The National Comprehensive Cancer Network’s guidelines for the management of non-bulky diffuse large B-cell lymphoma (DLBCL) recommend that, for patients with early-stage disease, shortening the number of cycles of chemotherapy and adding radiation therapy to the treatment regimen is a viable treatment option. However, results from an analysis of data from the National Cancer Data Base (NCDB) found that this option is not widely used – and differs according to access to care.
“Clinical and observational studies provide conflicting assessments of relative effectiveness of these strategies, and it is unclear what drives the decision-making in the community,” the authors, led by Jaleh Fallah, MD, from the Department of Medicine at Memorial Hospital of Rhode Island in Pawtucket, Rhode Island, and colleagues wrote.
To better understand patterns of care among patients with DLBCL and if administering six versus three cycles of R-CHOP plus radiation therapy (ACt/RT) is a practical treatment option, Dr. Fallah and colleagues evaluated factors determining treatment selection and affecting overall survival from data within the National Cancer Data Base (NCDB) – a national registry capturing >70 percent of incident cancers in the United States.
Out of 31,236 patients with stage I/II DLBCL, 5,464 (17.5%) received ACt/RT; this proportion remained unchanged between 2004 and 2012 (p=0.69 for trend). “Despite encouraging results of clinical trials such as SWOG 0014, only one in six Americans with early-stage DLBCL received ACt/RT in the R-CHOP era,” the authors observed, “possibly reflecting a discomfort with abbreviating chemotherapy on the part of physicians and patients.”
The use of ACt/RT “markedly” varied by primary site of DLBCL, from 3 percent for abdominal nodes to 35 percent for breast (p<0.0001). Despite treatment guidelines, 46 percent of patients with testicular DLBCL did not receive any RT, they added.
So, what accounted for variations in treatment selection? The likelihood of receiving ACt/RT did not vary based on age or gender (joint p=0.14), but did vary significantly based on the following factors:
- Black race (odds ratio [OR] = 0.77; p=0.015)
- Presence of two or more comorbidities (OR=0.70; p=0.0001)
- No insurance or insured by Medicaid (OR=0.69; p=0.0004)
- Stage IB/II lymphoma (OR=0.56 to 0.24; p<0.0001)
Availability and access to care also affected whether or not patients received RT. When RT was not available on site or if the facility was more than 30 miles from the patient’s residence, patients were less likely to receive ACt/RT (OR=0.45 and 0.74, respectively; p<0.0001 for both). Its use did not differ between community and academic hospitals, though.
Overall, the authors found that variations in treatment selection were most commonly attributable to unobserved patient-level preference (including bulky tumors and interim response; 65 percent of cases) and to patients’ recorded DLBCL-related factors (including stage and primary site; 22% of cases). Other factors influencing treatment variation included unobserved hospital-level preference (8% of cases), recorded hospital characteristics (3% of cases), and patient-specific and socio-demographic factors (2% of cases).
ACt/RT was associated with an unadjusted five-year overall survival of 81.4 percent, compared with 68.7 percent for other treatment strategies.
The effect of treatment on overall survival was confounded by unobserved factors, the authors noted, so they estimated the effect of treatment according to on-site availability of RT or distance to facility. Again, there was no significant difference based on these factors: adjusted hazard ratios were 1.10 for on-site availability of RT (95% CI 0.57-2.31) and 1.00 for a greater than 30 miles distance to facility (95% CI 0.52-2.16).
There were limitations for these analysis, however, including a lack of information about tumor size, presence of bulky tumors, and specific information about chemotherapy regimens.
“Treatment choice is largely determined by clinical circumstances, but its association with socioeconomic factors suggests disparate access to care,” Dr. Fallah and authors concluded. “For patients whose treatment is selected on the basis of local RT availability, overall survival does not significantly differ whether they undergo ACt/RT or a strategy involving longer chemotherapy.”
Fallah J, Thomas AG, Olszewski AJ. Abbreviated chemotherapy and radiation for early-stage diffuse large B-cell lymphoma (DLBCL) in the R-CHOP era: analysis of the National Cancer Data Base (NCDB). Abstract #3. Presented at the ASH Meeting on Hematologic Malignancies; September 18, 2015; Chicago, IL.