In the 1970s, cancer survival was poor for young children and older adults in the United States, as shown by data published in the Journal of the National Cancer Institute.
Great progress has been made since the 1970s, but improvements in outcome have been less impressive for cancer patients aged 15-39 years, as shown by research published in Cancer.
Patients aged 15-39 years have been designated by the National Institutes of Health (NIH) as “adolescents and young adults (AYAs),” and the lag in survival benefit has been termed “the AYA gap.”
The AYA gap persists in lymphoma patients, and an expert panel recently outlined differences between lymphoma in AYAs and lymphoma in other age groups.
The experts spoke at a special session of the AACR Virtual Meeting: Advances in Malignant Lymphoma moderated by Somali M. Smith, MD, of the University of Chicago.
Factors that contribute to the AYA gap
About 89,000 AYAs are diagnosed with cancer each year in the United States, according to data from the National Cancer Institute (NCI). Lymphomas and thyroid cancer are the most common cancers among younger AYAs, aged 15-24 years.
In a report commissioned by the NIH in 2006, many factors contributing to the AYA gap were identified. Chief among them were:
- Limitations in access to care.
- Delayed diagnosis.
- Inconsistency in treatment and follow-up.
- Long-term toxicity (fertility, second malignancies, and cardiovascular disease).
These factors compromise health-related survival, even when cancer-specific survival is improved.
Panelist Kara Kelly, MD, of Roswell Park Comprehensive Cancer Center in Buffalo, N.Y., noted that there are additional unique challenges for AYAs with cancer. These include:
- Pubertal changes.
- Developmental transition to independence.
- Societal impediments such as insurance coverage and disparities in access to specialized centers.
- Psychosocial factors such as health literacy and adherence to treatment and follow-up.
Focusing on lymphoma specifically, Dr. Kelly noted that lymphoma biology differs across the age spectrum and by race and ethnicity. Both tumor and host factors require further study, she said.
Clinical trial access for AYAs
Dr. Kelly emphasized that, unfortunately, clinical research participation is low among AYAs. A major impediment is that adult clinical trials historically required participants to be at least 18 years old.
In addition, there has not been a focused effort to educate AYAs about regulatory safeguards to ensure safety and the promise of enhanced benefit to them in NCI Cancer Trials Network (NCTN) trials. As a result, the refusal rate is high.
A multi-stakeholder workshop, convened in May 2016 by the American Society of Clinical Oncology and Friends of Cancer Research, outlined opportunities for expanding trial eligibility to include children younger than 18 years in first-in-human and other adult cancer clinical trials, enhancing their access to new agents, without compromising safety.
Recently, collaborative efforts between the adult and children’s NCTN research groups have included AYAs in studies addressing cancers that span the age spectrum, including lymphoma.
However, as Dr. Kelly noted, there are differences in AYA lymphoid malignancy types with a transition from more pediatric to more adult types.
Hodgkin lymphoma and primary mediastinal B-cell lymphoma
Panelist Lisa G. Roth, MD, of Weill Cornell Medicine, New York, reviewed the genomic landscape of Hodgkin lymphoma (HL) and primary mediastinal B-cell lymphoma (PMBCL).
Dr. Roth explained that both HL and PMBCL are derived from thymic B cells, predominantly affect the mediastinum, and are CD30-positive lymphomas. Both are characterized by upregulation of JAK/STAT and NF-kappaB as well as overexpression of PD-L1.
Dr. Roth noted that HL is challenging to sequence by standard methods because Reed Sternberg (HRS) cells represent less than 1% of the cellular infiltrate. Recurrently mutated genes in HL cluster by histologic subtype.
Whole-exome sequencing of HRS cells show loss of beta-2 microglobulin and MHC-1 expression, HLA-B, NF-kappaB signaling, and JAK-STAT signaling, according to data published in Blood Advances in 2019.
Dr. Roth’s lab performed immunohistochemistry on tissue microarrays in 145 cases of HL (unpublished data). Results showed that loss of beta-2 microglobulin is more common in younger HL patients. For other alterations, there were too few cases to know.
Dr. Roth’s lab is a member of a pediatric/AYA HL sequencing multi-institutional consortium that has been able to extract DNA and RNA from samples submitted for whole-exome sequencing. The consortium’s goal is to shed light on implications of other genomic alterations that may differ by age in HL patients.
Dr. Roth cited research showing that PMBCL shares molecular alterations similar to those of HL. Alterations in PMBCL suggest dysregulated cellular signaling and immune evasion mechanisms (e.g., deletions in MHC type 1 and 2, beta-2 microglobulin, JAK-STAT, and NF-kappaB mutations) that provide opportunities to study novel agents, according to data published in Blood in 2019.
By early 2021, the S1826 and ANHL1931 studies, which have no age restriction, will be available to AYA lymphoma patients with HL and PMBCL, respectively, Dr. Roth said.
Follicular lymphoma: Clinical features by age
Panelist Abner Louissaint Jr, MD, PhD, of Massachusetts General Hospital in Boston, discussed age-related differences in follicular lymphoma (FL).
He noted that FL typically presents at an advanced stage, with low- or high-grade histology. It is increasingly common in adults in their 50s and 60s, representing 20% of all lymphomas. FL is rare in children and AYAs.
Dr. Louissaint explained that the typical flow cytometric findings in FL are BCL2 translocations, occurring in up to 85%-90% of low-grade and 50% of high-grade cases. The t(14;18)(q32;q21) translocation juxtaposes BCL2 on 18q21 to regulatory sequences and enhances the expression of elements of the Ig heavy chain.
Malignant cells in FL patients express CD20, CD10, CD21, and BCL2 (in contrast to normal germinal centers) and overexpress BCL6 (in contrast to normal follicles), Dr. Louissaint noted. He said the Ki-67 proliferative index of the malignant cells is typically low.
Pediatric-type FL is rare, but case series show clinical, pathologic, and molecular features that are distinctive from adult FL, Dr. Louissaint explained.
He then discussed the features of pediatric-type FL in multiple domains. In the clinical domain, there is a male predilection, and stage tends to be low. There is frequent involvement of nodes of the head and neck region and rare involvement of internal lymph node chains.
Pathologically, the malignant cells appear high grade, with architectural effacement, expansile follicular pattern, large lymphocyte size, and an elevated proliferation index. In contrast to adult FL, malignant cells in pediatric-type FL lack aberrant BCL2 expression.
Most importantly, for pediatric-type FL, the prognosis is excellent with durable remissions after surgical excision, Dr. Louissaint said.
Follicular lymphoma: Molecular features by age
Because of the excellent prognosis in pediatric-type FL, it is important to assess whether young adults with FL have adult-type or pediatric-type lesions, Dr. Louissaint said.
He cited many studies showing differences in adult and pediatric-type FL. In adult FL, the mutational landscape is characterized by frequent chromatin-modifying mutations in genes such as CREBBP, KM22D, and EP300.
In contrast, in pediatric-type FL, there are frequent activating MAPK pathway mutations, including mutations in the negative regulatory domain of MAP2K1. These mutations are not seen in adult FL.
Dr. Louissaint noted that there may be mutations in epigenetic modifiers (CREBBP, TNFRSF14) in both adult and pediatric-type FL. However, CREBBP is very unusual in pediatric-type FL and common in adult FL. This suggests the alterations in pediatric-type FL do not simply represent an early stage of the same disease as adult FL.
Despite a high proliferating fraction and absence of BCL2/BCL6/IRF4 rearrangements in pediatric-type FL, the presence of these features was associated with dramatic difference in progression-free survival, according to research published in Blood in 2012.
A distinct entity
In 2016, the World Health Organization recognized pediatric-type FL as a distinct entity, with the following diagnostic criteria (published in Blood):
- At least partial effacement of nodal architecture, expansile follicles, intermediate-size blastoid cells, and no component of diffuse large B-cell lymphoma.
- Immunohistochemistry showing BCL6 positivity, BCL2 negativity or weak positivity, and a high proliferative fraction.
- Genomic studies showing no BCL2 amplification.
- Clinical features of nodal disease in the head and neck region, early clinical stage, age younger than 40 years, typically in a male with no internal nodes involved.
When FL occurs in AYAs, the diagnostic findings of pediatric-type FL suggest the patient will do well with conservative management (e.g., excision alone), Dr. Louissaint noted.
Two sizes do not fit all
The strategies that have improved cancer outcomes since the 1970s for children and older adults have been much less successful for AYAs with cancer.
As an oncologic community, we should not allow the AYA gap to persist. As always, the solutions are likely to involve focused clinical research, education, and communication. Effort will need to be targeted specifically to the AYA population.
Since health-related mortality is high even when cancer-specific outcomes improve, adopting and maintaining a healthy lifestyle must be a key part of the discussion with these young patients.
The biologic differences associated with AYA lymphomas demand participation in clinical trials.
Oncologists should vigorously support removing impediments to the participation of AYAs in prospective clinical trials, stratified (but unrestricted) by age, with careful analysis of patient-reported outcomes, late adverse effects, and biospecimen collection.
As Dr. Kelly noted in the question-and-answer period, the Children’s Oncology Group has an existing biobank of paraffin-embedded tumor samples, DNA from lymphoma specimens, plasma, and sera with clinically annotated data that can be given to investigators upon request and justification.
Going beyond eligibility for clinical trials
Unfortunately, we will likely find that broadening eligibility criteria is the “low-hanging fruit.” There are protocol-, patient-, and physician-related obstacles, according to a review published in Cancer in 2019.
Patient-related obstacles include fear of toxicity, uncertainty about placebos, a steep learning curve for health literacy, insurance-related impediments, and other access-related issues.
Discussions will need to be tailored to the AYA population. Frank, early conversations about fertility, sexuality, financial hardship, career advancement, work-life balance, and cognitive risks may not only facilitate treatment planning but also encourage the trust that is essential for patients to enroll in trials.
The investment in time, multidisciplinary staff and physician involvement, and potential delays in treatment initiation may be painful and inconvenient, but the benefits for long-term health outcomes and personal-professional relationships will be gratifying beyond measure.