Introduction

In Medicine, it is understood that a person is more than their disease. This has been well recognized, for example in 1927, Francis Peabody remarked, “The secret to the care of the patient is caring for the patient”1 whilst in 1908, Sir William Osler stated, “A physician is obligated to consider more than a diseased organ, more even than the whole man - he must view the man in his world”2.

However, the most cutting-edge recent development in biomedical science and clinical care, Personalized (or Precision) Medicine, has largely neglected the person. For instance, the American Cancer Society describes Personalized Medicine on its website as “a way health care providers can offer and plan specific care for their patients, based on the particular genes, proteins, and other substances in a person’s body”3. They go on to say, “Before starting treatment, doctors can test the cancer cells for certain gene and protein changes to help determine which treatments are likely to work best.… The two types of treatment most often used in precision medicine are targeted drug therapy (designed to attack a specific target on cancer cells) and immunotherapy (medicines used to help the body’s immune system attack the cancer)”3. However, as we highlight in this Comment, this was not how Personalized Medicine was intended to work. The person has gone missing in Personalized Medicine.

Considering the disease and the person with the disease

One of the earliest examples of personalized medicine was a study of an aggressive form of Non-Hodgkin’s Lymphoma, Diffuse Large B-Cell Lymphoma. (DLBCL) In that study, investigators analyzed lymphocyte samples from individuals with DLBCL, performing nearly 2 million gene expression measurements. They identified two distinct molecular subtypes with markedly different clinical outcomes: the activated B-like DLBCL group and the germinal center group. Having shown that these molecular signatures captured biologically distinct forms of the disease associated with different clinical courses, the investigators then turned their attention to differences among the individuals themselves. Using the International Prognostic Index (IPI) validated for BLBCL, they examined whether molecular differences in the tumor and clinical characteristics of the patients represented overlapping or distinct determinants of prognosis4.

The IPI incorporates a person’s physical performance capability, age, and disease severity. Among individuals with low clinical risk based on the IPI, those with the activated B-like subtype of DLBCL had significantly worse overall survival than those with the germinal center subtype. These findings show that gene-expression-based molecular classification and the IPI capture different but complementary factors, and when combined, can influence patient outcomes4.

This early demonstration of the power of Personalized Medicine also set an intellectual framework that emphasized the importance of differentiating the disease using molecular markers while also differentiating the individual. However, that framework has been lost in recent approaches, leading to “de-personalized” medicine.

Cancer arises, in part, from the accumulation of mutations in genes that regulate cell growth. These mutations can vary by organ, resulting in different types of cancer with unique genetic profiles. Understanding these distinct mutation patterns has made it possible to develop personalized therapies tailored to the specific molecular changes in each cancer.

For example, colorectal cancer often involves mutations in the KRAS (Kirsten Rat Sarcoma) gene, which have well-established prognostic significance and can guide treatment decisions5. Similarly, EGFR (Epidermal Growth Factor Receptor) mutations are known to drive the development of non-small cell lung cancer. In patients with these mutations, research shows that targeted therapies—such as tyrosine kinase inhibitors—can block the mutated receptor, helping to slow or stop the growth of cancer cells6.

By focusing these discussions exclusively on the molecular profile of the disease, distinctive features of the individual with the disease, which are also likely to modify the individual’s response to treatment, are ignored. There is evidence that all drug treatments, regardless of the therapeutic target, are affected by an array of features of the individual that operate through several key mechanisms. Age, for example, substantially affects treatment through changes in metabolism, organ function, and drug clearance7. Ancestry may play an important role, as some populations have higher frequencies of genetic variants that affect drug response8. Also, as illustrated in the DLBCL example mentioned earlier, comorbidities and individual-level performance status are independent prognostic factors in many cancer outcomes.

A person’s biography (or lived experience) can also impact treatment response9. The biography encompasses the social, behavioral, psychological, and environmental influences on a person that accumulate over the lifespan10. Biographies are shaped across time by both moments of pain, such as divorce or discrimination, and by moments of joy, such as the birth of a child or a long-awaited success, as well as by everyday stress, support, work, and rest. All these factors can impact our biology.

A recent study underscored the importance of personal biography by demonstrating a strong link between emotional distress and poorer clinical outcomes in individuals with advanced non-small cell lung cancer undergoing treatment with immune checkpoint inhibitors11. In the primary endpoint analysis, participants who reported emotional stress at baseline (111 individuals) had a significantly shorter median progression-free survival than those who did not (116 individuals; 7.9 months versus 15.5 months, P = 0.022). Similar patterns were observed for secondary endpoints, including lower objective response rates, reduced two-year overall survival, and greater declines in quality of life among those with higher distress levels. Elevated blood cortisol levels in these individuals further supported the biological plausibility of the association. Trials are now underway to test whether the co-administration of beta-blockers might improve the outcomes in individuals with emotional distress.

These findings highlight the role of how someone feels (i.e., their levels of distress or resilience) in altering the very biology of disease progression and treatment response. One way to understand the results of this study is to focus solely on the impact of emotional stress on individuals with NSCLC on their treatment response to immune checkpoint inhibitors. But doing so, physicians would miss the larger point that an individual’s lived experience profoundly affects personalized medicine, both in susceptibility to disease as well as response to treatment12. Biology and biography are intimately related. They cannot be separated, either in research or in practice. We have also come to learn that the biosocial mechanisms that drive these relationships are diverse and increasingly highlight the ways that social and physical environments interact with the person to influence disease risk and the response to therapy13.

Allostatic load is a concept introduced by Sterling and Eyer in 1988 that describes how the body maintains balance by adapting to repeated stress over time14. Unlike broad population-level metrics of distress, allostatic load describes a highly individual biosocial mechanism as it tracks how a person’s physiology, including their hormones, immune system, and neural circuits, adjusts or falters during life. In contrast, environmental epigenetics helps us understand how shared or long-term social exposures, such as poverty, isolation, or early childhood adversity, can leave highly stable molecular marks, often across groups or generations (also known as intergenerational trauma).

One striking example in model organisms showed that fruit flies, when raised in social isolation, were shaped by their lived experience. In this study, the investigators divided the flies into two groups: one group was kept together, and the other group was isolated. Interestingly, the flies exhibited normal behavior when housed in groups but showed impaired sleep and eating behavior when isolated from the other flies. When the investigators examined the fly heads, they found 214 genes, many of which were already known to be involved in eating behavior and sleep patterns15. The study was a remarkable demonstration that social influences have profound implications for underlying biology and suggests the role of biosocial experiences in precision medicine.

Concluding Remarks

It is insufficient to differentiate a disease based solely on its distinctive molecular characteristics and claim that the treatment is personalized. Scientific policy cannot continue to separate biology and biography in studies of disease risk and treatment response. The integration of biology and biography can be facilitated through advanced analytics, machine learning, and clinical decision support systems that enable researchers and clinicians to select the best treatment options for the individual while minimizing adverse effects. This multi-dimensional approach, which differentiates the disease from characteristics of the person, represents the core promise of personalized medicine. Moving beyond a one-size-fits-all treatment to therapy tailored to the unique biology of the disease and the biography of the person with the disease is key to progress medicine and uphold the implementation of Precision Medicine.