Approximately 20% of invasive ductal breast malignancies are classified as human epidermal growth factor receptor 2 (HER2)–positive because of overexpression of the HER2 protein on the surface of tumor cells or gene amplification.1 Classification of a breast malignancy as HER2-positive or HER2-negative is an essential component of clinical care, as patients with HER2-positive breast cancer receive specific targeted HER2 therapies that reduce the risk of death, whereas patients with HER2-negative breast cancer do not.2,3 An important question is whether HER2 status could be heterogeneous between sites of malignancy in a single patient. If heterogeneity exists, then classification of HER2 status from 1 biopsy may inaccurately describe the extent of disease in a patient with the potential to respond to HER2-targeted therapy. Indeed, heterogeneity of tumors both within and across lesions in a single patient is increasingly being recognized.4 Investigators have identified patients with HER2-positive primary malignancies in which metastases have lost HER2 positivity.5 Potentially more clinically significant, investigators have identified patients with HER2-negative primary malignancies in which metastases have acquired HER2 positivity. These patients may benefit from HER2-targeted therapies that they are not receiving because they are not considered to have HER2-positive disease.
It is currently unclear how to best identify patients who have malignancy with heterogeneous HER2 expression. Identification by biopsy would be difficult because only small samples of tissue from a limited number of lesions are normally sampled. We hypothesized that HER2 heterogeneity may be identified by imaging with a HER2-targeting radiotracer and designed a prospective clinical trial to evaluate HER2 heterogeneity with 89Zr-trastuzumab PET/CT, which allows for visualization of HER2-positive malignancy.6 We published the initial results from the first 9 patients and demonstrated the ability of 89Zr-trastuzumab PET/CT to identify HER2-positive metastases in patients with HER2-negative primary breast malignancy.7 We present the results of the second group of patients, comprising 11 patients.
PATIENTS AND METHODS
A single-center, prospective, institutional review board–approved protocol (ClinicalTrials.gov identifier NCT02286843) was performed. Patients receiving treatment for metastatic HER2-negative primary breast cancer were identified as potential candidates. Inclusion criteria were as follows: (1) biopsy-proven HER2-negative primary malignancy; (2) biopsy-proven metastatic disease; (3) foci of demonstrable metastases on imaging modalities within 6 weeks of enrollment; (4) women 18 or older; and (5) Eastern Cooperative Oncology Group performance score of 0 to 2. Exclusion criteria were as follows: (1) creatinine more than 2 times the upper limit of normal; (2) aspartate aminotransferase/alanine aminotransferase more than 2 times the upper limit of normal; (3) life expectancy of less than 3 months; (4) pregnancy or lactation; and (5) inability to undergo PET/CT scanning because of weight limits. All patients enrolled in the protocol provided written informed consent.
Confirmation of HER2 Status of the Primary Breast Malignancy
The primary breast cancer tissue sample for each patient was retested for HER2 status to ensure proper classification as HER2-negative. Human epidermal growth factor receptor 2 protein overexpression was evaluated by immunohistochemistry (IHC) using a US Food and Drug Administration (FDA)–approved monoclonal antibody (clone 4B5; Ventana, Tucson, AZ) directed against the internal domain of the c-erbB-2 oncoprotein (HER2). The IHC results were categorized as follows: 0, 1+ = negative result, 2+ = equivocal result, and 3+ = positive result, according to the published American Society of Clinical Oncology (ASCO) guidelines8 (Table 1). Tissues with 2+ staining (equivocal) were assessed for HER2 amplification with fluorescence in situ hybridization (FISH) as per ASCO guidelines8 using an FDA-approved probe set (HER2 IQFISH pharmDx; Dako, Pittsburgh, PA) and defined as HER2/CEP17 ratio of 2.0 or greater. Carcinoma with 0 or 1+ IHC staining or 2+ IHC staining and concurrent negative HER2 FISH (HER2/CEP17 ratio <2.0) were classified as HER2-negative primary breast malignancies.
Patients with confirmed HER2-negative primary breast malignancy underwent 89Zr-trastuzumab PET/CT. 89Zr-trastuzumab is composed of the native trastuzumab, an FDA-approved HER2-targeting monoclonal antibody, conjugated with desferrioxamine and labeled with the positron-emitting metalloradionuclide 89Zr. 89Zr has a half-life of 78 hours, long enough to allow favorable biodistribution of radiolabeled intact antibodies. Our institution has an FDA Investigational New Drug for human 89Zr-trastuzumab imaging (Memorial Sloan Kettering Cancer Center Investigational New Drug no. 119907). 89Zr-trastuzumab was produced under Good Manufacturing Practice conditions using previously described methodology.9 The final drug product for human use was manufactured, and quality control tested prior to being released for patient administration to ensure that it conformed to the established acceptance specifications for appearance, pH, endotoxin content, sterilizing filter integrity, radiochemical purity, and radionuclidic identity. Sterility and immunoreactivity determinations were performed after release.
Patients were intravenously administered 185 MBq ± 10% of 89Zr-trastuzumab over 5 minutes. Radiolabeled 89Zr-trastuzumab was brought up to a final mass dose of 50 mg by adding nonradiolabeled trastuzumab at the end of the production because this has been shown to help optimize tumor targeting.6 The final mass dose of 50 mg was provided by the Radiochemistry and Molecular Imaging Probe Core at Memorial Sloan Kettering Cancer Center to the clinic for patient administration. Five or 6 days following 89Zr-trastuzumab administration, patients underwent PET/CT from the midskull to midthigh on a dedicated research GE Discovery PET/CT 710 scanner (GE Healthcare, Chicago, IL), with an 80-mA CT component for attenuation correction and lesion localization. PET/CT images were reconstructed using iterative reconstruction and displayed in multiplanar reconstructions. 89Zr-trastuzumab PET/CT scans were interpreted by 2 different nuclear medicine experts (G.A.U., J.A.C.), both experienced in the use of PET antibody radiotracers. Both readers were able to compare 89Zr-trastuzumab PET/CT results with prior imaging studies. Physiologic 89Zr-trastuzumab uptake was expected in the blood pool, liver, gallbladder, bowel, and kidney. Radiotracer uptake in areas that are not physiologic were graded both qualitatively and semiquantitatively. For qualitative scoring, lesions were scored as suspicious or not suspicious. Only those foci qualitatively scored as suspicious by both readers were considered suspicious foci. Semiquantitative analysis of tracer uptake was performed by recording the SUVmax of suspicious lesions. Three-dimensional volumes of interest were placed in these areas, and tracer uptake was quantified using SUV, calculated as decay-corrected mean region of interest activity (μCi/mL) / (injected dose (μCi) / body weight (g)).
Biopsy and Pathologic Evaluation of Suspicious 89Zr-Trastuzumab Foci
Image-guided biopsy was selected in concert with an experienced oncologic interventional radiologist to minimize risks to the patient while obtaining high-quality samples. Biopsy specimens were evaluated by board-certified breast pathology specialists. The IHC results were categorized according to published ASCO guidelines,8 as discussed previously and in Table 1. Carcinomas with 3+ IHC staining or 2+ IHC staining and concurrent positive HER2 FISH were classified as HER2-positive metastases.
Between December 2015 and May 2016, a second group of 11 women with HER2-negative primary invasive ductal breast cancer completed the study protocol. These were designated patient numbers 10 to 20, following the first 9 patients described in the initial publication.7 Patient characteristics and study results are summarized in Table 2.
Sites of Distant Metastases at Time of Protocol Enrollment
All 11 patients had at least 1 site of metastatic disease at the time of protocol enrollment. The most common sites of distant metastases were bone (n = 7), nodal (n = 7), and liver (n = 7), followed by lung (n = 2), pleural (n = 1), and chest wall (n = 1). Nine of the 11 patients had multiple organ system metastatic involvement.
Confirmation of HER2 Status of the Primary Breast Malignancy
On IHC retesting of the patients’ archived primary breast cancer specimens, all 11 patients had confirmed HER2-negative primary malignancies according to published ASCO guidelines.8 Three patients had HER2 IHC scores of 0, whereas 7 had HER2 IHC scores of 1+, and 1 had a HER2 IHC score of 2+ with a FISH of 1.4.
All 11 patients underwent 89Zr-trastuzumab PET/CT. Patients were monitored for 30 minutes after tracer injection and called the following day; no adverse effects were observed or reported. Vital signs were recorded before and after injection, and no changes of clinical significance were observed.
In 4 (36%) of 11 patients, both readers observed foci of 89Zr-trastuzumab avidity considered suggestive of HER2-positive disease. In 2 patients, the suspicious 89Zr-trastuzumab foci were in pleura, whereas 1 patient exhibited suspicious 89Zr-trastuzumab foci in the bone, and 1 patient exhibited suspicious 89Zr-trastuzumab foci in the liver.
Biopsy and Pathologic Evaluation of Suspicious 89Zr-Trastuzumab Foci
Image-guided biopsy was performed in all 4 patients with suspicious 89Zr-trastuzumab–avid foci. Patient 11 was a 40-year-old woman who was diagnosed in October 2013 as having an estrogen receptor (ER)-positive/HER2-negative primary invasive ductal carcinoma. At the time of protocol enrollment, an FDG PET/CT demonstrated FDG-avid bone, liver, nodal, and chest wall malignancy. Bone and chest wall lesions had been pathologically confirmed as metastases. 89Zr-trastuzumab PET/CT was performed in March 2016 and demonstrated 89Zr-trastuzumab–avid osseous and nodal lesions. The most avid lesion was a right femoral focus (SUVmax, 10.1). Biopsy of the right femoral focus demonstrated HER2 IHC of 2+ and FISH of 1.6. As FISH was less than 2.0, this was considered a false-positive 89Zr-trastuzumab focus.
Patient 14 was a 47-year-old woman diagnosed in June 2011 as having an ER-positive/HER2-negative primary invasive ductal carcinoma. At the time of protocol enrollment, an FDG PET/CT demonstrated FDG-avid liver, nodal, and pleural malignancy (Fig. 1). Hepatic and pleural lesions had been pathologically confirmed as metastases. 89Zr-trastuzumab PET/CT was performed in May 2016 and demonstrated 89Zr-trastuzumab–avid pleural (SUV, 6.9) and liver (SUV, 10.9) lesions. The pleural lesion was selected for biopsy as the lesion with the lowest risk for sampling. Biopsy of the pleural lesion demonstrated HER2 IHC of 2+ and FISH of 2.4. As FISH was greater than 2.0, this was considered a true-positive 89Zr-trastuzumab focus for a HER2-positive distant metastasis.
Patient 17 was a 51-year-old woman diagnosed in November 2010 as having an ER-positive/HER2-negative primary invasive ductal carcinoma. At the time of protocol enrollment, an FDG PET/CT demonstrated FDG-avid bone, nodal, and lung malignancy (Fig. 2). A lung lesion had been pathologically confirmed as metastasis. 89Zr-trastuzumab PET/CT was performed in May 2016 and demonstrated 89Zr-trastuzumab–avid lung (SUV, 11.3) lesions. Biopsy of a lung lesion demonstrated HER2 IHC of 1+, which was considered a false-positive 89Zr-trastuzumab focus.
Patient 20 was a 58-year-old woman diagnosed in August 2003 as having an ER-positive/HER2-negative primary mixed invasive ductal and lobular carcinoma. At the time of protocol enrollment, an FDG PET/CT demonstrated FDG-avid bone, liver, and nodal malignancy (Fig. 3). Bone and liver lesions had been pathologically confirmed as metastases. 89Zr-trastuzumab PET/CT was performed in June 2016 and demonstrated 89Zr-trastuzumab–avid liver (SUV, 15.7) lesions. Biopsy of a liver lesion demonstrated HER2 IHC of 1+, which was considered a false-positive 89Zr-trastuzumab focus. The remaining 7 patients did not have suspicious foci of 89Zr-trastuzumab avidity.
This study describes a second group of 11 patients in a prospective clinical trial of HER2-targeted imaging with 89Zr-trastuzumab PET/CT to evaluate for unsuspected HER2-positive metastases in patients with HER2-negative primary breast cancer. It demonstrates that in a small subset of patients with HER2-negative primary breast cancer HER2-targeted imaging can detect previously unsuspected HER2-positive disease.
Between the current group of 11 patients and a previously published initial group of 9 patients,7 a completed cohort of 20 patients with HER2-negative primary breast cancer have now been evaluated. In total, 3 (15%) of 20 patients were demonstrated to have biopsy-proven unsuspected HER2-positive metastases. In another 6 (30%) of 20 patients, suspicious 89Zr-trastuzumab–avid foci resulted in biopsies with HER2-negative pathology, which are being conservatively considered false-positive 89Zr-trastuzumab–avid foci.
While medical imaging of oncology patients has traditionally been used for the detection of tumors, staging, and evaluation of treatment response, this study suggests that targeted medical imaging may be able to help select patients for targeted systemic therapies. More than 900,000 women are living with metastatic breast cancer, with more than 50,000 diagnosed each year.10 Eighty percent of these women have HER2-negative primary malignancies. If 15% of patients with metastatic HER2-negative primary breast cancer could be found to harbor HER2-positive metastases, this would represent a current population of 135,000 women. Thus, targeted HER2 imaging has the potential to play a clinically valuable role in identifying patients who may benefit from HER2-targeted therapies.
A limitation of this approach is the currently high level of presumably false-positive suspicious 89Zr-trastuzumab–avid foci. This leads to unneeded workup and biopsies. The reason for these false-positive foci is unclear. One possibility is pathologic sampling or processing errors. For example, decalcification of osseous lesions may decrease the intensity of HER2 IHC staining, resulting in a false-negative pathologic result.11 If a lesion exhibits heterogeneity of HER2 expression, it is possible that in any one biopsy the HER2-positive malignancy was undersampled. Although this could result in erroneous pathologic analysis, given the high number of false-positives, it is likely that at least some represent nonspecific 89Zr-trastuzumab uptake in HER2-negative lesions. Another possible explanation is free 89Zr being released from its chelator during the long 5-day uptake period. When loosely chelated, 89Zr is known to be a bone seeker.12 Indeed, evidence suggests that when 89Zr is chelated to antibodies with desferrioxamine radioactivity accumulates in the bone.13 Thus, nonspecific binding of potentially free 89Zr at sites of osseous turnover associated with bone metastases could result in false-positive osseous foci. Another possible explanation for presumed false-positive 89Zr-trastuzumab–avid foci could be related to in vivo HER2 internalization rates or affinity differences that would not necessarily be reflected by IHC or FISH.14 More specific HER2-targeting radiotracers will need to be explored. One potential alternative is 89Zr-pertuzumab, which demonstrated HER2 specificity in animal models.15
Human epidermal growth factor receptor 2–negative patients have been reported to respond to trastuzumab therapy. It would be of interest to evaluate if patients with positive 89Zr-trastuzumab scans but negative HER2 status by conventional criteria respond to trastuzumab. This would raise the potential of using 89Zr-trastuzumab as a predictive biomarker16 and selecting patients for HER2-target therapy based on imaging.
It is interesting to note that the intensity of 89Zr-trastuzumab avidity did not discriminate between true-positive and false-positive lesions for HER2 expression. The most 89Zr-trastuzumab–avid lesion in the study was a hepatic lesion (patient 20; SUV, 15.7), which had a HER2-negative pathology. Likewise, a substantially less 89Zr-trastuzumab–avid lesion (patient 14; SUV, 6.9) had HER2-positive pathology and was a true-positive.
Another interesting issue was that even when a 89Zr-trastuzumab–avid focus was considered a false-positive, there could be noticeable heterogeneity in HER2 expression between the primary malignancy and the avid metastasis. For example, patient 11 had a primary malignancy with an IHC of 0. The 89Zr-trastuzumab–avid osseous metastasis that was biopsied had an IHC of 2+ and FISH of 1.6. While IHC 2+/FISH 1.6 is considered HER2-negative by ASCO guidelines, it still represents a change from the primary malignancy. The value of 89Zr-trastuzumab PET/CT for evaluating this type of heterogeneity needs further evaluation.
The strength of this study is its design as a prospective clinical trial, whereas its weakness is still the small sample size. Ethical standards and logistical obstacles prevent biopsy of all 89Zr-trastuzumab foci; thus, pathology is classified on the basis of a limited number of biopsies. Likewise, ethical standards prevent biopsy confirmation of HER2-negative status of metastases in patients without suspicious foci on 89Zr-trastuzumab imaging.
In conclusion, 89Zr-trastuzumab PET/CT can be used to detect unsuspected HER2-positive metastases in a subset of patients with HER2-negative primary breast cancer. This demonstrates that targeted imaging can identify patients amenable to targeted therapies. Presumed false-positive results currently limit the translation of these techniques, but there is great potential for more specific methodology to have substantial clinical value in the future.
1. Slamon DJ, Clark GM, Wong SG, et al. Human breast cancer
: correlation of relapse and survival with amplification of the HER-2/neu oncogene. Science
2. Slamon DJ, Leyland-Jones B, Shak S, et al. Use of chemotherapy plus a monoclonal antibody against HER2
for metastatic breast cancer
that overexpresses HER2
. N Engl J Med
3. Romond EH, Perez EA, Bryant J, et al. Trastuzumab
plus adjuvant chemotherapy for operable HER2
-positive breast cancer
. N Engl J Med
4. McGranahan N, Swanton C. Biological and therapeutic impact of intratumor heterogeneity in cancer evolution. Cancer Cell
5. Niikura N, Liu J, Hayashi N, et al. Loss of human epidermal growth factor receptor 2 (HER2
) expression in metastatic sites of HER2
-overexpressing primary breast tumors. J Clin Oncol
6. Dijkers EC, Oude Munnink TH, Kosterink JG, et al. Biodistribution of 89
and PET imaging of HER2
-positive lesions in patients with metastatic breast cancer
. Clin Pharmacol Ther
7. Ulaner GA, Hyman DM, Ross DS, et al. Detection of HER2
-positive metastases in patients with HER2
-negative primary breast cancer
. J Nucl Med
8. Wolff AC, Hammond ME, Hicks DG, et al. Recommendations for human epidermal growth factor receptor 2 testing in breast cancer
: American Society of Clinical Oncology/College of American Pathologists clinical practice guideline update. J Clin Oncol
9. Vosjan MJ, Perk LR, Visser GW, et al. Conjugation and radiolabeling of monoclonal antibodies with zirconium-89 for PET imaging using the bifunctional chelate p-isothiocyanatobenzyl-desferrioxamine. Nat Protoc
10. Johnson RH, Chien FL, Bleyer A. Incidence of breast cancer
with distant involvement among women in the United States, 1976 to 2009. JAMA
11. Darvishian F, Singh B, Krauter S, et al. Impact of decalcification on receptor status in breast cancer
. Breast J
12. Abou DS, Ku T, Smith-Jones PM. In vivo biodistribution and accumulation of 89
Zr in mice. Nucl Med Biol
13. Deri MA, Ponnala S, Kozlowski P, et al. p-SCN-Bn-HOPO: a superior bifunctional chelator for (89)Zr immunoPET. Bioconjug Chem
14. Rudnick SI, Lou J, Shaller CC, et al. Influence of affinity and antigen internalization on the uptake and penetration of Anti-HER2
antibodies in solid tumors. Cancer Res
15. Marquez BV, Ikotun OF, Zheleznyak A, et al. Evaluation of (89)Zr-pertuzumab in breast cancer
xenografts. Mol Pharm
16. Ulaner GA, Riedl CC, Dickler MN, et al. Molecular imaging of biomarkers in breast cancer
. J Nucl Med
. 2016;57 suppl 1:53S–59S.