Ethical considerations for hematologic precursor conditions Open Access

TO THE EDITOR:

Blood cancer precursor conditions such as clonal hematopoiesis (CH), monoclonal B-cell lymphocytosis, monoclonal gammopathy of uncertain significance, and smoldering multiple myeloma occupy a unique space in hematologic and oncologic care. These conditions are associated with increased risk of developing overt blood cancers, with limited actionability beyond surveillance in the present era. The ethical implications of testing and preemptively treating these have not been explored. Focusing primarily on CH, we aim to detail ethical considerations relevant to decisions to test for, and potentially treat, these precursor conditions.

Many people develop CH as they age and the vast majority will not develop blood cancers; however, those diagnosed with the CH subtypes of clonal hematopoiesis of indeterminate potential (CHIP) or clonal cytopenia of uncertain significance (CCUS) have an increased risk of developing myeloid malignancies such as myelodysplastic syndromes (MDS) and acute myeloid leukemia. A high-risk subgroup has been identified with >50% chance of myeloid malignancy within a 10-year period.¹ A clinical trial (NCT06802146) is being conducted to test the feasibility and safety of treating high-risk CCUS with the standard-of-care MDS therapy, hypomethylating agents. If effective, it would identify a potential pathway toward MDS/acute myeloid leukemia prevention. Subsequent confirmation of efficacy in larger randomized studies could fundamentally change management of these blood cancer precursors.

In 1996, Jonsen et al² warned that the proliferation of genetic testing would create a new category of “unpatients” diagnosed with risk factors. This group would be at risk of adverse psychological effects of an uncertain diagnosis, vulnerable to overtreatment with excessive testing and unnecessary interventions or undertreatment due to a cavalier approach by some physicians toward individuals with hypochondriacal concerns (viewed cynically as the worried well). As Jonsen et al predicted, advances in understanding of germ line and somatic hematologic genetic risk factors have produced countless “unpatients,” whose identities have shifted from healthy to “at risk.”³ 

Society is traditionally divided into discrete categories of sick and well—each with corresponding social expectations. The sick role consists of leniency toward some social responsibilities (eg, work productivity) along with added responsibility to seek treatment and participate in recovery.⁴ Patients with blood cancer precursor conditions occupy a liminal position between these roles; they are offered no relief from social obligations, yet may have some responsibility to submit to medical monitoring and potential risk-reducing interventions. Such groups with elevated cancer risk have been dubbed “previvors.”⁵ The ambiguity inherent to a previvor identity highlights ethical considerations for related but distinct decisions: when to test for precursor conditions, and whether preventive intervention should be offered.

The ethical obligation to disclose genetic risk information is tied to whether one can act to meaningfully reduce the risk or severity of the condition they hope to prevent (collectively termed as actionability).⁶ However, patients and clinicians may view actionability differently. A survey of patients and oncologists⁷ compared attitudes toward disclosure of genomic tumor testing results of uncertain actionability: 74.6% of patients were interested in receiving results, yet only 23.6% of clinicians would choose to share them. This disparity can limit productive provider-patient decision-making.

Moreover, individual patient differences affect whether knowledge of risk factors empowers or harms. One way to manage this differential reaction is to defer to patient autonomy, with the hope that patients can anticipate how they will react to a precursor diagnosis. In current practice, however, CH and CCUS are often incidental findings identified during genetic testing for other conditions such as hereditary cancer syndromes.⁸ Thus, patients can receive these diagnoses unexpectedly. This may violate a patient’s “right not to know,” an ethical principle debated in the context of sharing genetic testing results.⁹ For example, in a survey of breast cancer survivors,⁸ 54.5% expressed concern that CHIP testing may lead to a diagnosis “that they could do nothing about” and 30.9% “that they did not want to know about.” Still, 78.6% expressed interest in testing.

Whether or not precursor patients take it into consideration, actionability has ethical relevance in balancing benefits and harms of testing. At present, actionable outcomes for CH testing are limited, and studies evaluating the risks and benefits of testing are needed. Testing may result in unnecessary further workup at substantial emotional and financial cost. Ultrasensitive assays capable of detecting CH-associated mutations at very low variant allele frequencies¹⁰ (so-called “micro-CHIP”) may greatly expand the number of patients exposed to such diagnostic odysseys. Until the outcomes associated with micro-CHIP are better defined, these highly sensitive assays should be reserved for research use. Overall, patients may benefit more from focusing on other health-promoting activities. Early identification of hematologic malignancy could be beneficial, but only in instances where early intervention improves outcomes which currently remains unknown. Regular contact with hematology specialists, or health care providers more broadly, may also result in indirect health benefits by identifying CH-associated comorbidities such as cardiovascular disease.¹¹ Moreover, a precursor diagnosis can create opportunities for research participation, yielding potential individual and societal benefits. Table 1 outlines ethical considerations in testing for blood cancer precursors, including added considerations for preventive treatment. It presents many potential risks and benefits, which do not yet have evidence behind them, and will likely evolve and require refinement as data on testing and risk modification strategies mature.

Treating cancer requires a delicate balance between optimizing survival and quality of life (beneficence) and avoiding harms of toxic therapies (nonmaleficence). Jacobson¹² illustrated this dilemma in the context of medical oncology through his essay “The Oncologist’s Wager.” When a patient has no evidence of disease after surgery, adjuvant therapy is often pursued to reduce risk of recurrence. By doing so, many patients experience the harms of chemotherapy and/or radiation who would never have needed it, whereas others will recur despite treatment.

Hematologists face an analogous dilemma (Table 2). For clinicians, quadrant D represents the worst outcome: the opportunity to intervene was not taken, and the patient developed malignancy. Jacobson¹² warned that clinicians’ fear of this outcome and malpractice can drive overtreatment. From the patient’s perspective, the worst outcome may instead be quadrant B; exposure to chemotherapy which did not prevent cancer.

Precursor-associated anxiety may also affect treatment decisions. A study comparing patients with myeloma precursors (monoclonal gammopathy of uncertain significance and smoldering multiple myeloma) to those with multiple myeloma found no significant differences in mental health–related quality of life, distress, or anxiety.¹³ As illustrated by research on BRCA1/2 previvor decisions about prophylactic surgery,¹⁴ precursor-associated anxiety may bias patients toward preventive treatment. Those with greater cancer worry were found to be more likely to choose surgery, whereas those with less cancer worry tended to opt for surveillance. This is an example of how CH can be contextualized within a broader genetic risk framework familiar to oncologists and genetic counselors.

It is reasonable to offer CH testing even in the context of limited actionability (ie, surveillance only). Health information of uncertain actionability can still be valuable to patients^7,8 and can offer opportunities to participate in research that ultimately may yield valuable knowledge for precursor patients themselves. There may even be psychological benefit from discussion with an expert hematologist, which is distinct from taking any action to reduce risk.

On the other hand, given that patients have limited time and resources to invest in health, we must keep in mind that testing for and treating CH may not be the best use of finite personal resources. There is a burgeoning literature assessing “time toxicity” of cancer therapies¹⁵; adding such measures to CH trials could help to quantify time commitment of preventive chemotherapy vs surveillance. Societal resources are also not infinite, and the added costs to the health care system for testing, surveillance, and potentially treating CH need to be justified by evidence of benefit.

Finally, by offering CH treatment, hematologists may be primarily treating the anxiety caused by a precursor diagnosis with little evidence that intervening ameliorates anxiety. One way to generate such evidence would be to incorporate patient-reported outcomes in clinical trials of preventive chemotherapy for CH. Collecting patient-reported outcomes such as health-related quality of life would allow for a robust accounting of benefits, risks, and opportunity costs of preventive chemotherapy regardless of ultimate disease outcome. Doing so would respect the autonomy of future previvors, who will make decisions partially based on these data.

Acknowledgment: This cross-institutional collaboration was supported, in part, by Break Through Cancer.

Contribution: All authors contributed to writing the manuscript and approved the final version.

Conflict-of-interest disclosure: L.D.W. is an investigator in the NCT06802146 clinical trial referenced in this article and has consulted for Vertex and Sobi. A.E.D. is an investigator in the NCT06802146 clinical trial referenced in this article, and participated in advisory boards, and/or had a consultancy with and received honoraria from Bristol Myers Squibb, Agios, Novartis; and served on clinical trial committees or data and safety monitoring boards for Novartis, Agios, AbbVie, Kura, Geron, Servier, Keros, Shattuck Labs, and Bristol Myers Squibb. G.A.A. has served as a consultant for Novartis and Geron. E.C.B. declares no competing financial interests.

Correspondence: Gregory A. Abel, Division of Population Sciences, Dana-Farber Cancer Institute, 450 Brookline Ave, Boston, MA 02215; email: gregory_abel@dfci.harvard.edu.