Autologous hematopoietic cell transplantation: An update for clinicians

Abstract

High-dose chemotherapy followed by transplantation of autologous hematopoietic progenitor cells has a proven track record of safety and efficacy in hematological malignancies and select solid tumors. The near-universal use of peripheral blood stem cells as source for autografts, routine growth factor support, and antimicrobial prophylaxis post transplantation has improved the safety of this procedure. However, the advent of highly active novel therapies in the last few years warrants reappraisal of the role of autologous transplantation in the therapeutic armamentarium of malignant disorder. This review summarizes the current role of autologous transplantation for hematological malignancies, discusses modern standards for patient selection, and highlights long-term care issues of transplant survivors from an internist's perspective. Role of tumor purging in autologous transplantation, novel transplant conditioning regimens, and post-transplant therapies to prevent disease relapse are reviewed.

Key words::

• Autologous transplantation
• lymphoma
• myeloma
• purging
• stem cell mobilization

Key messages

• Advances in supportive care have dramatically improved the safety of autologous transplantation in the modern era, with expected rates of treatment-related mortality below 2%–3%.

• Peripheral blood has replaced bone marrow as the preferred source of stem cells.

• Autologous transplantation is the standard of care and potentially curative option for patients with relapsed Hodgkin and aggressive B-cell lymphomas.

• Autologous transplant survivors require meticulous follow-up from primary care physicians to screen for long-term complications.

Introduction

Many decades ago researchers determined that bone marrow cells transplanted from one animal to another could restore blood production (1). These preclinical investigations were taken to into the clinical arena, and hematopoietic cell transplantation (HCT) evolved as a potentially curative treatment modality for patients with selected neoplastic and, rarely, certain non-neoplastic diseases. Many tumors show a steep dose-response to cytotoxic therapy, i.e. an increase in the dose of drug or radiation markedly increases the number of cancer cells killed. Cure or effective disease control depends on delivering intensive doses of chemotherapy or chemotherapy plus radiation therapy. This intensity often exceeds the tolerance of bone marrow. Hematopoietic cell function must therefore be restored after such therapy, or these patients will experience fatal bone marrow aplasia. This is accomplished by infusing hematopoietic cells intravenously afterward; cells then ‘home’ to marrow to reconstitute lymphohematopoietic function. Thus, it is essential that blood production be restored using either autologous (‘self’) cells or cells collected from another individual, i.e. allogeneic donor cells.

In chemotherapy-sensitive hematologic malignancies, autologous HCT can provide long-term disease control, while avoiding the immunologic complications and delayed immune reconstitution inherent to allogeneic HCT. The phases of autologous HCT include: collection of autologous hematopoietic progenitor cells (HPCs) and their cryopreservation; administration of high-dose therapy (HDT; also called conditioning or preparative regimen); thawing and intravenous administration of the autologous cells; and then vigorous supportive care including antibiotics, blood component transfusions, and other measures such as electrolyte replacement therapy while awaiting recovery of hematopoiesis.

Pre-transplant patient evaluation

Recommending HCT to a potential candidate is a complex decision process intricately dependent on several variables, including underlying diagnosis, prior therapies, patient's age, performance status, degree of impairment in the vital organ function, and socio-economic support structure. Careful patient selection is crucial to successful outcomes after autologous HCT (2).

Infectious disease evaluation

Careful evaluation of the patients is necessary to minimize the risk not only of new post-HCT infectious complications, but also of reactivation of past fungal and viral infections. Detailed questioning regarding sexual orientation, history of unprotected sexual intercourse, needle sharing, remote blood transfusions, travel history, etc., can help direct the infectious disease screening work-up. All HCT recipients should be tested for the presence of anti-cytomegalovirus (CMV) IgG antibodies. Testing for serum anti-herpes simplex virus IgG and anti-varicella zoster IgG is also routinely performed. All patients must be screened for HIV infection. Autologous HCT in HIV-positive patients with a CD4 count ≥ 100/μL and an undetectable or low viral load (< 10,000 copies/mL) is well tolerated and is not associated with unusually high morbidity or mortality (3).

Pre-transplant dental evaluation and medically necessary oral care can eliminate potential sites of infection and trauma (4).

Patients undergoing HCT should be screened for infection with hepatitis B virus (HBV) and hepatitis C virus (HCV). Infection with HBV or HCV is not an absolute contraindication for HCT. However, these patients are at increased risk of post-transplant viral reactivation, veno-occlusive disease (VOD), liver cirrhosis, and even fulminant hepatic failure (5).

Organ function evaluation

Karnofsky performance status (KPS) is a useful tool for assessing a patient's functional status. In patients undergoing autologous HCT a KPS score of ≥ 70 is desirable. Some (6), but not all studies have suggested an association between decreased pre-transplantation left ventricular ejection fraction (LVEF) and risk of developing post-HCT cardiotoxicity (7,8). Demonstration of LVEF of ≥ 45%–50% is arbitrarily considered a HCT eligibility requirement by most transplant centers.

Pulmonary function testing (PFT) performed before HCT helps to identify patients at increased risk of developing post-transplant pulmonary events. Detection of abnormal diffusion capacity for carbon monoxide (DLCO) and alveolar-arterial oxygen gradient on PFT are independent predictors of the need for mechanical ventilation and mortality post HSCT (9). A serum creatinine value of ≤ 1.5 mg/dL (SI unit 132 μmol/L) and a creatinine clearance above 60 mL/min (SI unit 1 mL/s) before transplantation are desirable.

Assessment of hepatic function with the measurement of serum bilirubin concentration, transaminases, and albumin level is routinely performed during pre-transplant evaluation. Elevation of serum transaminases and alkaline phosphatase before HCT are risk factors for developing VOD following transplantation (10). Determination of serum ferritin, iron profile, hepatic magnetic resonance imaging, and/or liver biopsy can help identify patients with transfusion-related iron overload. Institution of appropriate chelation therapy can lead to improvement in hepatic function.

Tables I and II list common diagnostic tests and physiological criteria used by the majority of transplant centers to evaluate patient eligibility for HCT.

Table I. Recommended evaluation in a HCT candidate.

Routine work-up

1. History and physical examination

2. Confirmation of histological diagnosis and remission status (review of biopsies by an experienced hematopathologist and radiological assessments by experienced radiologist)

3. Karnofsky performance score

4. ABO typing

5. Complete blood count

6. Serum tests of liver and kidney function

7. Left ventricular ejection fraction (two-dimensional echocardiography or MUGA scan)

8. EKG

9. Chest X-ray

10. Pulmonary function testing including FEV1, DLCO, arterial blood gas

11. Pregnancy test (female patients)

12. Lumbar puncture (high-risk patients^a and those with history of central nervous system involvement)

13. Dental evaluation

14. Cytomegalovirus serologies

15. Herpes simplex viral serologies

16. Varicella zoster virus serologies

17. HIV testing

18. Hepatitis B and C serologies

^aLymphoma presenting in sinuses, breast, testes; large cell lymphoma in bone marrow. Blastoid MCL patients. HIV+ patients. Patients with signs and symptoms suspicious of CNS disease.

Table II. Ideal physiological criteria for selecting patients for HCT.

Suggested selection criteria
Upper age limit (years)	70–75
Karnofsky performance status	≥ 60–70
Left ventricular ejection fraction (%)	≥ 45
Cardiac rhythm	No uncontrolled tachy- or brady-arrhythmia
Pulmonary function:
FEV1	≥ 50%
DLCO (adjusted)	≥ 50%
Hepatic function:
Serum bilirubin	≤ 2 mg/dL or ≤ 32 μmol/L^a
ALT/AST	≤ 2 × normal
Second active malignancy	Must be absent
Pregnancy test (females)	Negative
Serum creatinine	≤ 1.5 mg/dL or ≤ 132 μmol/L
Active uncontrolled infections	Must be absent

ALT/AST = alanine aminotransferase/aspartate aminotransferase; DLCO = diffusion capacity; FEV1 = force expiratory volume in 1 second.

^aUnless due to Gilbert's syndrome.

Counseling

Assessment of patients’ psychological health before transplantation cannot be overemphasized. It is imperative to recognize and remedy factors impairing patients’ social and psychosocial well-being, before committing them to a life-changing event such as HCT. Identifying and addressing alcohol or substance abuse issues before HCT is the key for good post-HCT patient compliance. Undergoing transplantation generally means being off work for a number of months post HCT. Support from a transplant social worker to ensure insurance coverage of transplantation, arranging for temporary disability for those in need, providing assistance with child-care, transportation, lodging, etc., are small details which produce huge impact on patients’ mental well-being and post-transplant follow-up compliance.

Advice regarding smoking cessation is often required. Smoking increases the pulmonary transplant-related mortality risk 5-fold (11), underscoring the importance of smoking cessation counseling during pre-transplant evaluation (12).

Hematopoietic progenitor cell collection

Autologous HPCs are collected from patients’ peripheral blood or bone marrow. These HPCs are capable of giving rise to the entire lymphohematopoietic system. These early HPCs are identified as CD34 + by immunophenotyping. Bone marrow harvesting was the traditional method of collecting HPCs for HCT. Marrow is removed from the posterior superior iliac crests with the patient in the prone position under spinal or general anesthesia. In a typical harvest, 1000 mL of marrow is aspirated (10–15 mL/kg patient weight), which requires manually entering the bone marrow space about 100 times.

Harvesting HPCs from peripheral blood is clearly an easier way to collect stem cells than traditional bone marrow harvests in the operating room. Today, nearly all autologous HCTs use peripheral blood progenitor cells as the graft source. HPCs are collected from the peripheral blood using an apheresis instrument that removes the cells by a centrifugation density gradient and returns the unwanted blood cells and plasma back to the patient. This process usually takes about ≥ 3–4 hours, and about 12–18 liters of blood are processed in the procedure. The minimum HPC dose to perform an autologous HCT successfully is generally considered to be 2 million CD34 + cells/kg recipient body weight. A dose of ≥ 4–6 million CD34 + cells/kg is considered optimal (13,14).

HPC mobilization is performed by using cytokines, most commonly granulocyte-colony stimulating factor (G-CSF), either alone or in combination with chemotherapy or plerixafor (Figure 1) (14–16). Plerixafor is a small molecule that inhibits chemokine stromal cell derived factor 1-alpha from binding to CXC chemokine receptor 4, resulting in increased HPC migration into peripheral blood. The HPC mobilization method with the best risk, benefit, and cost ratio is controversial. While many lymphoma patients can successfully mobilize HPCs with ‘cytokine only’ strategies, mobilization failure rates remain a concern in patients with risk factors such as prior radiotherapy, heavily pretreated disease with more intensive chemotherapy regimens such as hyper-CVAD (cyclophosphamide, vincristine, doxorubicin and dexamethasone) or fludarabine-containing regimens, or radioimmunotherapy (RIT), advanced age, and bone marrow involvement (17–20). In such cases, consideration should be given to chemotherapy- or plerixafor-based mobilization strategies (13). In myeloma patients who are potential autologous HCT candidates, it is important to avoid known factors that impair HPC mobilization, e.g. melphalan-based induction regimens, long-term use of lenalidomide, and radiation therapy to a significant volume of hematopoietic marrow areas (21,22).

Figure 1. Methods of HPC mobilization for autologous transplantation.

Purging contaminating tumor cells from the HPC product

Relapse after autologous HCT derives usually from proliferation of a chemotherapy-resistant clone of malignant cells surviving the HDT, or rarely from reinfusion of an autograft contaminated by tumor cells. Ex vivo purging (by monoclonal antibodies, CD34 + cell selection, etc.) (23,24) or ‘in vivo purging’ (e.g. rituximab therapy given to lymphoma patients during mobilization) (25,26) of autologous HPC grafts appear to reduce or eliminate the risk of tumor-cell reinfusion at HCT. Reduction of autograft contamination with tumor cells, however, may not lead to improved outcomes. The CUP trial, comparing salvage chemotherapy alone to chemotherapy followed by either immune-magnetically purged or unpurged HCT in relapsed follicular lymphoma (FL) (27), reported no significant difference in the outcomes of purged compared to unpurged autografts. The lack of clear benefit in randomized data and a possible increase in infectious complications with ex vivo purging (28,29) preclude routine use of this approach.

Preparative (conditioning) regimens

In autologous HCT, the preparative regimen, which consists of high-dose chemotherapy and/or total-body irradiation (TBI), is administered first in an attempt to eliminate the malignant disease. The exact type of autologous HCT conditioning regimen depends to a certain extent on the histological diagnosis of the patient.

Several conditioning regimens have been evaluated for autologous HCT in myeloma. A randomized phase III trial compared intravenous melphalan at 200 mg/m² (MEL200) to a combination of melphalan at 140 mg/m² (MEL140) and 8 Gy TBI (30). The MEL140-TBI regimen caused significantly more toxicity. Overall survival (OS) and progression-free survival (PFS) were similar. Another recent phase III study compared MEL200 with MEL140 + busulfan (Bu-Mel) (31). In this study Bu-Mel was associated with significantly lower complete remission (CR) rates, more frequent grade 3–4 toxicities, and similar OS or PFS compared to MEL200. MEL200 is considered the standard of care for conditioning before autologous HCT in myeloma in those who can tolerate the procedure (e.g. younger patients with acceptable renal function). Advanced renal impairment is frequent (20%–30%) in myeloma (32,33). Autologous HCT is feasible even in myeloma patients with advanced renal impairment (serum creatinine > 3 mg/dL [SI unit > 265 μmol/L] or receiving hemodialysis). In this setting a melphalan dose of 140 mg/m² is preferred (34,35). Unlike MEL200 there is no single, widely used and accepted conditioning regimen for lymphoid malignancies. Commonly used regimens include: CBV (cyclophosphamide, etoposide, carmustine), BEAM (carmustine, etoposide, cytarabine, melphalan), BEAC (carmustine, etoposide, cytarabine, cyclophosphamide), and TBI-containing regimens (36), with limited retrospective data suggesting less treatment-related mortality (TRM) and more favorable outcomes following non-TBI-containing conditioning regimens (37).

Radioimmunotherapy in autologous transplantation conditioning

RIT with monoclonal antibodies conjugated to a radionuclide is effective against B-cell non-Hodgkin lymphoma (NHL). Two main approaches have evolved for applying RIT as autologous HCT conditioning. One uses high-dose myeloablative RIT (with or without chemotherapy), and the other combines standard-dose RIT with HDT. A number of small phase II studies utilizing high-dose RIT with iodine-131 tositumomab (¹³¹I-T), in mostly chemosensitive relapsed B-cell NHLs, demonstrated 4-year PFS and OS of ∼ 40% and ∼ 65%, respectively, with acceptable toxicities (38–40). However, since ¹³¹I-T emits γ-radiation, its administration is complicated by requirements for prolonged patient isolation, special infusion equipment, caregiver/health care worker exposure precautions, and complex dosimetry facilities. Hence, high-dose RIT by and large remains confined to centers with available expertise. To circumvent logistical challenges associated with high-dose RIT, several studies combining standard-doses of RIT with HDT to intensify auto-HCT conditioning reported encouraging outcomes (41–43). However, the BMT-CTN 0401 trial which randomized chemosensitive, relapsed diffuse large B-cell lymphoma (DLBCL) to either yttrium-90 ibritumomab tiuxetan-BEAM or rituximab-BEAM conditioning did not show a benefit of RIT conditioning for disease control or survival (44). The published evidence currently does not support routine addition of standard-dose RIT to auto-HCT conditioning.

Blood component transfusion, supportive care, and infection prophylaxis

After the completion of conditioning, HPCs are infused in order to rescue autologous hematopoiesis. Infused HPCs require time to regenerate hematopoiesis, and during this time the patient requires intensive supportive care to prevent complications arising from myeloablation, prolonged cytopenias, and organ damage induced by HDT. Breakdown of the normal mucosal barriers in the gastrointestinal tract leading to oral mucositis, pain, diarrhea, and increased susceptibility to infectious complications is common. Prophylaxis with fluoroquinolones (generally ciprofloxacin or levofloxacin) to prevent Gram-negative infections and an antifungal agent (e.g. fluconazole) until neutrophil recovery is recommended. Autologous HCT recipients are at risk of herpes reactivation for several months post transplantation (especially those with prior bortezomib exposure), and antiviral prophylaxis (e.g. acyclovir or valacyclovir) for 12 months post transplantation is common practice. Sulfamethoxazole/trimethoprim (or equivalent) is instituted generally after robust neutrophil recovery for Pneumocystis jirovecii prophylaxis post autologous HCT for 6–12 months. It is, however, worth mentioning that the type and duration of infectious disease prophylactic regimens display significant variability across different transplant centers and countries. Readers are referred to the Infectious Diseases Society of American website for details regarding guidelines for preventing infectious complications in HCT recipients (45).

Hematopoietic recovery after autologous HCT takes approximately 2–3 weeks, and recombinant hematopoietic growth factor support early after HPC infusion (within 1–5 days) usually can reduce the duration of severe neutropenia by 2–4 days. Red blood cell and platelet transfusions are essential to prevent or treat complications and symptoms of prolonged cytopenia. Red blood cell transfusions are used to keep the hematocrit typically over 25%. Platelet transfusions are used to keep the platelet count above 10,000/μL to minimize bleeding; platelet transfusions may be required at a higher threshold in patients who demonstrate a hemorrhagic diathesis. All blood products must be irradiated to prevent inadvertent engraftment of ‘contaminating’ allogeneic lymphocytes from the transfused unit that may cause lethal transfusion-associated graft-versus-host disease (GVHD).

Acute toxicities of autologous transplantation

When the neutrophil count drops below 500/μL, there is a markedly increased susceptibility to bacterial and fungal infections. The disturbance of the mucosal barrier due to the conditioning regimen and use of intravascular access devices add to the patient's risk of infection. The longer the neutropenic period, the higher the infectious risk. Major infections encountered in autologous HCT are with Gram-negative and Gram-positive bacteria; Clostridia difficile toxin-associated diarrhea; herpes simplex virus reactivation; and fungi, particularly Candida spp. and Aspergillus spp. The use of prophylactic antibacterial, antiviral, and antifungal agents, recombinant hematopoietic growth factors, and blood HPCs has decreased the incidence of severe infections (46,47). Other common and reversible toxicities include nausea, vomiting, alopecia, poor appetite, rash, edema, and flushing.

Visceral organs may also be damaged by high-dose therapy. Rare but serious problems include hepatic VOD (also known as sinusoidal obstruction syndrome), idiopathic pneumonia syndrome, and diffuse alveolar hemorrhage (48,49). Hepatic VOD usually presents within the first two to three weeks of the HCT as tender hepatomegaly, jaundice, and fluid retention. Lung injury can present early or later after HCT and may take months to resolve. Compared to allogeneic HCT, treatment-related mortality is low, typically < 3% during the first 100 days after transplant; relapse remains the most common cause of death (Figure 2). Table III summarizes post-autologous HCT complications and their prophylaxis (if available) and management.

Figure 2. Causes of death after autologous HCT (US data) reported to CIBMTR during 2010–2012.

Table III. Complications following autologous transplantation.

Complications^a	Prophylaxis/management
Conditioning regimen-related toxicities
Nausea/vomiting	Antiemetics
Diarrhea	Antidiarrheals and fluids to prevent dehydration
Alopecia
Loss of appetite	Dexamethasone to stimulate appetite
Altered taste sensation	Zinc tablets or lemon candy
Mucositis	Pelifermin, anesthetic mouth wash, supersaturated calcium phosphate oral rinse, opioid analgesic, parenteral nutrition in severe case
Myelosuppression	G-CSF, blood component transfusions
Fevers	Evaluate for infections, antipyretics
Fatigue	Exercise
Infections (bacterial, viral, fungal, etc.)	Antibiotic prophylaxis/treatment as appropriate
Clostridium difficile diarrhea	Hand washing/metronidazole, oral vancomycin
Pulmonary toxicity (carmustine, busulfan, radiation)	Careful patient selection, smoking cessation/steroids for therapy
Cardiac toxicity	Careful patient selection
Renal toxicity	Patient selection, avoiding nephrotoxic agents
Hemorrhagic cystitis (cyclophosphamide)	Mesna for prophylaxis
Dermatitis (etoposide)
Cataracts	Topical corticosteroids or emollients
Parotitis (radiation)	Chewing gums, lemon drops
Infertility	Sperm or oocyte cryopreservation
Hypothyroidism	Avoid TBI
Osteoporosis	Avoid TBI
Stem cell infusion
Flushing	Interrupting and slowing infusion rate
Hypotension
Breath odor due to DMSO
Allergic reactions	Acetaminophen, histamine blockers, corticosteroids
Chest tightness/dyspnea
Miscellaneous
Transfusion-associated graft-versus-host disease	Irradiated blood products; only curative therapy for tMDS/AML is allogeneic HCT
Second malignancies (especially tMDS/AML)	Short-course corticosteroids; high-dose corticosteroids and platelet transfusions
Engraftment syndrome
Diffuse alveolar hemorrhage

^aLong-term complications are indicated as bold text.

Indications for autologous transplantation

The predominant indication for autologous HCT is the treatment of cancer, although it is sometimes performed for non-neoplastic diseases (Figure 3). This section focuses on the indications of autologous HCT in the setting of oncology (50).

Figure 3. Common indications for autologous HCT in the United States, as reported to CIBMTR during 2010–2012. AL amyloidosis=Light chain amyloidosis, ALL=acute lymphoblastic leukemia, AML=acute myeloid leukemia, CLL=chronic lymphocytic leukemia, HL=Hodgkin lymphoma, MM=myeloma, NHL=non-Hodgkin lymphoma.

Autologous HCT for plasma cell myeloma

In the United States, myeloma and other plasma cell disorders are the most common indications for autologous HCT (Figure 3). The majority of autologous HCTs for myeloma are performed as a planned procedure in newly diagnosed patients after a defined initial phase of induction therapy—‘early or upfront autologous HCT’. As myeloma is incurable in most patients, even those that do not receive an upfront autograft may undergo transplant at relapse (‘delayed autologous HCT’).

Early autologous HCT for myeloma

Randomized trials in the 1990s compared autologous HCT to conventional chemotherapy in newly diagnosed myeloma patients. The IFM (Intergroupe Francophone Myeloma) (51) and the Medical Research Council (52) led clinical trials demonstrating that patients randomized to receive upfront autologous HCT had significantly higher CR rates, event-free survival (EFS), and OS. Other randomized studies (Table IV), however, have not uniformly demonstrated a survival benefit (53–57). In general, these studies do indicate that autologous HCT improves CR rates and EFS in comparison to conventional chemotherapy. Survival benefit has not been demonstrated in studies that permitted or planned for delayed autologous HCT at relapse, nor in studies conducted in later time periods when novel anti-myeloma agents (lenalidomide, bortezomib, carfilzomib, and pomalidomide) were available as an option after relapse.

Table IV. Selected randomized trials of conventional chemotherapy compared to single autologous HCT as upfront therapy in multiple myeloma.

Study (year) (ref.)	Transplant arm^a	No.; age (y)	CR% CC vs. auto HCT	Comments
Attal (1996) (51)	MEL 140 + TBI 8 Gy	200; < 65	5 vs. 22^b	OS and EFS superior for HCT
Child (2003) (52)	MEL 200	401; < 64	8 vs. 44^b	OS and EFS superior for HCT
Blade, PETHEMA (2005) (53)	MEL 200 or MEL 140 + 12 Gy TBI	164; < 65	11 vs. 30^b	Only responders randomized. No OS and EFS benefit of HCT
Palumbo (2004) (139)	MEL 100 × 2 auto-HCT	194; 50–70	6 vs. 25^b	EFS and OS superior for autologous HCT
Barlogie (2006) (56)	MEL 140 + TBI 12 Gy	899; < 70	15 vs. 17	No OS and EFS benefit for HCT vs. conventional chemotherapy; 55% of patients relapsing after conventional chemotherapy received delayed HCT
Fermand (1998) (54)	BCNU + etoposide + cyclophosphamide + MEL and TBI	185; < 56	20 vs. 57	EFS superior for autologous HCT. No OS difference as patients in conventional chemotherapy randomized to HCT at relapse; QOL superior for HCT
Fermand (2005) (55)	MEL 200 or MEL 140 + busulfan	55–65	20 vs. 36	EFS and OS similar to conventional chemotherapy; QOL superior for early autologous HCT

CC = chemotherapy; EFS = event-free survival; MEL = melphalan; OS = overall survival; QOL = quality of life.

^aChemotherapy doses in mg/m². Definitions of response were based on non-uniform criteria.

^bStatistically significant difference (P < 0.05).

Tandem autologous HCT

Barlogie and co-workers pioneered the tandem autologous transplant approach in their Total Therapy (TT) program (58,59). The IFM94 was the first randomized trial (60) comparing single to tandem autologous HCT in previously untreated myeloma patients. The projected 7-year EFS and OS benefits were significantly better for the double autologous HCT arm. In an unplanned subgroup analysis, there was no benefit for the second autologous HCT for patients who were in a very good partial remission (VGPR) status or better after the first autograft. Another randomized study (Bologna 96) confirmed the observation that the second HCT does not benefit patients in VGPR or better after the first transplant (61). More importantly this study did not show an OS benefit for the tandem procedure despite superior CR rates and EFS compared to single HCT. Most recently, in a combined Netherlands/German (HOVON65/GMMG-HD4) study, patients receiving tandem autologous HCT in the German (GMMG) component of the study were shown to have superior OS (70% versus 55% at 5 years) over otherwise similarly treated patients receiving a single autologous HCT (62). Randomized studies show tandem autologous HCT produce higher CR rates and superior EFS overall, but not OS benefit (with the exception of IFM94), compared to single autologous HCT. Moreover, benefits of the second autograft appear to be limited to patients who are not in a VGPR after the first autologous HCT.

Delaying autologous HCT until relapse

The timing of autologous HCT (early versus delayed) warrants reappraisal. Mature data from randomized studies, conducted in pre-novel therapy era, suggest that survival is similar whether autologous HCT is performed early or in the delayed setting after relapse (54,56). Early autologous HCT was, however, associated with a longer treatment-free interval and improved quality of life (QOL) in the French MAG-95 trial (54). In the novel therapy era, Palumbo and colleagues randomized patients after lenalidomide/dexamethasone induction to tandem autologous HCT or chemotherapy maintenance. Although CR rates and OS were similar, early HCT reduced the risk of progression by 50%, and median PFS (41 versus 18 months) strongly favored early autologous HCT (63). Data from this trial caution us against abandoning the upfront autologous HCT strategy for myeloma.

Role of autologous transplantation in high-risk disease

Myeloma is a biologically diverse disease with clonal heterogeneity and genomic instability (64,65). The presence of t(4;14) or del(17p), or a high serum beta 2-microglobulin concentration predicts inferior OS (66). Autologous HCT previously has yielded suboptimal results in high-risk myeloma (67), and the benefit of autografting has been questioned in these patients. Recent studies of bortezomib in induction, consolidation, and maintenance phases of therapy programs incorporating tandem transplantation, as in the Arkansas TT-3 trials as well as the HOVON-65/ GMMG-HD4 study, support the role of autologous HCT in high-risk myeloma (62,68,69).

Autologous HCT for uncommon plasma cell dyscrasias

Immunoglobulin light chain amyloidosis (AL), light chain deposition disease, heavy chain deposition disease, and POEMS syndrome are characterized by the presence of clonal plasma cells and light chain or heavy chain production leading to tissue injury. HDT with melphalan has proven effective in many of these disorders, although randomized trials are generally lacking (70,71). Special challenges for HCT in these disorders include higher rates of TRM, engraftment syndrome, and severe fluid retention (72,73). In AL, a disorder where pre-transplant organ dysfunction is common, cardiac involvement, effusions, and significant fluid retention during G-CSF administration for mobilization have emerged as predictors of higher TRM (72). Careful patient selection, dose-adapted melphalan use (depending on patient age, renal function, and comorbidities), and center experience are critical. A randomized multicenter trial failed to show benefit for HCT over conventional chemotherapy in AL (72,74), while experience from specialized centers seems to show a significant benefit, by inducing CRs after HCT that lead to organ function recovery from AL over time (71). Referring AL patients to high-volume centers for HCT should be strongly considered.

Autologous HCT for diffuse large B-cell lymphoma

HCT for high-risk DLBCL in first remission

Studies using autologous HCT as consolidation after first-line therapies in aggressive (mostly DLBCL) NHL in the rituximab era have reported contradictory findings. While two reports showed no benefit (75,76), the intergroup US study suggested improved PFS and OS with autologous HCT consolidation in high-risk International Prognostic Index (IPI) DLBCL (77). Considering these discordant data and the high cure rates of DLBCL with modern therapies (78), upfront use of autologous HCT as consolidation therapy is not recommended.

HCT for DLBCL—in relapsed disease

The role of autologous HCT in relapsed DLBCL is well-defined. The PARMA trial (Table V) (79) established that DHAP (dexamethasone, high-dose ara-C, and cisplatin) salvage chemotherapy and HCT provided a significantly enhanced OS benefit in subjects with relapsed, chemotherapy-sensitive disease. Several registry-based (80–83) and prospective studies in the rituximab era (Table V) (84) have reproduced these results. Autologous HCT is thus standard-of-care for relapsed/chemotherapy-sensitive DLBCL.

Table V. Select studies addressing the role of autologous transplantation in lymphoid malignancies.

Study (year) (ref.)	n	Study design	TRM	EFS/PFS (y)	OS (y)	Comments
Follicular lymphoma in first remission
GLSG (2004) (93)	307	Phase III Auto vs. Chemo	< 2.5% in both arms	Auto = 64%^a Chemo = 33% (5 y)	Not reported	Significantly more sMDS/AML with Auto (3.5%^a vs. 0%; P = 0.02)
GELA (2006) (94)	402	Phase III Auto vs. Chemo	Not reported	Auto = 38% Chemo = 28% (7 y)	Auto = 76% Chemo = 71% (7 y)	Second malignancies similar in both groups
GOELAMS (2009) (95)	166	Phase III Auto vs. Chemo	Not reported	Auto = 64%^a Chemo = 39% (9 y)	Auto = 76% Chemo = 80% (9 y)	Auto associated with significantly more second malignancies
GITMO (2008) (96)	136	Phase III Auto vs. Chemo	Auto = 3 Chemo = 2 at day 100	61%^a vs. 28% (4 y)	81% vs. 80% (4 y)	4-year sMDS/AML was higher with Auto (6.6% vs. 1.7%)
Relapsed follicular lymphoma
CUP (2003) (27)	89	Phase III Chemo vs. Purged Auto vs. Unpurged Auto	Not reported	Chemo = 26% Unpurged = 58%^a Purged = 55%^a (2 y)	Chemo = 46% Unpurged = 71%^a Purged = 77%^a (4 y)	No benefit of purging seen; done in pre-rituximab era
CIBMTR (2004) (99)	728	Retrospective	Purged = 14% Unpurged = 8%	Purged = 39% Unpurged = 31% (5 y)	Purged = 62%^a Unpurged = 55% (5 y)	OS favored purged autografts
Mantle cell lymphoma in first remission
Dreyling (2005) (85)	122	Phase III Auto vs. Chemo	Not reported	Auto = 54%^a Chemo = 25% (3 y)	Auto = 83% Chemo = 77% (3 y)	Performed in pre-rituximab era; no survival benefit
Hermine (2012) (90)	455	R-CHOP vs. R-CHOP/R-DHAP induction	4% vs. 4%	49 months vs. 84 months^a (median)	82 months vs. not reached^a (median)	Addition of cytarabine in induction improved OS
Relapsed diffuse large B-cell lymphoma
Philip (1995) (79)	215	Phase III Auto vs. Chemo	Not reported	Auto = 46%^a Chemo = 12% (5 y)	Auto = 53%^a Chemo = 32% (5 y)	Performed in pre-rituximab era
CORAL (2010) (84)	206	Prospective RICE vs. R-DHAP salvage	Not reported	53% (3 y)	Not reported	Established continued feasibility of auto-HCT for relapsed DLBCL
Relapsed Hodgkin lymphoma
CIBMTR (2001) (138)	414	Retrospective	7% at day 100	46%–64% (3 y)	58%–75% (3 y)	Patients in second CR had better outcomes
CIBMTR (1999) (116)	122	Retrospective	12% at day 100	38% (3 y)	50% (3 y)	Study established that primary refractory patients can be salvaged by auto-HCT
T-cell NHL in first remission
Corradini (2006) (106)	62	Phase II	4.8%	30% (12 y)	34% (12 y)	ALK+ anaplastic T-cell patients and ones in CR had favorable outcomes
Reimer (2009) (109)	83	Phase II	3.6%	36% (3 y)	48% (3 y)	66% of enrolled patients received auto-HCT
NLG-T-01 (2012) (110)	166	Phase II	4%	44% (5 y)	51% (5 y)	ALK-negative anaplastic T-cell had the best outcomes

AML = acute myeloid leukemia; Auto = autologous transplant arm; Chemo = chemotherapy; CHOP = cyclophosphamide, doxorubicin, vincristine, and prednisone; CIBMTR = Center for International Blood and Marrow Transplant Research; DHAP = dexamethasone, cytarabine, cisplatin; EFS/PFS = event-free survival/progression-free survival; GELA = Groupe d’Etude des Lymphomes de l’Adulte; GITMO = Gruppo Italiano Trapianto di Midollo Osseo; GLSG = German Low Grade Lymphoma Study Group; GOELAMS = Groupe Ouest-Est des Leucémies et Autres Maladies du Sang; OS = overall survival; RICE = rituximab, ifosphamide, etoposide, carboplatin; sMDS = secondary myelodysplastic syndrome; TRM = treatment-related mortality.

^aStatistically significant value.

Autologous HCT for mantle cell lymphoma

The European Mantle Cell Lymphoma (MCL) Network trial randomized MCL patients after first-line chemotherapy to either autologous HCT consolidation (‘upfront HCT’) or interferon-alpha maintenance (Table V) (85) and demonstrated a superior PFS but no OS benefit for autologous HCT. Similar randomized trials in the rituximab-era are not available. Prospective trials examining upfront autologous HCT in MCL patients following rituximab-containing induction regimens have reported encouraging 5-year PFS and OS approaching 60% and 70%, respectively (Table V) (86–91). Despite the lack of randomized data, autologous HCT consolidation for MCL in first remission could be considered standard practice. A CIBMTR (Center for International Blood and Marrow Transplant Research) analysis reported a 5-year OS rate of 61% for early upfront HCT versus 44% after HCT performed later in the disease course for chemotherapy-sensitive relapse (92). In relapsed MCL patients with chemotherapy-sensitive disease who are not candidates for allogeneic HCT, consolidation with autologous HCT is reasonable; however, it is not recommended for therapy-refractory MCL.

Autologous HCT for follicular lymphoma

HCT for FL—first remission

The role of autologous HCT as consolidation after initial therapy for advanced stage FL patients in first remission was examined in three trials conducted in the pre-rituximab era (93–95) and one in the rituximab era (Table VI) (96). Autologous HCT provided a PFS benefit in three out of these four trials. OS, however, was not improved as this therapy was associated with a higher risk of second malignancies (95) including therapy-related myelodysplastic syndrome/acute myeloid leukemia (tMDS/AML) (93,96). The use of autologous HCT consolidation for FL in first remission is not recommended.

Table VI. Post-autologous HCT vaccination schedule.

Vaccine	Time post transplantation to initiate vaccination	Recommended number of doses
Inactivated influenza	4–6 months	1 (followed by yearly vaccination)
Pneumococcal conjugate (PCV)	3–6 months	3 (after the 3 PCV doses, a dose of 23-valent polysaccharide pneumococcal vaccine to broaden the immune response might be given)
Haemophilus influenzae conjugate	6–12 months	3
Tetanus, diphtheria, acellular pertussis (DTaP)	6–12 months	3
Meningococcal conjugate	6–12 months	1
Inactivated polio	6–12 months	3
Recombinant hepatitis B	6–12 months	3
Measles, mumps, rubella	24 months	1–2

Modified from: Tomblyn M, Chiller T, Einsele H, et al. Guidelines for Preventing Infectious Complications among Hematopoietic Cell Transplantation Recipients: A Global Perspective. Biol Blood Marrow Transplant. 2009;15: 1143–238.

HCT for FL—relapsed disease

Use of autologous HCT in relapsed FL, too, is controversial primarily because this modality is generally not curative. In the pre-rituximab era, the CUP study compared salvage chemotherapy alone versus chemotherapy followed by either unpurged or purged HCT for relapsed FL; overall, the investigators reported a PFS and OS benefit with autologous HCT (Table V) (27). In the modern era, the superiority of autologous HCT over modern salvage chemoimmunotherapies continues to be debated (97). Registry data from the European Blood and Marrow Transplant Group (EBMT) (98) and CIBMTR show no plateau in relapse rates of FL post autologous HCT (99) and a 5%–15% risk of second malignancies (Table V) (98,100). Autologous HCT for relapsed FL should be considered in the context of alternative treatment choices, patient age, remission status, comorbidities, and a small but definite risk of secondary cancers. Autologous HCT is best reserved for chemotherapy-sensitive, relapsed FL in those subjects who are not candidates for curative intent allogeneic transplantation.

Autologous HCT for transformed follicular lymphoma

An EBMT report of 50 patients receiving autologous HCT for chemotherapy-sensitive FL transforming to DLBCL described 5-year PFS and OS of 30% and 51% (101). A prospective Norwegian study also reported 5-year PFS and OS of 32% and 47% rates, respectively (102). In the rituximab-era, a cohort analysis from the University of British Columbia suggested a survival benefit for patients undergoing autologous HCT, compared with those receiving salvage chemoimmunotherapy alone (103,104). Autologous HCT is reasonable for transformed FL patients with non-bulky (no nodal areas ≥ 3 cm in size), chemotherapy-sensitive disease.

Autologous HCT for T-cell lymphomas

The outcomes of relapsed T-cell NHL are dismal with conventional chemotherapy alone (105), and autologous HCT has been explored as initial therapy or at the time of tumor progression. Phase II studies (with heterogeneous histological subtypes) examining upfront HCT in T-cell NHL (Table V) (106–110) suggest a 3–4 year PFS ranging from 30% to 50%. Although randomized data are not available in T-cell NHL, considering the suboptimal outcomes with conventional front-line therapies it is appropriate to consider autologous HCT in most chemotherapy-sensitive patients in first remission. Since ALK-positive anaplastic large cell lymphoma has an excellent prognosis with standard chemotherapies, autologous HCT in first CR is not recommended for this specific subgroup (111). Relatively encouraging outcomes for a highly select group of relapsed T-cell NHL patients with sensitive disease also have been shown with 3–5 years PFS and OS rates of approximately 15%–30% and 30%–45%, respectively (112–114). Autologous HCT is also reasonable in relapsed/chemotherapy-sensitive T-cell NHL patients not deemed candidates for allogeneic HCT.

Autologous HCT for Hodgkin lymphoma

Approximately 15%–20% of Hodgkin lymphoma (HL) patients relapse and are candidates for autologous HCT. The British National Lymphoma Investigation study randomized relapsed HL subjects to chemotherapy salvage alone or to consolidation with autologous HCT (115) and reported a PFS benefit in favor of the HDT arm (53% versus 10%; P = 0.02). Autologous HCT is recommended in chemotherapy-sensitive first relapse of HL because of the remarkable improvement in PFS, low TRM, and the benefit of avoiding future relapses. HL patients not achieving a CR after first-line chemotherapies (primary refractory disease) have a poor prognosis. Lazarus et al. (116) showed that autologous HCT could cure about one-third of primary refractory HL patients (Table V). Chemotherapy-unresponsive disease at autologous HCT is a poor prognostic factor, with 5-year PFS rates generally < 20% (117). Autologous HCT is standard-of-care for relapsed HL, including those with primary refractory disease.

Autologous HCT for primary central nervous system (CNS) lymphoma

Upfront application of autologous HCT in primary CNS lymphomas has been investigated in a number of small phase II studies (118,119). While this approach is not considered standard-of-care, it is often considered in patients not achieving a CR after first-line therapies. Unlike the upfront setting, in patients with relapsed or refractory disease long-term disease control with chemotherapy or radiotherapy alone is rare. In such patients HDT is associated with CR rates of ∼50%–60%, 2–3 year OS of 40%–45%, and acceptable rates of TRM ∼ 10% (120,121).

Autologous transplantation in the elderly patients

Historically autologous HCT was offered only to younger patients (≤ 60–65 years). However, the advent of non-TBI-based conditioning, use of peripheral blood grafts, growth factor support, antibiotic prophylaxis, systematic revaccination, and overall improved supportive care have dramatically improved TRM of autologous HCT. As such, these improvements have translated into a consistently greater utilization of HDT in patients above 60–65 years of age. Contemporary registry data from the CIBMTR and EBMT seem to validate this practice, with seemingly no increased TRM or compromised survival outcomes in patients > 65 years of age undergoing auto-HCT (122–124). Autologous HCT is feasible and safe in carefully selected elderly patients (70–75 y) with excellent organ function and performance status, and few or no comorbidities.

Role of maintenance therapies following autologous transplantation

Disease relapse following autologous HCT is unfortunately common, especially in patients with myeloma and low-grade NHL. Various consolidation and maintenance strategies have been evaluated to delay post-HCT progression/relapse in myeloma. In general, novel anti-myeloma agents have shown superior results as maintenance therapy. Lenalidomide maintenance was studied in the Cancer and Leukemia Group B (CALGB) 100104 and IFM 05–02 trials (125,126). Both studies demonstrated a superior PFS and EFS with lenalidomide maintenance, while CALGB 100104 also demonstrated an OS benefit in the lenalidomide arm. There was, however, a 2- to 3-fold increase in second primary malignancies in both studies associated with the lenalidomide maintenance. Pre-transplant bortezomib induction coupled with post-autologous HCT bortezomib maintenance in the HOVON-65/GMMG-HD4 trial was shown to have superior PFS and OS (62). Based on these data, the risks/benefits and health care costs of novel agent maintenance should be discussed with myeloma patients post autologous HCT.

Due to the known benefits of rituximab maintenance in indolent B-cell NHL (127), this strategy has been studied in the post-HCT setting (28,128). A randomized EBMT study assigned relapsed FL patients to either rituximab maintenance or observation alone following autologous HCT (129). Maintenance therapy was well tolerated and was associated with superior 10-year PFS (54% versus 37%; P = 0.01), but no OS benefit was noted. In patients with DLBCL, the CORAL study examined post-autologous HCT rituximab maintenance and reported no PFS or OS. An increased risk of adverse events with maintenance was seen in this study (130). Routine use of rituximab maintenance after HCT in B-cell NHL is not recommended.

Follow-up after autologous HCT

Patients undergoing HDT in the outpatient setting generally require daily outpatient visits for hematopoietic growth factor support, blood count monitoring, and electrolyte replacements. Following neutrophil and platelet engraftment, most transplant centers reassess response to HDT approximately 2–3 month post transplantation, with history, physical examination, laboratory evaluations, radiographic/nuclear medicine imaging, and bone marrow biopsies as appropriate. Patients with no evidence of disease and resolved acute toxicities from this point onward are generally followed long-term. Once a year follow-up is recommended in long-term survivors. In (secretory) myeloma patients, monoclonal protein assessment in serum and urine, every 3–6 months, is common practice. Many programs perform surveillance bone marrow biopsies (in myeloma and acute leukemia patients) and PET and/or CAT-scans (in lymphoma patients) to screen for relapsed disease. No data, however, exist to suggest any benefit of surveillance with marrow and radiograph evaluations in otherwise asymptomatic patients.

Post-autologous HCT immunizations

HCT results in profound immunosuppression, and a decline in protective antibody titers to vaccine-preventable diseases (e.g. measles, mumps, rubella) is observed following autologous HCT, if the recipient is not revaccinated (http://www.cdc.gov/vaccines/pubs/hemato-cell-transplts.htm). All HDT recipients should be revaccinated routinely after transplantation. Inactivated influenza vaccine should be administered beginning at least 6 months after autologous HCT and annually thereafter. Sequential administration of three doses of pneumococcal conjugate vaccine is recommended, beginning 3–6 months after the transplant, followed by a dose of pneumococcal polysaccharide vaccine. A three-dose regimen of Haemophilus influenzae Type b (Hib) vaccine is administered beginning 6 months after transplant; at least 1 month should separate the doses. Measles, mumps, and rubella (MMR) vaccine should be administered 24 months after transplant, if the HCT recipient is immunocompetent. Table VI summarizes a suggested post-autologous HCT vaccination schedule. Centers for Disease Control (CDC) recommend caution regarding live varicella vaccine (since it is a live attenuated vaccine) (131) among HCT recipients. After considering the risk/benefits, if a decision is made to vaccinate with varicella vaccine, CDC recommends that the vaccine should be administered a minimum of 24 months after HCT and only if the recipient is presumed to be immunocompetent.

Late complications and survivorship issues after autologous HCT

Autologous HCT recipients require long-term follow-up and have a reduced life expectancy compared to the general population. These patients are at increased risk for late complications such as opportunistic infections, iron overload, endocrine abnormalities, osteoporosis, infertility, psychiatric disturbances, difficulty in returning to the work-force, and second malignancies including tMDS/AML (49,132–135). Recipients should be monitored for these potential maladies at least yearly, including administering all age-appropriate cancer screening and anti-infective vaccinations. Formal guidelines derived from prospectively observed patient cohorts to define optimal screening procedures in autologous HCT survivors are not available. American Society of Blood and Marrow Transplantation consensus guidelines (based primarily on expert opinion) recommended long-term autograft survivors to undergo yearly ocular examinations for cataracts (especially after TBI) and refractive errors, yearly assessment of cardiovascular risk factors, and education/counseling on ‘heart’-healthy lifestyle, blood pressure monitoring, and aggressive management of hypertension to prevent chronic kidney disease, screening for osteopenia/osteoporosis, yearly thyroid function evaluation (especially after TBI or RIT exposure), sexual, gonadal function, and fertility assessment in appropriate age groups, and periodic psychological evaluations (135).

Conclusions and future questions

Tens of thousands of patients have derived significant benefit and many have been cured in the decades after the introduction of autologous HCT as a life-saving treatment of malignant disorders. Through painstaking clinical trials, the techniques of hematopoietic cell collection and graft manipulation, cryopreservation, and reinfusion have improved. Further, conditioning regimens and intensive supportive care have evolved considerably, dramatically advancing patient safety. Practitioners now are generally aware of the appropriate indications for referral and use of this modality as a routine tool. Sophisticated strategies include incorporating novel targeted agents into the mobilization and conditioning regimens, initiation of targeted therapy in the post-transplant period to enhance immunity and reduce relapse, and prevent late consequences.

Current directions being explored in HCT-based treatments for myeloma include newer conditioning regimens and planned post-transplant maintenance or consolidation strategies that aim to induce deeper response states. Approaches at delivering higher doses of radiation therapy to the marrow prior to autologous HCT without increasing overall toxicity are also being explored with helical tomotherapy or bone-seeking radioisotopes (136,137). In lymphoid malignancies results of RIT-based conditioning have been disappointing, but the future ought to look at applying novel monoclonal antibodies, targeted therapies as post-transplant maintenance and/or consolidation. Selective application of such novel maintenance and consolidation or immunotherapy strategies post HCT may help eradicate minimal residual disease (MRD) leading to durable disease control. Table VII summarizes agents with potential to advance autologous HCT outcomes in the future.

Table VII. Novel investigational agents and strategies in development.

Experimental agents and novel strategies in development
Novel conditioning regimens	Helical tomography
	Novel agents (e.g. bortezomib, carfilzomib) in conditioning regimens
	Propylene glycol-free melphalan
Novel mobilization strategies	FLT3 ligand-augmented mobilization
Myeloma	Novel maintenance and consolidation strategies with daratumumab (CD38 antibody), pomalidomide, MLN9708, carfilzomib, elotuzumab
	MRD-directed therapies
B-cell lymphomas	Novel maintenance (with MEDI-551, ABT-199, GA101) therapies post auto-HCT
	Post auto-HCT lenalidomide
	Post-transplant idiotype vaccines
	Post auto-HCT cellular therapy (e.g. CART)
	Tandem auto-allogeneic HCT for FL
	Ibrutinib maintenance/consolidation in DLBCL or MCL
	Disabling immune tolerance by PD1 blockade
Hodgkin lymphoma	Brentuximab vedotin maintenance
	HDAC inhibitors, lenalidomide, everolimus as maintenance therapies
T-cell lymphomas	Lenalidomide or alisertib maintenance
	Brentuximab vedotin maintenance or conditioning in CD30 +variants

Declaration of interest: The authors report no conflicts of interest.