3

Late Effects of Hematopoietic Stem Cell Transplantation

Doug Rizzo, attending physician at the Medical College of Wisconsin and senior scientific director at the Center of International Bone Marrow Transplant Research (CIBMTR), opened the second session by saying that the prior speakers laid a foundation on which to consider hematopoietic stem cell transplantation (HSCT) as a curative therapy that may lead to disability. Recapping session one, Rizzo reiterated that recipients of a transplant do not generally return to a pretransplant life expectancy, and they often have significant late effects that affect morbidity, mortality, work status, and quality of life. Lee had noted that 50 to 80 percent of survivors return to work, said Rizzo, and also that changing trends in transplantation could lead to an increase in disability as the number of recipients and survivors increase. Rizzo said some of the late effects associated with HSCT are consistent with those experienced with other cancer survivors, but some are unique, chronic conditions that affect HSCT recipients specifically. Although children generally recover better from HSCT than adults, their longer life span means there is a risk of long-term implications. The speakers in this session, said Rizzo, focus on three topics related to late effects of HSCT: immune recovery, graft-versus-host disease (GVHD), and childhood transplant recipients.

IMMUNE RECOVERY IN PATIENTS

The main causes of death for patients who have undergone HSCT using an unrelated donor are relapse, GVHD, infection, and toxicity, said Miguel-Angel Perales, who is the chief of the Adult Bone Marrow Transplantation

Service at the Memorial Sloan Kettering Cancer Center. Infection can occur before the patient has engrafted, during the early posttransplant and engraftment period, or as a late effect, and can be bacterial, viral, or fungal. Delayed immune recovery can be caused by a number of factors, including neutropenia, barrier breakdown, impaired cellular and humor immunity, and slow recovery of B and T cells. Acute and chronic GVHD are particular drivers of delayed immune recovery, said Perales, and the treatment for GVHD usually involves immunosuppression. Perales said that while allogeneic transplant recipients have some additional complications compared to autologous transplant recipients, there are many similar issues in both populations.

Perales shared data from a trial that was aimed at developing new approaches to reduce the risk of chronic GVHD (Pasquini et al., 2018). The trial enrolled patients under 65 years old who had a matched donor; they were randomized to three treatments: bone marrow with tacrolimus methotrexate (the standard of care), bone marrow with posttransplant cyclophosphamide, and bone marrow with a T cell-depleted graft using CD34 selection (ClinicalTrials.Gov NCT02345850). Perales explained that the latter approach selects out stem cells and removes the T cells. The primary endpoint for the study was a composite endpoint that included relapse-free survival and being free of GVHD. Results showed that patients with CD34 selection had a significantly lower risk of chronic GVHD; however, there was no difference between the three arms on the primary endpoint (Luznik et al., 2022). This was partly attributable to increased mortality in the CD34 arm, Perales said. One of the main drivers of increased mortality in the CD34 group was infection (Figure 3-1). Perales said that by removing the T cells from the graft in order to reduce GVHD, “the price that we pay” is increased toxicity-related infections and decreased survival. Other studies have also shown an association between low levels of T cells and increased mortality (Small et al., 1999; Goldberg et al., 2017; de Koning et al., 2021). These studies, said Perales, demonstrate that there is a direct relationship not only between delayed immune recovery and the risk of infection, but also between delayed immune recovery and increased mortality.

In the past year, numerous studies have examined the effect of COVID-19 on patients undergoing stem cell transplantation (e.g., Shah et al., 2020; Sharma et al., 2021; Sultan et al., 2021). At the same time, said Perales, there has been an opportunity to look closely at what happens to the immune system in these patients. One study (Shah et al., 2020) found that all of the parameters of the immune system—CD3 T cells, CD4 T cells, CD8 T cells, CD19 B cells, and NK cells—dropped with acute COVID-19 infection. An important finding, said Perales, was that lymphopenia with COVID-19 infection did not impair long-term recovery of the immune system. Another study (Aydillo et al., 2020) looked at HSCT patients with prolonged COVID-19 infection; these patients have an incompetent immune system that is unable to clear the virus. Three

patients were shedding viable SARS-CoV-2 for more than 20 days, and one patient had detectable viable virus for more than 60 days. Another group of researchers sequenced the virus found in these patients and found that mutations are more likely to arise in these patients (McCarthy et al., 2021). Perales said that these results demonstrate the importance of protecting HSCT patients from infection, both to protect the individual patient but also to protect the general population from virus mutations, such as the Delta variant of SARS-CoV-2.

Perales shared stories about some of his patients to demonstrate the effect of transplantation on returning to school and work. One patient, a 55-year-old New York City school teacher, had a transplant because of her acute myelogenous leukemia. Her posttransplant recovery was routine, with no GVHD and the expected delayed immune recovery. Despite her routine recovery, she was out of work for more than 18 months because of the requirement that she show proof of revaccination, combined with the timing of the school year. Another example, said Perales, is a 50-year-old emergency room physician with a diagnosis of T cell acute lymphoblastic leukemia. He also experienced a routine recovery after transplant, but his return to work was delayed because his employer needed proof of his immune system’s response to the MMR (measles, mumps, and rubella) vaccine. To demonstrate this, samples had to be sent to a specialized lab at the Centers for Disease Control and Prevention that used advanced testing to measure his immune response.

When considering the return to work and school for HSCT patients, Perales said it is critical to keep in mind that infection is a main cause of mortality after a transplant. One of the main drivers of infection is immune

reconstitution, and there is a direct relationship between immune recovery and the ability to mount responses to vaccines, said Perales. While response to COVID vaccines increases over time, it generally remains decreased compared to healthy controls (Tamari et al., 2021). Perales said that understanding the relationship between HSCT, immune recovery, and vaccine response is critical to our ability to protect patients and allow them to return to work with adequate protection to the environment.

GRAFT-VERSUS-HOST DISEASE

As other speakers have discussed, GVHD is a major cause of morbidity and mortality following an allogeneic stem cell transplant, said Betty K. Hamilton, the interim director of the Blood and Marrow Transplant Program at the Cleveland Clinic Taussig Cancer Institute. GVHD occurs when the transplanted donor immune system (the graft) recognizes the recipient (the host) as foreign, and the donor system mounts an immunologic attack against the recipient’s organs. Despite the current approaches used to prevent GVHD, acute GVHD occurs in 30 to 70 percent of patients, and 20 to 50 percent will develop chronic GVHD (Hill et al., 2018). Hamilton said that GVHD and its treatment, which usually involves prolonged immunosuppression, increase the risk of infection, organ impairment, poor quality of life, and survival.

Acute GVHD, said Hamilton, is classically characterized by either an inflammatory red maculopapular rash; elevated liver enzymes; gastrointestinal symptoms such as nausea, vomiting, or diarrhea; or a combination of any of these symptoms together. Chronic GVHD can affect a number of body systems, including skin, nails, scalp, hair, mouth, eyes, esophagus, lungs, muscles, joints, fascia, and genitalia. The diagnostic criteria for chronic GVHD were developed by experts at the National Institutes of Health (NIH) in 2014 (Jagasia et al., 2015), after many years with a lack of clear consensus for diagnosis, staging, and treatment. Importantly, said Hamilton, the criteria take into account the functional effect of GVHD as well as the physical manifestations. In addition to acute and chronic categories of GVHD, there is also late acute GVHD and overlap GVHD, which involves aspects of acute and chronic. In addition to GVHD, said Hamilton, patients can experience many other late effects and chronic conditions after transplant, including problems with cardiovascular and pulmonary function, infections, malignancies, and psychosocial issues. Chronic GVHD can interact with and accelerate some of these effects, said Hamilton.

The incidence of chronic GVHD has been increasing in recent years, said Hamilton. This is likely caused by a number of factors, including improvements in early mortality leading to more long-term survivors, the use of peripheral blood stem cell grafts, and the use of unrelated donors. Several

studies have found that chronic GVHD is the leading cause of nonrelapse related death after transplant (Wingard et al., 2011; DeFilipp et al., 2021), said Hamilton (Figure 3-2). A study led by a NIH consortium dedicated to clinical research in immune-mediated disorders after transplant (DeFilipp et al., 2021) found that the mortality risk associated with chronic GVHD does not plateau but continues to increase over time, and that moderate to severe GVHD and GVHD that affects the lungs are both associated with higher nonrelapse mortality.

Patient-reported outcomes have also been studied in chronic GVHD, said Hamilton. Patients with moderate or severe GVHD report worse quality of life, lower functional status, and higher symptom burden, and are more likely to be taking prescription medications for pain, anxiety, and depression, compared to patients who never had GVHD, had mild GVHD, or whose GVHD had resolved (Lee et al., 2018). Lee et al. (2018) found that up to 40 percent of patients with active chronic GVHD were unable to work for health reasons, compared to 12 percent of those with resolved chronic GVHD and 15 percent who never had chronic GVHD. Certain manifestations of GVHD have been associated with more permanent impairments, a longer duration of treatment, and a decreased ability to return to work or school, said Hamilton. These manifestations include moderate to severe eye involvement, scleroderma, joint fascia involvement, severe esophageal stricture, and any lung involvement (Fatobene et al., 2019). One study (Hamilton et al., 2020) enrolled patients with chronic GVHD and followed them for 18 months (Figure 3-3). At the time of enrollment, about half of patients reported GVHD-related disability, which was defined as a combination of GVHD criteria, along with a decline

in performance status or functional status. Of these patients, about half experienced a progression in disability by 18 months. Of the patients who were not disabled at the beginning of the study, half met the criteria for disability at the 18-month mark.

Hamilton told workshop participants about a survey conducted of patients living with chronic GVHD (Hamilton et al., 2021). It was a cross-sectional online survey of adult patients who reported having chronic GVHD in the previous 5 years. Respondents answered questions pertaining to demographics, disease diagnosis, work status, chronic GVHD symptoms, and effect on activities of daily living. The survey found that 63 percent of respondents reported some work disability, defined as taking disability leave at any time or leaving a job because of chronic GVHD. In addition, 67 percent reported severe physical disability (limitations in activities such as housework, using the restroom, and eating), and 47 percent reported severe cognitive disability (limitations in managing finances, social interactions, or using a computer).

As HSCT approaches and supportive care have improved, said Hamilton, the number of transplant survivors continues to grow. There are currently 200,000 HSCT survivors in the United States, and it is estimated that there will be more than half a million by 2030 (Battiwalla et al., 2017). As this population grows, so will the number of people at risk for chronic GVHD, other late effects, and death. Hamilton said that while all-cause mortality

declines as patients survive longer, nondisease-related mortality increases over time (Bhatia et al., 2007). Major causes of death that are not related to the primary disease include subsequent malignancies and cardiopulmonary disease (Wong et al., 2020). The proportion of subsequent malignancies continues to increase with longer survival (Baker et al., 2019), and the associated mortality risk significantly exceeds the corresponding risks in the general population, said Hamilton. People who survive HSCT are also at risk of living with chronic health conditions; by 15 years after a transplant, around 70 percent of allogeneic transplant survivors have one or more chronic health conditions, and many have more than one (Sun et al., 2010).

While transplant recipients have a generally positive perception of their own recovery at 1 year, said Hamilton, data demonstrate that many long-term survivors of HSCT experience multiple late effects of their transplants (Figure 3-4). According to research, said Hamilton, survivors with three or more late effects have lower physical functioning, lower likelihood of full-time work or study, and higher likelihood of limitations on their usual activities (Khera et al., 2012).

CONSIDERATIONS FOR CHILDREN AND ADOLESCENTS

In general, children have higher rates of survival after HSCT than adults, said Stella Davies, the director of Bone Marrow Transplantation at Cincinnati Children’s Hospital Medical Center. There are two major reasons for this difference, she said. First, children usually have fewer prior comorbidities. Second, children have an intact functioning thymus, a gland that helps T cells recover and reconstitute. In general, said Davies, immune reconstitution in

children is better than in adults, and as a result, infections are less prevalent and children recover more quickly.

However, HSCT does come with a cost for children, Davies said. Children who received a transplant are at higher risk of chronic conditions than children who were treated conventionally, and this heightened risk continues throughout life (Eissa et al., 2017). These chronic conditions cover a wide range of issues, from hypertension to severe physical limitations. Similarly, said Davies, an increased risk of mortality persists even decades after childhood HSCT (Holmquist et al., 2018). Davies noted, however, that there have been significant improvements in mortality over time; the incidence of mortality is lower for children who have had transplants more recently (Holmquist et al., 2018). Given the challenges associated with HSCT, asked Davies, “Why would anybody undergo this therapy?” It is important, she said, to consider the effects of transplant in the context of the primary disease. For example, sickle cell disease is a significantly damaging disease that can lead to stroke and cognitive decline; HSCT offers an opportunity to potentially halt progressive decline (King et al., 2019).

A number of factors influence a child’s recovery and potential for disability after HSCT, said Davies. Some of these factors are similar to those of adults, while others are very different. Davies described some of these similarities and differences (Table 3-1). For both children and adults, chronic GVHD is an extremely important factor in recovery; a patient with GVHD is likely to have a decreased ability to fight infections and increased disability.

Orthopedic issues are also important for both children and adults, although the specific risks vary somewhat, said Davies. In adults, there are age-related risks such as arthritis and joint damage, whereas in children the bones and joints are generally healthy. However, a complication that is sometimes seen in children is avascular necrosis of bone (death of a small area of the bone). This is a particular problem in teenagers and girls, said Davies, and can lead to the need for joint replacement or significant pain and limitation of mobility. Other orthopedic risks for children include reduced bone density, frailty, and risks related to certain primary diseases.

One factor that is particularly important for children is the effect of dealing with a prolonged, critical illness. During the transplantation process, there can be episodes of hospitalization, infection and other complications, and time away from school and family. Children with long illnesses can miss out on development opportunities that are necessary to progress from childhood to adulthood, including education and socialization. Davies said that behavioral issues can become more prominent, and returning to school after a period of illness can be very difficult for children who may be behind developmentally or be socially awkward because of being in the hospital and “around adults for a couple of years.” These types of issues, said Davies, can make it difficult for children to progress in school and eventually in employment.

One big difference between adults and children is the type of diagnoses that lead to HSCT. In adults, the majority of transplants are performed for malignancies. Children also receive transplants for malignancies, but there are also many other disorders that lead to transplant and some of these disorders are associated with cognitive deficits and orthopedic complications. Finally, the use of radiation is generally avoided in children because of the retardation of brain development; however, it is used for acute lymphoblastic leukemia because of a very clear survival benefit. As children who have received radiation get older, they will often need additional help in education on issues such as focus, attention, and other aspects of cognitive function, said Davies.

Several specific diagnoses are associated with a high burden of likely or expected disability, said Davies (Table 3-2). Leukodystrophies and mucopolysaccharidoses are childhood dementia disorders; children achieve milestones and then progressively lose their abilities. The goal with these disorders is to treat them early, before significant cognitive decline has occurred. However, said Davies, these diseases are unpredictable and can worsen during transplan-

TABLE 3-1 Similarities and Differences in Clinical Factors Influencing Disability in Children and Adults Who Receive Hematopoietic Stem Cell Transplants

SOURCE: As presented by Stella Davies, November 15, 2021.

TABLE 3-2 Diagnoses Associated with High Burden of Late Disability

SOURCE: As presented by Stella Davies, November 15, 2021.

tation, so patients may be left with permanent disabilities. Sickle cell anemia patients who receive a transplant are usually those with the most severe phenotypes; these patients may already have lung damage, cardiac damage, history of stroke, and vascular necrosis. These morbidities put these patients at a higher risk of disability after a transplant. Children with Fanconi anemia and dyskeratosis congenita are susceptible to progressive decline owing to the nature of the disorders, said Davies. Finally, patients with brain tumors may undergo resection of the tumor and multiple autologous transplants to substitute for the need for very high radiation doses in childhood; resulting disabilities may include cognitive impairment, motor impairment, and growth impairment.

Davies emphasized that it is critical to consider the diagnosis that led to the transplant when assessing disability. One important advancement that has reduced disabilities is the introduction of newborn screening, which is now active in every state in the United States, said Davies. Specific screenings vary by state, but can include leukodystrophies, sickle cell anemia, and mucopolysaccharidosis. Early diagnosis can lead to early treatment, including transplant and gene therapy, and early treatment can lead to better outcomes.

Cognitive difficulties are a common and important effect related to HSCT in children, said Davies. There are a variety of causes of cognitive impairment in this population, including the underlying diagnosis, exposure to radiation, missed education, encephalitis, sepsis, stroke, and thrombotic

microangiopathy. It is critical that children have access to appropriate educational services, particularly when their return to school is delayed because of immunosuppression or complex medical needs. Despite these challenges, most HSCT survivors generally do well, said Davies. Survivors of HSCT in childhood who were surveyed at a median of 25 years after transplant (Wu et al., 2021) reported scores comparable to the general population on a measure of cognitive function; impaired function on this measure was associated with hearing issues, history of stroke or seizure, and sleep disturbances. Between 18.9 and 32.5 percent of survivors reported impairments in task efficiency, memory, emotional regulation, or organization, compared to an expected 10 percent in the general population. Another study looked at employment outcomes for HSCT recipients, and found that only half of long-term survivors were working part- or full-time and 19 percent were on sick leave or disability, compared to 6 percent of the general population (Winterling et al., 2018).

In addition to the effect of HSCT on children, it is critical to also consider the effect on caregivers, Davies said. Children recovering from HSCT need a high level of care, including medication management, observation for infection, physical therapy at home, and frequent doctor visits. These responsibilities limit the ability of primary caregivers to work, particularly when patients have GVHD or other late complications. In addition, she said, caregivers may have their own disabilities caused by the stress and trauma of caring for an ill child, and they may be financially challenged owing to the inability to work regularly. Davies said that supporting caregivers is essential for recovery of the family unit as a whole.

DISCUSSION

Following the presentations, Rizzo led panelists in a discussion by asking a series of questions from the workshop participants.

Q1: How do the long-term effects vary between patients who do get hematopoietic stem cell transplants and patients who do not get transplants?

Rizzo noted that some of the data presented compared patients who received a transplant to the general population, and he asked panelists to comment on how transplant survivors compare to similar patients who did not receive a transplant. Hamilton said that it can be complicated to compare these populations, particularly for malignant diseases that are life threatening. There are some studies for nonmalignant diseases; this type of research requires good databases to follow patients’ chronic conditions and late effects over time. Rizzo said that some data suggest that there are additional comorbidities and additional complications in patients receiving HSCT compared to those who

had more conventional cancer treatment. He added that it can be challenging to examine outcomes in rarer diseases owing to a lack of sufficient numbers.

Q2: Are there patients who largely recover from moderate or severe GVHD? Are there biomarkers or other ways to predict who will experience GVHD, or how severe GVHD will be?

The first-line treatment for chronic GVHD is corticosteroids such as prednisone, said Hamilton. Approximately half of patients respond to this treatment, while the others go on to require second- or third-line therapies (Wolff et al., 2011). There are currently three U.S. Food and Drug Administration (FDA)-approved treatments for steroid refractory chronic GVHD. Certain types of chronic GVHD tend to be more difficult to treat, she said, including patients who experience sclerotic symptoms, fibrotic impairments, or lung GVHD. Hamilton said that while the presence of comorbidities does not necessarily predict the likelihood of developing GVHD, patients with higher comorbidities who do develop GVHD have a higher risk of nonrelapse mortality and other complications after a transplant. The field is interested in identifying ways to predict how a chronic GVHD case may develop or progress, she said; in 2020, working groups convened by the NIH Consensus Conference on Chronic Graft vs. Host Disease conveyed a lot of interest in determining clinical and biologic markers that can help predict who will develop chronic GVHD and severe GVHD.¹ Developing these markers, said Hamilton, can help identify patients who really need treatment or who may not need aggressive treatment. Rizzo suggested that as evidence in this area emerges, it could be useful to develop systematic risk models to inform clinical decision support.

Q3: What are the management strategies for GVHD in children, and are these useful in adults?

The treatment landscape for GVHD in the pediatric population has “changed radically just in the last 2 to 3 years,” said Davies. The standard treatment for GVHD is steroids, which are particularly toxic to a growing child. Three new, nonsteroid drugs are now available to treat GVHD: ibrutinib, ruxolitinib, and belumosodil. Each of these works in a slightly different way, has nonoverlapping toxicities with steroids, and is currently used in people who have an unsatisfactory response to steroids, she said. These drugs have

___________________

¹ See https://ncifrederick.cancer.gov/events/conferences/2020ChronicGVHDConsensus (accessed February 4, 2022).

been used with a good response in children, and Davies conveyed her hope that pediatric GVHD treatment will soon be able to transition to steroid-free treatment.

Rizzo asked Davies to comment on the use of nonpharmacological therapies for GVHD. She responded that for appropriate patients, there is great value in aggressive physical therapy. She noted that children respond well to creative games and competitions. Nutrition is always important, she added, and children who have received transplants can become overweight from steroids, or underweight from poor absorption or esophageal strictures. Perales added that supportive care is critical in both adult and child populations, and that working with a multidisciplinary team can help get patients the care they need for the symptoms they have. Hamilton agreed and added that it unfortunately “takes a lot of resources” to have the types of programs that adults need, such as physical therapy, social support, and employment assistance. Perales added that patients often have limited access to physical therapy and are only allowed a certain number of sessions based on their insurance coverage; many run out of sessions “before they’re anywhere close to being done with dealing with their complications.” Rizzo said that as we consider disability in HSCT patients, there may be an opportunity to find ways to improve access to rehabilitation in order to decrease long-term disability.

Q4: If a patient doesn’t respond to a hematopoietic stem cell transplant, what are their other treatment options?

If a patient experiences relapse or progression of the underlying disease after a transplant, said Perales, the next step could be another transplant or treatment with CAR T cells. However, in order to do a second transplant, a patient must be well enough to withstand the treatment. For many diseases, particularly the malignant diseases, there are limited options if a patient does not respond. Davies added that the goal is to “try and win the first time” because additional transplants come with a cumulative burden of treatment and additional complications.

Q5: Given the risks of infection, what should posttransplant care look like?

Traditional guidelines, said Perales, are to reimmunize your patients at particular times following transplantation. Another approach is to measure parameters of immune reconstitution, and immunize based on those results. However, there are still challenges in successfully building immunity in patients after HSCT. For example, Perales told participants about a patient who was several years removed from her transplant, met all of the parameters

for immune reconstitution, and demonstrated responses to the standard vaccines; despite this, she was admitted to the hospital with pneumocystis pneumonia. A gap in the immune system allowed an opportunistic infection to take hold, he said, and we need more research in this area.