Hodgkin Lymphoma

Radiation Therapy
In adult HL, the appropriate dose of radiation alone is 25 Gy to 30 Gy to clinically uninvolved sites and 35 Gy to 44 Gy to regions of initial nodal involvement.
These recommendations are often modified in pediatric or advanced-staged adult patients who also receive chemotherapy.
Treatment is usually delivered to the neck, chest, and axilla (mantle field) and then to an abdominal field to treat para-aortic nodes and the spleen (splenic pedicle). In some patients, pelvic nodes are treated with a third field.
The three fields constitute total nodal radiation therapy. In some cases, the pelvic and para-aortic nodes are treated in a single field called an inverted Y. In patients with a favorable prognosis, treatment of the pelvic lymph nodes is frequently omitted, since fertility can be preserved without affecting relapse-free survival.

Second Malignancies Acute nonlymphocytic leukemia may occur in patients treated with combined modality therapy or with combination chemotherapy alone, especially with increasing exposure to alkylating agents.
At 10 years after therapy with regimens containing MOPP, the risk of acute myelogenous leukemia (AML) is approximately 3%, with the peak incidence occurring 5 to 9 years after therapy.
The risk of acute leukemia at 10 years following therapy with ABVD appears to be less than 1%.
A population-based study of more than 35,000 survivors during a 30-year time span identified 217 patients who developed AML; the excess absolute risk is significantly higher (9.9 vs. 4.2 after 1984, P < .001) for older patients (i.e., >35 years at diagnosis) versus younger survivors.

An increase in second solid tumors has also been observed, especially cancers of the lung, breast, thyroid, bone/soft tissue, stomach, esophagus, colon and rectum, uterine cervix, head and neck, and mesothelioma.
These tumors occur primarily after radiation therapy or with combined modality treatment, and approximately 75% occur within radiation ports.
At a 15-year follow-up, the risk of second solid tumors (cumulative incidence of a second cancer) is approximately 13%; at a 20-year follow-up, the risk is approximately 17%; at a 25-year follow-up, the risk is approximately 22%; and at a 40-year follow-up, the risk is approximately 48%.
In a cohort of 18,862 5-year survivors from 13 population-based registries, the younger patients had elevated risks for breast, colon, and rectal cancer for 10 to 25 years before the age when routine screening would be recommended in the general population. Even with involved-field doses of 15 Gy to 25 Gy, sarcomas, breast cancers, and thyroid cancers occurred with similar incidence in young patients receiving higher-dose radiation.

Lung cancer is seen with increased frequency, even after chemotherapy alone, and the risk of this cancer is increased with cigarette smoking.
In a retrospective Surveillance, Epidemiology, and End Results (SEER) analysis, stage-specific survival was decreased by 30% to 60% in HL survivors compared with patients with de novo non-small cell lung cancer.
Breast cancer is seen with increased frequency after radiation therapy or combined modality therapy. The risk appears greatest for women treated with radiation before age 30 years and especially close to menarche, and the incidence increases substantially after 15 years of follow-up.
In two case control studies of 479 patients who developed breast cancer after therapy for HL, cumulative absolute risks for developing breast cancer were calculated as a function of radiation therapy dose and the use of chemotherapy.
With a 30-year to 40-year follow-up, cumulative absolute risks of breast cancer with exposure to radiation range from 8.5% to 39.6%, depending on the age at diagnosis.
A family history of breast cancer or ovarian cancer does not confer a greater increased risk than that of radiation therapy for one of these cohorts. These cohort studies show a continued increase in cumulative excess risk of breast cancer beyond 20 years of follow-up.

In a nested case control study and subsequent cohort study, patients who received both chemotherapy and radiation therapy had a statistically significant lower risk of developing breast cancer than those treated with radiation therapy alone.
Reaching early menopause with less than 10 years of intact ovarian function appeared to account for the reduction in risk among patients who received combined modality therapy. Reduction of radiation volume also decreased the risk of breast cancer after HL.
The risk of non-HL is also increased, but this risk is not clearly related to type or extent of treatment.

Several studies suggest that splenic-field radiation therapy and splenectomy increase the risk of a treatment-related second cancer.
Late effects after autologous stem cell transplantation that is given for failure of induction chemotherapy include second malignancies, hypothyroidism, hypogonadism, herpes zoster, depression, and cardiac disease.

Adverse Effects of Therapy
A toxic effect that is primarily related to chemotherapy is infertility, usually after MOPP-containing or BEACOPP-containing regimens; After six to eight cycles of BEACOPP, most men had testosterone levels within normal range; however, among women younger than 30 years, 82% recovered menses (mostly within 12 months) but only 45% of women older than 30 years recovered menses. ABVD appears to spare long-term testicular and ovarian function.

Late complications primarily related to radiation therapy include hypothyroidism and cardiac disease, which may persist through to 25 years after first treatment.
The absolute excess risk of fatal cardiovascular disease ranges from 11.9 to 48.9 per 10,000 patient years and is mostly attributable to fatal myocardial infarction (MI).
The use of subcarinal blocking did not reduce the incidence of fatal MI in a retrospective review, perhaps because of the exposure of the proximal coronary arteries to radiation.
In a cohort of 7,033 HL patients, MI mortality risk persisted through to 25 years after first treatment with supradiaphragmatic radiation therapy (dependent on the details of treatment planning), doxorubicin, or vincristine.
HL patients treated with mediastinal radiation compared with a normal-matched population have been reported to be at increased risk with the use of cardiac procedures.

Impairment of pulmonary function may occur as a result of mantle-field radiation therapy; this impairment is not usually clinically evident, and recovery in pulmonary testing often occurs after 2 to 3 years.
Pulmonary toxic effects from bleomycin as used in ABVD are seen in older patients (especially those older than 40 years).
Avascular necrosis of bone has been observed in patients treated with chemotherapy and is most likely related to corticosteroid therapy.

Bacterial sepsis may occur rarely after splenectomy performed during staging laparotomy for HL; it is much more frequent in children than in adults.
The Advisory Committee on Immunization Practices recommends that all patients with HL, whether or not they have had a splenectomy, should be immunized with Haemophilus influenzae type b conjugate, meningococcal, and pneumococcal vaccines at least 1 week before treatment.
Some investigators recommend reimmunization with all three vaccines 2 years after completion of treatment and with pneumococcal vaccine every 6 years thereafter.

Fatigue is a commonly reported symptom of patients who have completed chemotherapy. In a case-control study design, a majority of HL survivors reported significant fatigue lasting for more than 6 months after therapy compared with age-matched controls.
Quality-of-life questionnaires given to 5,306 patients on German Hodgkin Study Group trials showed that 20% of patients complained of severe fatigue 5 years after therapy, and those patients had significantly increased problems with employment and financial stability.

Recommendations for screening for secondary malignancies or follow-up of long-term survivors are consensus based rather than derived from randomized trials.

Patients older than 60 years with HL may experience more treatment-related morbidity and mortality; maintaining dose intensity of standard chemotherapy may be difficult.
Alternative therapies have been proposed for elderly patients, but no randomized trials have been conducted with these regimens.
A series of 27 previously untreated patients older than 60 years, judged by the investigator to be in too poor a condition to undergo chemotherapy, received brentuximab. A 92% overall response rate and 73% complete remission rate were reported.[Level of evidence: 3iiiDiv]

Recommendations for posttreatment follow-up are not evidence based, but a variety of opinions have been published for high-risk patients who present with advanced-stage disease or for patients who achieve less-than-complete remission by positron emission tomography–computed tomography scans at the end of therapy.

Historically, radiation therapy alone had been the primary treatment for patients with early favorable HL, often after confirmatory negative staging laparotomy.
A randomized, prospective trial involving 542 patients with early favorable HL compared MOPP-ABV for three cycles plus involved-field radiation therapy (IF-XRT) with subtotal nodal radiation; with a median follow-up of 7.7 years, combined modality was favored in terms of 5-year event-free survival (98% vs. 74%, P < .001) and 10-year overall survival (OS) (97% vs. 92%, P = .001).[Level of evidence: 1iiA]
The late mortality from solid tumors, especially in the lung, breast, gastrointestinal tract, and connective tissue, and from cardiovascular disease makes radiation therapy a less attractive option for the best-risk patients, who have the highest probability of cure and long-term survival. Recent clinical trials have focused on regimens with chemotherapy and IF-XRT or with chemotherapy alone.

A randomized, prospective trial from the National Cancer Institute of Canada involving 123 patients with early favorable HL compared ABVD for four to six cycles with subtotal nodal radiation; with a median follow-up of 11.3 years, no difference was observed in event-free survival (89% vs. 86%; P = .64) or in OS (98% vs. 98%; P = 0.95).[Level of evidence: 1iiA]

In a randomized study from the Milan Cancer Institute of patients with clinical early-stage HL, 4 months of ABVD followed by either IF-XRT or extended-field radiation therapy (EF-XRT) showed similar OS and freedom-from-progression with a 10-year median follow-up, but the study had inadequate statistical power to determine noninferiority of IF-XRT versus EF-XRT.[Level of evidence: 1iiDii]

The German Hodgkin Lymphoma Study Group (GHSG) randomly assigned 1,190 patients with early favorable HL to the following:

• Two cycles of ABVD plus 30 Gy of IF-XRT.
• Two cycles of ABVD plus 20 Gy of IF-XRT.
• Four cycles of ABVD plus 30 Gy of IF-XRT.
• Four cycles of ABVD plus 20 Gy of IF-XRT.

With a 7.6-year median follow-up, no differences were observed in freedom-from-progression (97%) or OS (98%) for all four groups.[Level of evidence: 1iiA]

The GHSG study (HD 13) compared reduced chemotherapy schedules while maintaining IF-XRT at 30 Gy: two cycles of ABVD, two cycles of ABV, two cycles of AVD, or two cycles of AV among 1,502 patients.
After 5 years, freedom from treatment failure was significantly worse when dacarbazine, bleomycin, or both were omitted. This trial suggests that ABVD remains the standard therapy when given in four doses with radiation therapy for early-stage favorable disease.

A specialized approach to therapy can be taken when patients with nonbulky lymphocyte–predominant disease presenting in unilateral high neck (above the thyroid notch) or epitrochlear locations require only IF-XRT after clinical staging.
A retrospective report of 426 cases of lymphocyte-predominant HL (including the so-called nodular lymphocyte–predominant and lymphocyte-rich classical subtypes) showed that more patients died of treatment-related toxicity (both acute and long term) than from recurrence of HL.[Level of evidence: 3iiiA]
Limitation of radiation dose and radiation fields and avoidance of leukemogenic chemotherapeutic agents, along with watchful waiting policies, should be investigated for these subgroups.
Patients with nonbulky nodular sclerosing disease presenting in the anterior mediastinum only after clinical staging also do well with mantle radiation alone.

Treatment options include the following:

• ABVD for four to six cycles.
• ABVD for two cycles plus IF-XRT (20 Gy or 30 Gy).
• Radiation therapy alone in special circumstances.

Patients with early unfavorable HL showed relapse rates greater than 30% at 5 years with radiation therapy alone, prompting evaluation of chemotherapy plus involved-field radiation therapy (IF-XRT) versus chemotherapy alone.
The late mortality from solid tumors, especially in the lung, breast, gastrointestinal tract, and connective tissue, and from cardiovascular disease makes radiation therapy a less attractive option unless therapeutic benefits exceed the long-term complications.

A randomized, prospective trial from the National Cancer Institute of Canada (NCIC) involving 276 patients with early unfavorable HL compared ABVD for four to six cycles with ABVD for two cycles plus extended-field radiation therapy (EF-XRT);
with a median follow-up of 11.3 years, the freedom-from-progression favored combined modality therapy (86% vs. 94%; P = .006), but the overall survival (OS) was better for ABVD alone (92% vs. 81%; P = .04).[Level of evidence: 1iiA]
The trend toward a worse survival for the combined modality arm was attributed to excess secondary malignancies and cardiovascular deaths. In this trial, the extended-field radiation used higher doses and significantly larger exposure to body sites than are employed in current practice.

A randomized study from the Southwest Oncology Group of clinically staged patients (no laparotomy) compared subtotal lymphoid radiation with 3 months of AV followed by subtotal lymphoid radiation therapy; the combined modality arm showed superior failure-free survival (94% vs. 81%; P < .001) but not OS at 3.3 years' median follow-up.[Level of evidence: 1iiDiii]

In a randomized study from the Milan Cancer Institute of patients with clinical early-stage HL, 4 months of ABVD followed by either IF-XRT or EF-XRT showed similar OS and freedom-from-progression with 10 years' median follow-up, but the study had inadequate statistical power to determine noninferiority of IF-XRT versus EF-XRT.[Level of evidence: 1iiDii]
Similarly, in a randomized study from the German Hodgkin Lymphoma Study Group (GHSG) of more than 1,000 patients with early unfavorable HL, 4 months of COPP plus ABVD followed by IF-XRT versus EF-XRT showed equivalent OS and freedom-from-treatment failure (FFTF) with 5 years' median follow-up.[Level of evidence: 1iiA]
Another randomized study of 996 patients with early unfavorable HL also showed no difference in OS and event-free survival at 10 years, comparing four to six cycles of MOPP-ABV plus IF-XRT versus the same chemotherapy plus subtotal nodal radiation therapy.[Level of evidence: 1iiA]

In the HD11 trial, the GHSG randomly assigned 1,395 patients with early unfavorable HL to the following:
• Four cycles of ABVD plus 30 Gy of IF-XRT.
• Four cycles of ABVD plus 20 Gy of IF-XRT.
• Four cycles of BEACOPP plus 30 Gy of IF-XRT.
• Four cycles of BEACOPP plus 20 Gy of IF-XRT.

With a 6.8 year median follow-up no differences were observed in OS (93%–96%) for all four groups.[Level of evidence: 1iiA]
In the arms of the study with 30 Gy of IF-XRT, there was no difference in FFTF between BEACOPP and ABVD (P = .65), but a significant difference in favor of BEACOPP was seen for FFTF when 20 Gy of IF-XRT was used (P = .02).[Level of evidence: 1iiD]

In the HD14 trial, the GHSG randomly assigned 1,528 patients with early unfavorable HL to either four cycles of ABVD plus 30 Gy of IF-XRT or two cycles of escalated BEACOPP followed by two cycles of ABVD plus 30 Gy of IF-XRT. With a median follow-up of 43 months, no difference was observed in OS.[Level of evidence: 1iiA]

A prospective, randomized trial from the European Organization for Research and Treatment of Cancer and Groupe d'Etudes de Lymphomes de L'Adulte of 808 patients with early unfavorable HL compared the following:
• Four cycles of ABVD plus 30 Gy of IF-XRT.
• Six cycles of ABVD plus 30 Gy of IF-XRT.
• Four cycles of BEACOPP plus 30 Gy of IF-XRT.

With a 64-month median follow-up, in a preliminary report in abstract form, no differences were observed in event-free survival (89%–92%; P = .38) or OS (91%–96%; P = .98).[Level of evidence: 1iiA]

In summary, these randomized trials support the use of ABVD for four cycles with 20 Gy to 30 Gy IF-XRT.
Could the radiation therapy be omitted to minimize late morbidity and mortality from secondary solid tumors and from cardiovascular disease?
The NCIC study is the only trial to address this question in patients with early unfavorable HL; although four to six cycles of ABVD alone has improved OS compared with a combined modality approach, the use of EF-XRT in the combined modality arm is excessive by current standards, and late effects will be magnified with these larger fields. In addition, chemotherapy alone was 8% worse in freedom-from-progression compared with the combined modality approach.

How can we balance an improvement in freedom-from-progression using radiation therapy with chemotherapy against late morbidity and mortality from late effects?
Randomized studies with or without IF-XRT would be required, but no such studies are currently under way.
An indirect comparison for using ABVD alone is that the 94% OS seen for early unfavorable patients in the NCIC study at 11 years is equivalent to the survival seen in the GHSG's HD6 [NCT00002561], ™HD10 [NCT01399931], and HD11 trials using combined modality therapy at 11 years.
A Cochrane meta-analysis of 1,245 patients in five randomized, clinical trials suggested improved survival for combined modality therapy versus chemotherapy alone (HR, 0.40; 95% CI, 0.27–0.61). However, the NCIC study does demonstrate a 92% OS for ABVD alone at a median follow-up of 11.3 years. This would support the use of ABVD for patients with early unfavorable disease. Long-term follow-up, which would account for late toxicities and deaths from combined modality therapy, will not be forthcoming from these trials.

Patients with bulky disease (≥10 cm) or massive mediastinal involvement were excluded from most of the aforementioned trials. Based on historical comparisons to chemotherapy or radiation therapy alone, these patients currently receive combined modality therapy.[Level of evidence: 3iiiDiii]

Treatment options include the following:

• Four to six cycles of ABVD.
• Four cycles of ABVD plus IF-XRT (20 Gy–30 Gy).

Patients who experience a relapse after initial wide-field, high-dose radiation therapy have a good prognosis.
Combination chemotherapy results in 10-year disease-free survival (DFS) and overall survival (OS) rates of 57% to 81% and 57% to 89%, respectively.
For patients who experience a relapse after initial combination chemotherapy, prognosis is determined more by the duration of the first remission than by the specific induction or salvage combination chemotherapy regimen.
Patients whose initial remission after chemotherapy was longer than 1 year (late relapse) have long-term survival with salvage chemotherapy of 22% to 71%.
Patients whose initial remission after chemotherapy was shorter than 1 year (early relapse) do much worse and have long-term survival rates of 11% to 46%.

Because of CD30 expression on malignant Reed-Sternberg cells of Hodgkin Lymphoma (HL), but limited expression on normal cells, CD30 is a target for therapy.
Brentuximab vedotin is a chimeric antibody directed against CD30, which is linked to the microtubule-disrupting agent, monomethyl auristatin E.
For relapsing patients, response rates around 75% are seen with complete remissions around 30% to 50% and median progression-free survival (PFS) of 4 to 8 months.[Level of evidence: 3iiiDiv]
A series of 27 previously untreated patients older than 60 years, judged by the investigator to be in poor condition and unable to undergo chemotherapy, received brentuximab. A 92% overall response rate and 73% complete remission rate were reported.[Level of evidence: 3iiiDiv]
Successful treatment with brentuximab for relapsed patients has been reported with a response rate of 60%.[Level of evidence: 3iiiDiv]

Patients who relapse after initial combination chemotherapy can undergo reinduction with the same or another chemotherapy regimen followed by high-dose chemotherapy and autologous bone marrow or peripheral stem cell or allogeneic bone marrow rescue.
This therapy has resulted in a 3- to 4-year DFS rate of 27% to 48%. Patients who are responsive to reinduction chemotherapy may have a better prognosis.
In a retrospective review of 105 patients, those older than 60 years fared better with a combination of chemotherapy and salvage radiation therapy than with the use of intensified transplant consolidation.[Level of evidence: 3iiiDiv]

Two randomized trials have compared aggressive conventional chemotherapy versus high-dose chemotherapy with autologous hematopoietic stem cell transplantation for relapsed chemosensitive HL. Both trials show improvement in freedom from treatment failure at 3 years for the transplantation arm (75% vs. 45% and 55% vs. 34%, respectively); but no difference was observed in OS.[Level of evidence: 1iiDii]
A Cochrane meta-analysis also concluded that autologous stem cell transplantation after reinduction chemotherapy improves relapse-free survival by 20% to 30% over chemotherapy alone but without an OS benefit.[Level of evidence: 1iiDii]

In two retrospective reviews of patients who underwent autologous bone marrow transplantation (ABMT) for relapsed or refractory disease, a comparison was made of those who received involved-field radiation therapy (IF-XRT) for residual masses after high-dose therapy versus no further treatment.
Those who received IF-XRT had improved PFS. Normalization of fluorine F 18-fludeoxyglucose PET-CT scans after reinduction therapy predicted a much better outcome after stem cell transplantation, with an event-free survival rate of 80% versus 29% in one phase II trial.[Level of evidence: 3iiiDi]

For 329 patients at high risk of residual HL after stem cell transplant, the double-blind AETHERA trial [NCT01100502] evaluated brentuximab vedotin versus placebo; the median PFS of 42.9 months for the brentuximab group was better than the 24.1 months for the control group (hazard ratio, 0.57; 95% confidence interval [CI], 0.40–0.81; P = .0013).[Level of evidence: 1iDiii]

The use of human leukocyte antigen-matched sibling marrow (allogeneic transplantation) results in a lower relapse rate, but the benefit may be offset by increased toxic effects. Reduced-intensity conditioning for allogeneic stem cell transplantation is also under clinical evaluation.

The anti-PD1 monoclonal antibody nivolumab, one of the new immune checkpoint inhibitors, has shown an overall response rate of 65% to 87% and a complete response rate of 16% to 28%, with durations usually exceeding 1 year for heavily pretreated, relapsed patients.[Level of evidence: 3iiiDiv]

A phase II trial reported a response rate higher than 50% for bendamustine in relapsing patients.[Level of evidence: 3iiiDiv] For patients with recurrent disease after ABMT, weekly vinblastine therapy has provided palliation with minimal toxic effects.[Level of evidence: 3iiiDiv] For the small subgroup of patients with only limited nodal recurrence following initial chemotherapy, radiation therapy with or without additional chemotherapy may provide long-term survival for about 50% of these highly selected patients.

Patients who do not respond to induction chemotherapy (about 10%–20% of all presenting patients) have less than a 10% survival rate at 8 years. For these patients, high-dose chemotherapy and autologous bone marrow or peripheral stem cell or allogeneic bone marrow rescue are under clinical evaluation. These trials have resulted in a 3- to 5-year DFS rate of 17% to 48%.

Treatment Option Overview for HL During Pregnancy
HL that is diagnosed in the first trimester of pregnancy does not constitute an absolute indication for therapeutic abortion. Each patient must be looked at individually to take into account the stage and rapidity of growth of the lymphoma and the patient's wishes.
If the HL presents in early stage above the diaphragm and appears to be growing slowly, patients can be followed carefully with plans to induce delivery early and proceed with definitive therapy.
Alternatively, these patients can receive radiation therapy with proper shielding.
Investigators at MD Anderson Cancer Center reported no congenital abnormalities in 16 babies delivered after the mothers had received supradiaphragmatic radiation while shielding the uterus with five half-value layers of lead.
Because of theoretical risks that the fetus might develop future malignancies from even minimal scattered radiation doses outside the radiation field, radiation therapy should be postponed, if possible, until after delivery.

Chemotherapy that is administered in the first trimester has been associated with congenital abnormalities in as many as 33% of infants.
However, in one series, there were no adverse effects in 14 children of mothers who received a combination of mechlorethamine, vincristine, procarbazine, and prednisone (MOPP) or a combination of doxorubicin, bleomycin, vinblastine, and dacarbazine (ABVD) during gestation, five of whom began treatment during the first trimester.
Consequently, some women may opt to continue the pregnancy and agree to radiation therapy or chemotherapy if immediate treatment is required.

In the second half of pregnancy, most patients can be followed carefully and can postpone therapy until induction of delivery at 32 to 36 weeks.
If chemotherapy is mandatory before delivery, such as for patients with symptomatic advanced stage disease, vinblastine alone (given at 6 mg/m² IV q 2 wk until induction of delivery) may be considered because it has never been associated with fetal abnormalities in the second half of pregnancy.
Steroids are employed both for their antitumor effect and for hastening fetal pulmonary maturity.
As an alternative, a short course of radiation therapy can be used before delivery in cases of respiratory compromise caused by the rapidly enlarging mediastinal mass.
Combination chemotherapy with ABVD appears to be safe in the second half of pregnancy. If chemotherapy is required after the first trimester, many clinicians prefer the combination of drugs over single-agent drugs or radiation therapy.

A multicenter retrospective analysis of 40 patients described pregnancy termination in 3 patients, deferral of therapy to postpartum in 13 patients (median 30 wk gestation), and antenatal therapy applied to the remaining 24 patients (median 21 wk gestation, all done after the first trimester). With a median follow-up of 41 months, the 3-year progression-free survival (PFS) was 85%, and the overall survival (OS) was 97%, often using ABVD.[Level of evidence: 3iiiDiv]

A retrospective analysis from MD Anderson of 39 patients described pregnancy termination in 3 patients, deferral of therapy to the postpartum period in 12 patients, and antenatal therapy in 24 patients. Two women miscarried after receiving doxorubicin-based chemotherapy in the first trimester.
With a median follow-up of 68 months from diagnosis, the 5-year rate of PFS was 75%, and the OS rate was 82%. These rates did not differ between the antenatal and postpartum timing of therapy.[Level of evidence: 3iiiDiv]

In one study, the 20-year survival rate of pregnant women with HL did not differ from the 20-year survival rate of nonpregnant women who were matched for similar stage of disease, age at diagnosis, and calendric year of treatment.
The long-term effects on progeny after chemotherapy in utero are unknown, though present evidence tends to be reassuring. There is no evidence that a pregnancy after completion of therapy increases the relapse rate for patients in remission.