This article is more than 5 years old. Information may no longer be current.
Outpatient models offer tremendous potential for transplantation and beyond
A few weeks ago in clinic, I had the pleasure of seeing several of my patients with leukemia whose diseases were in remission.
Remission in acute myeloid leukemia is a precarious place. Although we are able to achieve remission for many of our patients, the resulting freedom from disease is usually not indefinite. The reality of relapse has led to increasingly heroic attempts to consolidate initial responses, and the advent of reduced-intensity conditioning regimens has made allogeneic stem cell transplantation a possibility for many of our older patients with this disease.
Although allogeneic transplant makes cure possible, many patients with high-risk diseases experience relapse anyway. In some instances, we can think about allogeneic transplant as an immune-based therapeutic modality for patients whose diseases may have been incurable from the start — an essentially palliative approach that extends life relative to best supportive care, not unlike the use of chemotherapy in metastatic solid tumor malignancies.
This reality, though, begs the question: How much toxicity is acceptable for patients with a low probability of cure? Is it fair to allow our patients to decide this?
An acceptable price
The experience of allogeneic transplant is so complicated and so life-changing that, despite our best attempts at pre-transplant education, it is hard for our patients to understand this until they live it. Is it possible to accurately describe degrees of transplant treatment intensity? Until we have reliable and valid patient-reported outcomes to help us, it may not be.
William Wood
That day in clinic, I spent some time with a patient whose treatment trajectory was arguably better than most. He had been diagnosed with AML arising from myelodysplastic syndrome, and never underwent induction chemotherapy, achieving remission with hypomethylating therapy alone. He went on to a reduced-intensity allogeneic transplant, was discharged after 2.5 weeks, and remained home with the exception of a 3-day re-hospitalization.
His disease returned. After decitabine and donor lymphocyte infusion, he was again in remission, which continues to date. I don’t know how long his remission will last — hopefully indefinitely — but I do know that it has been a small victory that he’s largely avoided prolonged hospitalizations with the exception of his transplant admission.
For many of our patients — particularly those with difficult diseases and a high probability of relapse — this may seem to be an acceptable price to pay for the possibility of freedom from disease, whether temporary or permanent.
Many programs have become interested in and have implemented outpatient transplant programs for reasons related to costs and resources. I also think this makes sense, though, from a patient-centered perspective and, perhaps, even an ethical perspective. However, outpatient transplants must be more than
procedures that begin in the outpatient environment but progress to several weeks of inpatient hospitalization.
What can we do to allow our transplant patients to stay out of the hospital? I can think of a few components that will be important to ensuring the long-term success of this objective.
Risk projection
We must project who is most at risk for transplant-related toxicity and complications.
In other areas of medicine, risk projection has allowed models of triaging patients up front by the amount of surveillance and resource investment they are likely to require during the subsequent period of risk. EDs are an example of this in the acute setting, where certain beds are allocated to patients by risk category at triage. In a longitudinal setting, high- and average-risk obstetric services have different models of outpatient follow-up and resource investment during pregnancy.
Although one could argue that all transplant patients are at significant risk of treatment-related complications, there are clearly many patients who go through the early and longer-term periods of post-transplant care with minimal issues, and others whose care becomes extraordinarily complex and intensive. Can we predict who might fall into which category? Besides broad categories of types of transplant — autologous, standard reduced intensity allogeneic, standard myeloablative allogeneic or alternative donor transplants — data are emerging from models such as the hematopoietic cell transplantation-specific comorbidity index (HCT-CI) to suggest that underlying illness before transplant can predict transplant-related morbidity.
Others, such as myself, are investigating whether functional tests and other direct measures of physiologic reserve, such as cardiopulmonary exercise testing, might offer similar prognostic information. Reliable models that project high vs. standard risk could allow arrangement of resources to most effectively meet the outpatient needs of these different subpopulations.
Effective monitoring
Effective monitoring during the post-transplant period for the prediction and detection of illness is essential.
Increasingly, post-transplant surveillance protocols within the outpatient transplant clinics allow the detection of disease and the initiation of treatment before the signs and symptoms of significant illness — or sometimes any illness — become apparent. An example of this is the initiation of antiviral therapy as pre-emptive treatment for the detection of cytomegalovirus viremia.
The new era of mobile and digital health offers ever more sophisticated methods for longitudinal, granular surveillance of physiologic data. An example in another area of medicine is an electronic wireless tattoo — which is being developed at the University of California, San Diego, and recently received a grant from the Bill & Melinda Gates Foundation — that promises to eventually deliver wireless outpatient electronic fetal monitoring.
With time, wireless transmission of vital signs and, perhaps, laboratory data will become available for outpatient transplant recipients from the comfort of home. Additionally, I and others are demonstrating that frequent self-report of electronic patient-reported outcomes is feasible, even among acutely or chronically ill patients, and can complement physiologic data as a longitudinal toxicity output. It is not too difficult to imagine that some of the surveillance functions of an inpatient environment will become available for homebound outpatients and, when accompanied by alarms, can trigger appropriate interventions.
On top of this, we may be able to layer models of biomarker and “-omics” data onto this basic surveillance structure to dynamically understand and predict evolving risk for specific transplant-related complications. Within transplant, some groups have recently published promising acute graft-versus-host disease predictive biomarker data, and minimal residual disease measurements to predict relapse are becoming increasingly refined. Outside of transplant, a recent intriguing publication in Cell described how one investigator used his own “-omics” data to predict his previously unknown risk of type 2 diabetes, and then to observe and act upon the development of this disease after a respiratory syncytial virus infection. Could we imagine similar dynamic biomarker monitoring in the early post-allogeneic transplant period?
Health care availability
The final component involves the availability of appropriate health care when and where it is needed by patients.
As we understand more about monitoring for and managing specific transplant-related complications and relapse, excellent care can increasingly be delivered in an algorithmic way by providers specifically trained to do it. Functionally, and for practical reasons related to a growing transplant physician shortage, this means a growing reliance upon nurse practitioners, physician extenders, and pharmacists, a sophisticated work force ready to accomplish these complicated tasks.
To borrow phrases from The Innovator’s Prescription, transplant medicine is moving away from an “intuitive” medicine composed of “solution shops,” and toward “precision medicine” composed of “value-added processes.” It will be challenging to arrange outpatient resources in the most efficient manner, especially because complications can arise in the day or night, sometimes requiring the acuity of a routine clinic visit and other times needing the acuity of urgent or emergent intervention. Some tasks might be accomplished by telemedicine-facilitated videoconferencing and electronic prescriptions, whereas others might need an after-hours urgent/emergent care triage at the transplant hospital.
Conclusion
Clearly, a lot of work will need to be done to figure out how to make outpatient and even home-centered stem cell transplantation feasible, less toxic and fair in the level of risk required of patients with diseases that are exceptionally difficult to cure. The above concepts are just starting points for discussion as we look toward the future in this field.
Perhaps most exciting, though, is the promise that the potential of these ideas holds. Accurate risk projection, sophisticated physiologic surveillance and precision medicine are not just about the future of stem cell transplantation. They are key pillars of the future of all of medicine. If we can figure this out in transplant and other areas of oncology, we can create a proof-of-concept model for health care in general.
William Wood, MD, is assistant professor of medicine in the division of hematology/oncology at the University of North Carolina in Chapel Hill. He may be reached at william_wood@med.unc.edu, or follow him on Twitter (@WoodBD). Disclosure: Dr. Wood reports no relevant financial disclosures.
References:
- Chen R. Cell. 2012;148:1293-1307.
- Christensen CM. The Innovator’s Prescription: A Disruptive Solution for Health Care. New York: McGraw-Hill; 2008.
- Cleeland CS. J Clin Oncol. 2011;29:994-1000.
- Wood WA. Biol Blood Marrow Transplant. 2012;18:497-504.