Radiation Oncology: A Question-Based Review

Non-Hodgkin Lymphoma

Owen C. Thomas and John P. Plastaras

image Background

What is the pathologic definition of non-Hodgkin lymphoma (NHL)?

NHL is a monoclonal expansion of malignant B or T cells that lacks the pathologic characteristics of Hodgkin disease (HD) (no RS cells) and is typically characterized by nodal/focal involvement vs. the more disseminated presentation of leukemias.

How does the clinical presentation of NHL differ from that of HD?

NHL typically involves more nodes at presentation, is more likely to be extranodal, is more likely to spread in a noncontiguous fashion, and has a prognosis that is more strongly affected by histologic subtype than HD.

What are the most common presenting signs or Sx of NHL?

Painless adenopathy (axillary, inguinal, femoral) is the most common presenting sign of NHL. ~30% of pts have B Sx. Waxing and waning adenopathy suggests an indolent form of NHL. Tumor bulk may cause intestinal obstruction, urinary tract obstruction, or nerve compression.

What are B Sx?

B Sx include fever >38°C, >10% body weight loss in 6 mos, or drenching night sweats.

What is the NCI working formulation for NHL?

The working formulation groups NHL by clinical aggressiveness or grade with subgroups based on cell type or presentation.

1.     Low-grade NHL: follicular (grades 1–2), chronic lymphocytic leukemia (CLL), MALT, mycosis fungoides

2.     Intermediate-grade NHL: follicular (grade 3), mantle cell, diffuse large B-cell lymphoma (DLBCL), T/natural killer (NK) cell, peripheral T cell, anaplastic large cell

3.     High-grade NHL: Burkitt, lymphoblastic

What is the WHO classification of NHL?

The WHO classification divides NHL into B- and T-cell/NK cell neoplasms. The indolentaggressive, and highly aggressive subgroups roughly correlate to the aforementioned working formulation groups.

Is there a relationship between clinical aggressiveness and curability of NHL?

Advanced-stage indolent NHL is rarely curable. Intermediate-grade NHL may be curable even in advanced stages.

Without Tx, what is the life expectancy for pts with NHL of varying aggressiveness?

Pts with indolent NHL have survival measured in yrs. Pts with aggressive NHL have survival measured in mos, and those pts with highly aggressive Dz have an expected survival of wks.

What % of NHL is indolent, and what are the most prevalent subtypes?

~35% of NHL is indolent by the WHO classification. 95% of indolent NHL are follicular lymphoma (FL) (grades 1–2; 65%), small lymphocytic lymphoma (SLL) (18%), and marginal zone B-cell lymphoma or MALT lymphoma (12%).

What are common cytogenetic abnormalities associated with indolent NHL?

t(14;18) is seen in 90% of FLs. This results in overexpression of antiapoptotic Bcl-2. Chromosome 13 deletiont(14;19), and trisomy 12 are associated with SLL and CLL. Trisomy 3 (60%) and t(11:18) (25%–40%) are associated with MALT lymphoma.

How is FL graded?

FL demonstrates a mix of centrocytes (small, cleaved cells) and centroblasts (large, noncleaved cells). Grade correlates to the density of centroblasts (e.g., 0–5 centroblasts/high-power field (hpf), grade 1; >15 −centroblasts/hpf, grade 3a).

What is SLL?

SLL is the same Dz entity as CLL but with a predominant manifestation in the spleen, liver, or nodes as opposed to peripheral blood or BM.

What is Richter syndrome? What is its rate of occurrence?

Richter syndrome is the transformation of SLL or CLL into DLBCL. It occurs in roughly 5% of cases.

How is bulky mediastinal Dz commonly defined?

Bulky mediastinal Dz is commonly defined as a mass greater than one third of the intrathoracic diameter at T5-6 on upright PA film.

How is bulky Dz defined outside of the mediastinum?

Outside of the mediastinum, bulky Dz is variably defined in clinical trials but most often is either any mass >5 cm or any mass >10 cm.

image Workup/Staging

What are the pertinent focused aspects of the physical exam in a person with suspected NHL?

The physical exam should include complete nodal assessment including epitrochlear and popliteal groups. Cervical adenopathy palpable above the hyoid bone should prompt an ENT exam. (The Waldeyer ring is more frequently involved in NHL than in HD.) Exam of extranodal at-risk sites including the liver, spleen, testicles, bones, abdomen, and flanks is appropriate.

What lab studies should be performed?

Laboratory studies should include: CBC with differential, CMP, LDH, β2-microglobulin, serum protein electrophoresis, HIV, hepatitis B virus (essential as it may reactivate with rituximab Tx), and hepatitis C virus. BM Bx should be performed for all lymphomas. LP should be performed for CNS Sx, testicular or paranasal sinus involvement, or immunodeficiency.

What imaging studies should be performed?

The imaging workup should include CT C/A/P. PET is appropriate in most cases. MRI brain should be performed for CNS Sx, testicular or paranasal sinus involvement, or immunodeficiency.

How is NHL staged?

NHL is staged similar to HD using the Ann Arbor (AA) system:

1.     Stage 1: involvement of 1 LN region or localized involvement of 1 extralymphatic organ or site (IE)

2.     Stage 2: involvement of ≥2 LN regions on same side of diaphragm or localized involvement of 1 associated extralymphatic organ or site and its regional LN, with or without involvement of other LN regions on same side of diaphragm (IIE)

3.     Stage 3: involvement of LN regions on both sides of diaphragm, which may also be accompanied by localized involvement of an associated extralymphatic organ or site (IIIE)

4.     Stage 4: multifocal involvement of ≥1 extralymphatic organ, with or without associated LN involvement, or isolated extralymphatic organ involvement with distant nodal involvement

Note: Pts without B Sx are designated with an A, otherwise with a B. Pts with splenic involvement are designated with an S.

What is a major limitation to the AA staging of NHL (as opposed to HD)?

NHL typically spreads in a less contiguous fashion compared to HD and thus stage I NHL is very rare (10%).

How is NHL practically staged?

NHL is practically divided into limited stage and advanced stage. Limited stage consists of AA stage I–II pts with ≤3 adjacent LN regions, no B Sx, and no bulky (≥10 cm) lesions. Advanced stage includes all other pts. For FL, practical division is between AA stage I–II, nonbulky, nonabdominal Dz, and all others.

image Treatment/Prognosis

What are prognostic factors for NHL?

While grade remains the most important factor, several attempts have been made to combine multiple prognostic factors into a single numerical prognostic index to determine prognosis within a grade stratification of NHL. The most well known is the International Prognostic Index (IPI) based on aggressive NHL. Derivatives of the IPI include age adjusted, stage adjusted, and the follicular lymphoma International Prognostic Index (FLIPI).

What factors are included in the IPI, and what are the expected stratified survivals?

IPI factors: age >60 yrs, ECOG performance status ≥2, LDH > normal, >1 extranodal group, AA stages III–IV (Mnemonic: APLES)



5-YR OS (%)

Low risk



Low-intermediate risk



High-intermediate risk



High risk



What were the Dz characteristics and the Tx strategies employed for the British Columbia pts whose outcome data was used to generate the IPI formulation?

The data used to generate the IPI come from 308 pts with stage I–IIa nonbulky DLBCL treated at the British Columbia Cancer Agency from 1980–1998. All pts rcv 3 cycles of doxorubicin-containing chemo and involved-field radiation therapy (IFRT) (30 Gy for small-volume Dz, 35 Gy for larger-volume Dz). (Shenkier TN et al., JCO 2002)

What is the revised IPI (R-IPI) for intermediate-risk NHL incorporating the use of Rituxan?

The R-IPI incorporates the same 5 factors as the standard IPI but with substantial changes in the prognosis of these pts:



5-YR OS (%)

Very good









(Sehn LH et al., Blood 2007)

What factors are included in the FLIPI, and what are the expected stratified survivals?

FLIPI factors: Hgb <12 g/dL, age >60 yrs, stages III–IV, ≥5 extranodal sites, LDH > normal (Mnemonic: FLIPI is a HASSL). Note: These are FLIPI-specific nodal sites, not AA nodal groups.



10-YR OS (%)

Low risk



Intermediate risk



High risk



What was demonstrated by the Stanford retrospective series supporting RT alone in the management of stage I–II, low-grade FL?

The Stanford series of 177 pts treated from 1961–1994 (MacManus MP et al., JCO 1996) demonstrated an MS of 13.8 yrs and a 5-yr actuarial survival of 82%. RT included IFRT, EFRT, and TLI. Doses ranged from 35–50 Gy. Age <60 yrs was associated with better OS and freedom from recurrence (FFR). Only 5/47 pts who had FFR at 10 yrs relapsed subsequently. RT to both sides of the diaphragm was associated with better 10-yr FFR (67% vs. 36%).

What was demonstrated by the MDACC retrospective series supporting RT alone in the management of stage I–II, low-grade FL?

The MDACC series of 80 pts, treated from 1960–1988 (Wilder RB et al., IJROBP 2001) with similar Tx as in the Stanford series, demonstrated a 15-yr OS of 43% (vs. 44% for Stanford), no failures after 17 yrs, and better FFR for stage I vs. stage II pts (87% vs. 54% FFR at 15 yrs with no difference in OS). No difference in outcomes was appreciated for regional vs. extended-field RT.

What was demonstrated in the retrospective Stanford series of stage I–IIA, low-grade FL not treated immediately?

In this series, 43 pts (11 pts stage I) with a median age of 58 yrs rcv no initial Tx for various reasons. At a median follow-up of 86 mos, 63% had not been treated. Estimated survivals at 5, 10, and 20 yrs were 97%, 85%, and 22%, respectively. (Advani R et al., JCO 2004)

Has adj chemo demonstrated a benefit in randomized trials of early-stage, low-grade FL?

No. 5 randomized trials have failed to demonstrate significant benefit from RT → adj chemo (cyclophosphamide/vincristine/prednisone [CVP], chlorambucil, or cyclophosphamide HCl/doxorubicin/Oncovin/prednisone [CHOP] in early-stage FL:

1.     Nissen NI et al., Cancer 1983

2.     Monfardini S et al., IJROBP 1980

3.     Carde P et al., Radiother Oncol 1984

4.     Yahalom J et al., Cancer 1993

5.     Kelsey SM et al., Med Oncol 1994

What is the clinical evidence for combined modality Tx in the management of early-stage, low-grade FL?

A prospective series from MDACC followed 102 pts (83% FL) with low-grade NHL treated with 10 cycles of chemo and IFRT to 30–40 Gy after 3 cycles. 10-yr FFP and OS for pts with FL were 72% and 80%, respectively (Seymour JF et al., JCO 2003). Other retrospective series have also demonstrated good outcomes.

What remains the Tx standard for localized, low-grade FL?

Per NCCN (2010) guidelines, locoregional RT to 30–36 Gy remains standard. However, observation and combined modality Tx are considered viable options depending on pt and Dz characteristics.

What are the basic Tx principles for stage III–IV, low-grade FL?

No Tx is considered curative. Several randomized trials have indicated that therapy can be deferred without reducing survival. Tx is reserved for the following:

1.     Symptomatic Dz

2.     Threatened end organ dysfunction

3.     Cytopenias

4.     Bulky Dz

5.     Steady Dz progression

6.     Clinical trial

7.     Pt preference

What is the role of RT for stage III–IV, low-grade FL?

In advanced-stage indolent lymphomas, RT is reserved for palliation of Sx.

What is a typical RT Rx for symptomatic stage III–IV FL?

Traditionally, >20 Gy. A phase II study of RT for symptomatic local masses with indolent NHL demonstrated a 65% CR and 22% PR with a median duration of response of 22 mos with 2 Gy × 2 (Johannsson J et al., IJROBP 2002).

What is the role of RT in the Tx of SLL?

RT is used for palliation of symptomatic lesions in SLL. Consider 2 Gy × 2 regimens.

What is the role of RT in treating nodal marginal zone lymphomas?

RT is used for palliation of symptomatic lesions in advanced-stage nodal marginal zone lymphomas.

What is the most common initial multiagent chemo used in the management of intermediate-/high-grade NHL?

The most common initial multiagent chemo used in NHL is R-CHOP, which uses the following drugs:

1.     Rituximab

2.     Cyclophosphamide

3.     Hydroxydaunomycin (Adriamycin)

4.     Oncovin (vincristine)

5.     Prednisone

What are the current indications for RT regarding early-stage, intermediate-/high-grade NHL?

The inclusion of RT in early-stage, intermediate-/high-grade NHL is very institution dependent. It may be included as consolidation after 3–4 cycles of R-CHOP in favorable Dz, in pts with a PR to chemo, or in pts with bulky Dz. The inclusion of rituximab to CHOP in advanced Dz has resulted in significant OS benefit, and the results of trials including rituximab in the management of localized Dz may obviate the inclusion of RT for most localized NHL. Alternatively, improved systemic control may further increase the importance of LC and thus RT in early-stage Dz.

What is the present Tx paradigm for advanced stage, intermediate-/high-grade NHL?

Advanced-stage, intermediate-/high-grade NHL Tx paradigm: R-CHOP × 6–8 cycles. IFRT may be considered for initially bulky sites.

Estimate the prognosis of limited-stage aggressive B-cell lymphoma treated with R-CHOP and IFRT.

Long-term outcomes with R-CHOP and IFRT are limited. SWOG 0014 enrolled 60 pts with limited-stage aggressive NHL and at least 1 adverse risk factor and treated with R-CHOP × 3 + IFRT: 4-yr PFS was 88%, and OS was 92%. (Persky DO et al., JCO 2008)

What is the long-term DFS for pts with localized DLBCL treated with RT alone? What were the typical Tx doses used in clinical trials?

Using 45–50 Gy to maximize LC, only 40% of pts with localized DLBCL had long-term DFS based on historical RT-alone data. (Chen MG et al., Cancer 1979; Kaminski MS et al., Ann Intern Med 1986; Sweet DL et al., Blood 1981)

What was demonstrated in the initial publication of the SWOG 8736 study comparing chemo alone to abbreviated CRT in localized intermediate-grade NHL?

In SWOG 8736, 401 pts with stage I or IE (including bulky Dz) and stage II or IIE (nonbulky) intermediate-grade NHL were randomized to CHOP × 8 cycles vs. CHOP × 3 + IFRT. RT doses of 40–55 Gy were employed. At 5-yr follow-up, PFS and OS favored the combined therapy group (OS: 82% vs. 72%). (Miller TP et al., NEJM 1998)

What was shown in the updated analysis of SWOG 8736 at median follow-up of 8.5 yrs?

In the updated analysis of SWOG 8736, published in abstract form in 2001, PFS curves overlapped at 7 yrs and OS curves overlapped at 9 yrs. There were excess late relapses and deaths from advanced lymphoma in the combined arm seen in yrs 5–10. Results are stratified by IPI criteria:

1.     Stage I, no risk factors: 94% OS at 5 yrs

2.     Stage II (nonbulky) and/or 1+ adverse risk factor: 71% OS at 5 yrs

3.     3 risk factors (stage II may be 1 of them): 50% OS at 5 yrs

(Miller TP et al., ASH abstract 3024 2001)

What was demonstrated in the ECOG E1484 study randomizing postchemo complete responders to observation vs. IFRT?

In ECOG E1484, 352 pts with intermediate-grade, bulky stage I–IE or nonbulky stage II–IIE Dz were administered CHOP × 8 cycles. Complete responders (215 pts) were randomized to IFRT vs. observation. At 6 yrs, DFS favored IFRT (73% vs. 56%), but OS was equivalent. FFS was equivalent in partial responders administered IFRT and in CR pts. Failure at initial sites was greater in pts not given IFRT. (Horning SJ et al., JCO 2004)

What was demonstrated in the GELA LNH-93-1 study comparing aggressive chemo vs. standard chemo and RT in pts ≤60 yo?

In GELA LNH-93-1, 647 pts ≤60 yo with low-risk (IPI 0), stage I or II, intermediate-risk NHL (extranodal or bulky Dz allowed) were randomized to doxorubicin/cyclophosphamide/vindesine/bleomycin/prednisone (ACVBP) × 3, then methotrexate/etoposide/ifosfamide/cytarabine vs. CHOP × 3, then IFRT to 30–40 Gy. ACVBP improved EFS and OS regardless of the presence of bulky Dz. (Reyes F et al., NEJM 2005)

What was demonstrated in the GELA LNH-93-4 study evaluating pts age >60 yrs with low-risk, localized, intermediate-grade NHL?

In GELA LNH-93-4, 576 pts age >60 yrs with low-risk (age-adjusted IPI 0), stage I or II NHL (bulky [8%] or extranodal [56%] Dz allowed) were randomized to CHOP × 4 vs. CHOP × 4 + IFRT to 40 Gy. The 5-yr EFS and OS were equivalent in both Tx arms. (Bonnet C et al., JCO 2007)

What is the present Tx paradigm for relapsed intermediate-/high-grade NHL?

Relapsed intermediate-/high-grade NHL Tx paradigm: high-dose chemo + stem cell transplant

image Toxicity

What are the expected RT toxicities associated with Tx of NHL?

The RT toxicities depend on the site of Tx. B/c of high rates of long-term survival in these pts, particularly important late effects are coronary artery Dz, hypothyroidism, and 2nd malignancies. The later age at presentation, when compared to HL, should be considered when considering risk of 2nd malignancies.