Spot
Proteins have many roles in the body. They help repair and build the body’s tissues, allow metabolic reactions to take place, and coordinate bodily functions. Proteins are the blueprints to understanding everything that is taking place inside our body. Since every person’s biological system is different, each patient will respond differently to their given treatment. Looking at the tumor in order to understand and predict response is only one piece of the puzzle. To get the full picture, we should be looking at the proteins in the blood and at the tumor’s environment—connecting the drug-tumor-host triangle vertex.
Proteomic profiling involves analyzing a broad range of select proteins in a series of patient blood samples, with the first collected prior to treatment, the next collected after the first dose of treatment, and several blood draws taken as the treatment progresses. The tracked changes in the select series of proteins between the samples can then be used to serve as a predictive tool to determine the likelihood of the treatment’s success and further guide physicians in tailoring treatment plans for individual patients.  With this approach, physicians can better predict which treatment will work for each individual patient, optimizing patient care outcomes while reducing treatment costs and minimizing potential adverse treatment side effects. Through this process, physicians will be able to easily assess whether the chosen treatment plan will be accepted or rejected by the patient’s body.[
We as physicians and clinicians are blinded by the tumor’s micro-environment, neglecting the complexity of the human biological system. Proteomic profiling allows us to see the disease from a different angle, completing the picture and, most importantly, differentiating between those patients who will respond, and those who won’t. The next step is finding the right tool to conduct proteomic profiling.
The Promise of Machine Learning
To successfully measure, monitor, and improve oncology treatments, the application of machine learning is a necessity. The number of points and potential interactions in proteomic profiling is almost infinite, and machine learning helps us assess and digest this information, transforming it into clinical sense. This is the future of precision oncology.
Taking advantage of high-performance computing capabilities, machine learning algorithms can now achieve reasonable success in predicting risk in certain cancers by assessing multidimensional clinical and biological data. With just a simple blood test, machine learning approaches are enabling us to obtain explanations for patient-centric predictions, providing us with strong tools to achieve preferred clinical outcomes.
Essentially, precision oncology demands from us to take all of the moving parts of the disease into account if we want to be successful – the patient, the treatment, the tumor, and the host response. If we see the full picture and assess exactly what’s happening in the microenvironment and the body, the blind spot in cancer treatment can finally be revealed.
 Cancer Facts & Figures 2020. American Cancer Society. Online Last accessed on March 22, 2021.
 Falzone L, Salomone S, Libra M. Evolution of Cancer Pharmacological Treatments at the Turn of the Third Millennium. Front Pharmacol. 2018 Nov 13;9:1300. doi: 10.3389/fphar.2018.01300. PMID: 30483135; PMCID: PMC6243123.
 U.S. Breast Cancer Statistics. Breastcancer.org. Online. Last accessed on March 21, 2021
 Daenen LG, Houthuijzen JM, Cirkel GA, Roodhart JM, Shaked Y, Voest EE. Treatment-induced host-mediated mechanisms reducing the efficacy of antitumor therapies. Oncogene. 2014 Mar 13;33(11):1341-7. doi: 10.1038/onc.2013.94. Epub 2013 Mar 25. PMID: 23524584.
 Katz OB, Shaked Y. Host effects contributing to cancer therapy resistance. Drug Resist Updat. 2015 Mar;19:33-42. doi: 10.1016/j.drup.2014.12.002. Epub 2014 Dec 26. PMID: 25575621.
 Voloshin T, Voest EE, Shaked Y. The host immunological response to cancer therapy: An emerging concept in tumor biology. Exp Cell Res. 2013 Jul 1;319(11):1687-95. doi: 10.1016/j.yexcr.2013.03.007. Epub 2013 Mar 19. PMID: 23518388.
 Harel M, Lahav C, Jacob E, et al. 21 Plasma-based proteomic profiling as a tool for predicting response to immunotherapy in melanoma patients. Journal for ImmunoTherapy of Cancer 2020;8:doi: 10.1136/jitc-2020-SITC2020.0021
 Shaked Y, Lahav C, Harel M, et al.A proteomics-based platform for predicting response to immunotherapy and personalising treatment plans. Molecular Analysis for Precision Oncology| Volume 31, SUPPLEMENT 5, S1217, October 01, 2020
Featured image: © 2019 – 2021 OncoHost; Used with permission.